2013 02-12 asm full report

Active Distribution
System Management
A key tool for the smooth integration
of distributed generation
full discussion paper
A EURELECTRIC paper, FEBRUARY 2013

The Union of the Electricity Industry–EURELECTRIC is the sector association representing the common interests of
the electricity industry at pan-European level, plus its affiliates and associates on several other continents.
In line with its mission, EURELECTRIC seeks to contribute to the competitiveness of the electricity industry, to
provide effective representation for the industry in public affairs, and to promote the role of electricity both in the
advancement of society and in helping provide solutions to the challenges of sustainable development.
EURELECTRIC’s formal opinions, policy positions and reports are formulated in Working Groups, composed of
experts from the electricity industry, supervised by five Committees. This “structure of expertise” ensures that
EURELECTRIC’s published documents are based on high-quality input with up-to-date information.
For further information on EURELECTRIC activities, visit our website, which provides general information on the
association and on policy issues relevant to the electricity industry; latest news of our activities; EURELECTRIC
positions and statements; a publications catalogue listing EURELECTRIC reports; and information on our events and
conferences.
Dépôt légal: D/2013/12.105/7
EURELECTRIC pursues in all its activities the application of
the following sustainable development values:
Economic Development
Growth, added-value, efficiency
Environmental Leadership
Commitment, innovation, pro-activeness
Social Responsibility
Transparency, ethics, accountability

Active Distribution System Management
A key tool for the smooth integration of distributed
generation
--------------------------------------------------------------------------------------------------
TF Active System Management
Per HALLBERG (SE) Chair
Juan Jose ALBA RIOS (ES); Christer BERGERLAND (SE); Aurelio BLANQUET (PT); Marcel CAILLIAU (BE); Conor
CLIFFORD (SE); Paul DE WIT (NL); Ellen DISKIN (IE); Håkan FEUK (SE); Bruno GOUVERNEUR (BE); Mike KAY
(GB); Pauline LAWSON (GB); Marina LOMBARDI (IT); Pavla MANDATOVA (EURELECTRIC Secretariat); Ruud
OTTER (NL); Pablo SIMON CABALLERO (ES); Jorge TELLO GUIJARRO (ES); Walter TENSCHERT (AT); David
TREBOLLE (ES); Siegfried WANZEK (DE)
Contact:
Gunnar LORENZ, Head of Networks Unit – glorenz@eurelectric.org
Pavla MANDATOVA, Adviser Networks Unit - pmandatova@eurelectric.org

Contents
INTRODUCTION ................................................................................................................................1
1. INTEGRATION OF DISTRIBUTED GENERATION: A KEY CHALLENGE FOR DSOS..............................................2
1.1. Distributed Generation: Facts and Figures.....................................................................................2
1.2. Key Challenges for Current Distribution Networks........................................................................4
1.2.1. Network Reinforcement.................................................................................................4
1.2.2. Distribution Network Operation ....................................................................................6
1.2.3. Traditional Design of Distribution Networks .................................................................9
2. ACTIVE DISTRIBUTION NETWORKS................................................................................................10
2.1. Key Building Blocks.......................................................................................................................10
2.2. Distribution Network Development, Planning, Access & Connection .........................................14
2.2.1. Coordinated Network Development ............................................................................14
2.2.2. Network Capacity Management..................................................................................15
2.2.3. Connection ...................................................................................................................17
2.3. Active Distribution Network Operation .......................................................................................18
2.3.1. DSO System Services & ‘Traffic Lights’ Approach ........................................................19
2.3.2. Information Exchange..................................................................................................23
2.3.3. Voltage Control ............................................................................................................24
2.4. Technical Development: Towards Flexible Distribution Systems ................................................27
3. IMPLICATIONS FOR REGULATION AND MARKET DESIGN......................................................................28
KEY CONCLUSIONS ..........................................................................................................................30
POLICY RECOMMENDATIONS .............................................................................................................31
ANNEX 1 – EXAMPLE OF A POSITIVE CBA, THE IMPROGRESS PROJECT ......................................................33
ANNEX 2 – CASE STUDIES .................................................................................................................34
A. Demonstration of active power effect in voltage control – Networks 2025 Project, Spain........34
B. Active voltage management using wind generation voltage control and reactive capabilities –
ESB Networks, Ireland..................................................................................................................36
C. Technical Control Strategy for Active System Management – EDP Distribuicão, Portugal.........38
D. Protection and Automation Strategies in MV Networks – Enel Distribuzione, Italy ...................39
E. Integration of RENEWABLES in MV and LV Networks – E.ON Bayern, Germany ........................40
ANNEX 3 – SURVEY RESULTS .............................................................................................................42
A. Automation and Control in MV & LV Distribution Networks.......................................................42
B. Curtailment of Distributed Generation........................................................................................45
GLOSSARY.....................................................................................................................................46
REFERENCES & BIBLIOGRAPHY............................................................................................................47

List of Figures
Figure 1 Common voltage connection levels for different types of DG/RES .....................................................2
Figure 2 Distributed generation installed capacity and peak demand in Galicia, Spain ....................................3
Figure 3 Installed capacity of photovoltaic installations in the E.ON Bayern grid (Source: E.ON).....................3
Figure 4 Power flows between transmission and distribution network in Italy, 2010-2012 (Source: Enel
Distribuzione) .....................................................................................................................................................4
Figure 5 Relation between the degree of DG penetration and grid losses (Source: van Gerwent)...................5
Figure 6 Instability in distribution system (Source: Mainova)............................................................................6
Figure 7 Reverse power flows at a substation in northwest Ireland (Source: ESB Networks)...........................7
Figure 8 Current DSO networks..........................................................................................................................9
Figure 9 Three-Step Evolution of Distribution Systems....................................................................................11
Figure 10 DSO interactions with markets & TSO at different time frames ......................................................12
Figure 11 Variable access approach (Source: EWE Netz).................................................................................16
Figure 12 Extension of network capacity for peak load versus a load management solution (Source: EWE
Netz) .................................................................................................................................................................17
Figure 13 Market and network operations ......................................................................................................19
Figure 14 Instability in the distribution system................................................................................................21
Figure 15 Information Exchange Today and in the Future...............................................................................23
Figure 16 An illustrative example of a traditional approach to voltage control ..............................................24
Figure 17 Comparison of BAU distribution Investments and maintenance costs (WP4) and costs when
applying ANM solutions (WP5) for the Spanish network.................................................................................33
Figure 18 R/X ratio effect in voltage control ....................................................................................................34
Figure 19 Extra reactive power contribution to maintain a voltage setpoint..................................................34
Figure 20 DG effect in the voltage triangle design...........................................................................................35
Figure 21 Wind farm active and reactive capabilities as demonstrated in ESB Networks Volt / VAr trials.....36
Figure 22 Test network for field demonstration of coordinated Volt / VAr Control........................................37
Figure 23 Active and reactive power where wind farms connect to substation - as active export increases,
reactive power is imported to control PCC voltage .........................................................................................37
Figure 24 Hot spots areas in E.ON Bayern (installed RES capacity in EBY grid areas)......................................40
List of Tables
Table 1 Three-Step Evolution of Distribution Systems in detail.......................................................................13
Table 2 New System Services at Distribution Level..........................................................................................20
Table 3 Voltage control in current distribution networks................................................................................25
Table 4 Technology options for distribution network development ...............................................................28
Table 5 DG Contribution to voltage control by voltage level...........................................................................35

1
Introduction
The expansion of decentralised and intermittent renewable generation capacities introduces new
challenges to ensuring the reliability and quality of power supply. Most of these new generators
(both in number and capacity) are being connected to distribution networks – a trend that is set to
continue in the coming years.
This development has profound implications for distribution system operators (DSOs). Until
recently, DSOs designed and operated distribution networks through a top-down approach.
Predictable flows in the electricity network did not require extensive management and monitoring
tools.
But this model is changing. Higher shares of distributed energy sources lead to unpredictable
network flows, greater variations in voltage, and different network reactive power characteristics.
Local grid constraints will occur more frequently, adversely affecting the quality of supply. Yet
DSOs are nevertheless expected to continue to operate their networks in a secure way and to
provide high-quality service to their customers.
This EURELECTRIC report addresses a number of fundamental questions that arise from the
integration of distributed generation (DG) and other distributed energy resources (DER) into the
energy system:
 How can DSOs make the most efficient use of the existing network?
 When are new infrastructure and changes in system architecture needed to better
integrate DG and DER?
 Which types of system services are needed and how can they be procured?
 How can renewable energy sources (RES), DG, and DER contribute to system security?
 How should the regulatory framework develop?
Active distribution system management may provide some answers to these questions. Indeed,
distribution management will allow grids to integrate DER efficiently by leveraging the inherent
characteristics of this type of generation. The growth of DER requires changes to how distribution
networks are planned and operated. Bi-directional flows need to be taken into account: they must
be monitored, simulated and managed.
The report sets out implications for the tasks of system operators (TSOs and DSOs) and DG/RES
operators and outlines options for system planning and development, system operation, and
information exchange, thereby opening the door for further analysis. It focuses on outstanding
technical issues and necessary operational requirements and calls for adequate regulatory
mechanisms that would pave the way for these solutions.
The paper focuses largely on distributed generation – a challenge many DSOs are already facing
today. However, the presented possible solutions will generally also be applicable to other
flexibility providers like loads and electric vehicles, which fall under the umbrella of flexibility
offered by DER.

2
1. Integration of Distributed Generation: A Key Challenge for DSOs
Distributed energy resources (DER) include distributed/decentralised generation (DG) and distributed
energy storage (DS)1
. With the EU on its way to meeting a 20% target for RES in total energy consumption
by 2020, the share of electricity supply from RES is on the rise. Intermittent RES like solar and wind add an
additional variable to the system that will require more flexibility from generation (including RES) and
demand and investments in network infrastructure. Such intermittent RES will be connected largely to
European distribution systems. At the same time, electrification of transport will be needed to further
decarbonise the economy. For a significant deployment of electric vehicles by 2050, Europe needs to target
a 10% share of electric vehicles by 2020. These vehicles will need to be charged through the electrical
system. Together with the electrification of heating and cooling, these trends will contribute to further
evolution of European power systems.
1.1. Distributed Generation: Facts and Figures
Distributed/decentralised generation (DG) are generating plants connected to the distribution network,
often with small to medium installed capacities, as well as medium to larger renewable generation units.
Due to high “numbers”, they are important compared to the “size” of the distribution network. In addition
to meeting on-site needs, they export the excess electricity to the market via the local distribution network.
DG is often operated by smaller power producers or so-called prosumers.
Unlike centralised generation, which is dispatched in a market frame under the technical supervision of
TSOs, small DG is often fully controlled by the owners themselves. The technologies include engines, wind
turbines, fuel cells and photovoltaic (PV) systems and all micro-generation technologies. In addition to
intermittent RES, an important share of DG is made up of combined heat and power generation (CHP),
based on either renewables (biomass) or fossil fuels. A portion of the electricity produced is used on site,
and any remainder is fed into the grid. By contrast, in case of CHP the generated heat is always used locally,
as heat transport is costly and entails relatively large losses. Figure 1 provides an overview of generation
types usually connected at different distribution voltage levels.
Usual connection
voltage level
Generation Technology
HV
(usually 38-150 kV)
Large industrial CHP
Large-scale hydro
Offshore and onshore wind parks
Large PV
MV
(usually 10-36 kV)
Onshore wind parks
Medium-scale hydro
Small industrial CHP
Tidal wave systems
Solar thermal and geothermal systems
Large PV
LV (< 1kV) Small individual PV, Small-scale hydro
Micro CHP, Micro wind
Figure 1 Common voltage connection levels for different types of DG/RES
1
Storage is not only a resource but also an off take/load. Electric vehicles could be used as storage in the future.

3
The following examples demonstrate that a move from mere DG connection to DG integration is a must
already today in some countries. As illustrated in Figure 2, in Galicia, Spain, the installed capacity of DG
connected to the distribution networks of Union Fenosa Distribución (2,203 MW) represents 120% of the
area’s total peak demand (1,842 MW).2
Figure 2 Distributed generation installed capacity and peak demand in Galicia, Spain
(Source: Union Fenosa Distribución)
In the regional distribution network in the south of Germany (see Figure 3), the installed capacity of
intermittent renewable DG already represents a large percentage of the peak load. In many places, the
DG output of distribution networks already exceeds local load – sometimes by multiple times. From the
TSO point of view, the DSO network then looks like ‘a large generator’ in periods with high RES production.
*status for 31 December 2012
Figure 3 Installed capacity of photovoltaic installations in the E.ON Bayern grid (Source: E.ON)
2
Overall, DG covers more than 30% of Union Fenosa Distribución regional energy (MWh) demand already today.
Max. Hourly Average
capacity (MW)
Galicia
Demand
1.842
Installed
Capacity
(MW)
Percentage
(%)
CHP (Natural Gas) 166,9 7,6
Rest CHP 319,4 14,5
Wind Power 1.369,5 62,1
Photovoltaic (PV) 10,3 0,5
Hydro 306,1 13,9
Other Renewables 31,4 1,4
TOTAL Generation 2.203,6 100
NaturalGas
CHP7,6%
Rest CHP
14,5%
Wind power
62,1%
PV
0,5%
Hydro
13,9%
Other
Renewables
1,4%

4
In 2011, 10 GW of PV were newly connected to Italian distribution grids (Enel Distribuzione), the highest
yearly increase in distributed generation connected to the grid worldwide.
In northwest Ireland with a peak demand of 160 MW, 307.75 MW of distributed wind generation are
already connected to the distribution system, and a further 186 MW are contracted or planned. Beyond
this, another 640 MW of applications have been submitted.
1.2. Key Challenges for Current Distribution Networks
In theory, due to its proximity to the loads, distributed generation should contribute to the security of
supply, power quality, reduction of transmission and distribution peak load and congestion, reduced need
for long distance transmission, avoidance of network overcapacity, deferral of network investments and
reduction in distribution grid losses (via supplying active power to the load and managing voltage and
reactive power in the grid).
In reality, integrating distributed generation into DSO grids represents a capacity challenge due to DG
production profiles, location and firmness. DG is not always located close to load and DG production is
mostly non-dispatchable (cannot control its own output). Therefore, production does not always coincide
with demand (stochastic regime) and DG does not necessarily generate when the distribution network is
constrained. In addition, power injections to higher voltage levels need to be considered where the local
capacity exceeds local load. This poses important challenges for both distribution network development
and operation.
1.2.1. Network Reinforcement
The ability of DG to produce electricity close to the point of consumption alleviates the need to use
network capacity for transport over longer distances during certain hours. However, the need to design
the distribution networks for peak load remains undiminished and the overall network cost may even
increase. For example, peak residential demand frequently corresponds to moments of no PV production.
Figure 4 shows the situation in the southern Italian region of Puglia. It indicates the incredible increase in
the power installed and energy produced from PVs in recent years and the subsequent evolution of power
flows at the connection point between the transmission network and the distribution network. As the peak
load corresponds to literally zero PV production, there is no reduction in investment (“netting” generation
and demand).
Figure 4 Power flows between transmission and distribution network in Italy, 2010-2012 (Source: Enel Distribuzione)

5
Generally speaking, distribution networks have to be prepared for all possible combinations of
production and load situations. They are designed for a peak load that often only occurs for a few hours
per year, in what this paper refers to as the fit-and-forget approach. Even constraints of short duration
trigger grid adaptations (e.g. reinforcement).3
Distribution networks have always been designed in this
way, but with DG the utilisation rate of network assets declines even more.
Network connection studies and schemes for generators are designed to guarantee that under normal
operation all capacity can be injected at any time of the year.4
The current European regulatory framework
provides for priority and guaranteed network access for electricity from RES (Art. 16 of RES Directive
2009/28/EC) and RES-based CHP (Art. 14 of the new Energy Efficiency Directive 2012/27/EC). RES-E is
mostly connected on a firm/permanent network access basis (but cannot be considered as firm for such
design purposes). Generation and load of equivalent sizes imply different design criteria as e.g. wind and PV
has lower diversity than load. In addition, wider cables to lower the voltage might be needed. Overall, this
can lead to higher reinforcement cost and thus a rise in CAPEX for DSOs and/or higher connection costs
for DG developers. The contribution of DG to the deferral of network investments holds true only for a
relatively small amount and size of DG and for predictable and controllable primary sources.
The lead time needed to realise generation investment is usually shorter than that for grid
reinforcement. Article 25.7 of Directive 2009/72/EC requires DSOs to take into account distributed energy
resources and conventional assets when planning their networks. This may be complicated when
applications for connection are submitted at short notice and DSOs receive no information on connection
to private networks. Situations will occur when DSOs have large amounts of DER connected to their
network and the resulting net demand seen further up the system hierarchy is lowered. Virtual saturation –
a situation when the entire capacity is reserved by plants queuing for connection that may not eventually
materialise – may also occur as generator plans cannot be firm before the final investment decision.
However, even in the cases when the project is not built, it occupies an idle capacity which may lead new
grid capacity requests to face increased costs in case network reinforcements are needed. Temporary lack
of network capacity may result in ‘queuing’ and long waiting times, delaying grid connection of new
generators.
The situation is similar in case of grid losses (related costs are part of DSOs’ OPEX). Figure 5 demonstrates
that with a low share of DG these losses drop, but once there are large injections of DG into the DSO
network and load flows over the network, grid losses tend to increase. DG can reduce network costs in
transport levels but entails higher costs in the level to which it is connected.
Figure 5 Relation between the degree of DG penetration and grid losses (Source: van Gerwent)
3
In some countries, this regulation is part of the regulation for feed-in of renewable energy or other preferential
power production and remuneration.
4
In some countries this is also the case for the N-1 contingency state which is typically considered in a meshed
network and represents the fail of a grid element. Many MV networks are meshed even when they are operated
radially, so some N-1 is also possible in MV.

6
1.2.2. Distribution Network Operation
In addition, distributed generation, in particular intermittent RES, poses a challenge not only for system
balancing, but also for local network operation. The security and hosting capacity of the distribution
system is determined by voltage (statutory limits for the maximum and minimum voltage ensure that
voltage is kept within the proper margins and is never close to the technical limits of the grid) and the
physical current limits of the network (thermal rates of lines, cables, transformers that determine the
possible power flow).
A distribution system can be driven out of its defined legal and or physical operating boundaries due to one
or both of the following:
 Voltage variations: Injection of active power leads to voltage profile modifications. Voltage increase
(overvoltage) is the most common issue at the connection point for DG units and the relevant grid
area. Reversed power flows (flows from distribution to transmission) occur when DG production
exceeds local load. The more local production exceeds local demand, the stronger the impact on
voltage profiles. Figure 6 illustrates such situations.
DSOs may have difficulties in maintaining the voltage profile at the customer connection points, in
particular on LV level, as active voltage control is not in place. In most countries, monitoring of grid
values is missing and most distributed generators are not equipped to participate in system
management – no active contribution of generation to network operation is expected.5
As a result,
operational system security may be endangered and security of facilities (both customers’ installations
and the network as such) put at risk.
Figure 6 Instability in distribution system (Source: Mainova)
 Congestions: When excessive DG feed-in pushes the system beyond its physical capacity limits (P
G
- P
L
>
Pmax
), congestions may occur in distribution networks. This may lead to necessary emergency actions to
interrupt/constrain off generation feed-in or supply. A similar situation can occur in case of excessive
demand on the system (P
L
- P
G
> P
max
). This could apply to high load incurred e.g. by charging of electric
vehicles, heat pumps and electrical HVAC (heating ventilation and air-conditioning).
5
Where DSOs are already in the ‘reactive DSO’ phase (see section 2.1), these solutions have been implemented.

7
Generation curtailment is used in cases of system security related events (i.e. congestion or voltage
rise). The regulatory basis for generation curtailment in such emergency situations differs across
Europe.6
In some countries (e.g. in Italy, Spain, Ireland), the control of DG curtailment is de facto in TSO
hands: the DSO can ask the TSO, who is able to control active power of DG above a certain installed
capacity, to constrain DG if there is a local problem.7
As the TSO is not able to monitor distribution
network conditions (voltage, flows), DSOs can only react to DG actions. This can result in deteriorating
continuity on the distribution system which will impact both demand customers and DG. Annex 3.B
provides an overview of the situation in different countries.
In systems with a high penetration of DG, both types of unsecure situations already occur today. As a
result, DSOs with high shares of DG in their grids already face challenges in meeting some of their
responsibilities8
. These challenges are expected to become more frequent, depending on the different
types of connected resources, their geographic location and the voltage level of the connection.
Figure 7 shows the reverse power flows on an Irish substation in an area with high wind penetration.
Figure 7 Reverse power flows at a substation in northwest Ireland (Source: ESB Networks)
6
Curtailment/feed-in management rules are either not defined by law at all (e.g. Austria), defined at the TSO level
only (e.g. Spain or Italy) or defined at both the TSO and the DSO level (e.g. UK or Germany according to the revised
feed-in law).
7
The TSO might even issue V, Q or pf set points to DG on the distribution system.
8
DSOs are responsible for ensuring the long-term ability of the system to meet reasonable demands for the
distribution of electricity and for operating, maintaining and developing under economic conditions a secure, reliable
and efficient electricity distribution system in their area with due regard for the environment and energy efficiency
(Article 25(1) of Directive 2009/72/EC).

8
In sum, the key challenges for DSOs include
 Increased need for network reinforcement to accommodate new DG connections:
 Network operators are expected to provide an unconditional firm connection which may
cause delays or increase costs for connecting embedded generation;
 Increased complexity for extension (including permitting procedures) and maintenance of the
grid may require temporary limitation for connection of end customers.
 Problems with operation of the distribution grid:
 Local power quality/operational problems, in particular variations in voltage but also fault
levels and system perturbations like harmonics or flickers;
 Rising local congestions when flows exceed the existing maximum capacity, which may result
in interruptions of generation feed-in or supply;
 Longer restoration times after network failure due to an increased number and severity of
such faults.
BOX 1: Current DSO responsibilities
 Distribution planning, system development, connection & provision of network capacity
DSOs are in charge of developing their network. They design new lines and substations and ensure
that they are delivered or that existing ones are reinforced to enable connection of load and
decentralised power production. Depending on the size of a DG/RES & DER system, DSOs may require
a new connection at a particular voltage level. They are obliged to provide third party access to all end
customers and provide network users with all information they need for efficient access and use of
the distribution system. They may refuse access to the grid only when they can prove that they lack
the necessary network capacity (Art. 32 of Directive 2009/72/EC).
 Distribution network operation/ management and support in system operation
DSOs maintain the system security and quality of service in distribution networks. This includes
control, monitoring and supervision, as well as scheduled and non-scheduled outage management.
DSOs are responsible for operations directly involving their own customers. They support the TSOs,
who are typically in charge of overall system security, when necessary in a predefined manner, either
automatically or manually (e.g. via load shedding in emergency situations). Such systems of
cooperation for intervention in generation and demand in cases of system security events are defined
in detail in national regulations. A common basis for these rules is now being set in the EU-wide
network codes (operational security, balancing, congestion management, etc.).
 Power flow management: Ensuring high reliability and quality in their networks
 Continuity and capacity: DSOs are subject to technical performance requirements for quality of
service including continuity of supply (commonly assessed by zonal indexes such as average
duration of interruptions per customer per year (SAIDI) and average number of interruptions per
customer per yea (SAIFI) or individual indexes like number and duration of interruptions) and
power quality laid out in national law, standards and grid codes.9
They are also responsible for
voltage quality in distribution networks (maintaining voltage fluctuations on the system within
given limits). In planning, the DSO ensures that networks are designed to maintain these standards
regardless of power flow conditions. However in cases of network faults, planned outages or other
9
National regulatory authorities (NRA) have the duty of setting or approving standards and requirements for quality
of supply or contributing thereto together with other competent authorities (Article 37(1h) of Directive 2009/72/EC).

9
anomalous events, the DSO must undertake switching actions so that adequate supply quality is
maintained. While to date this has been rather static, increasingly automation or remote switching
will need to be undertaken to ensure near real-time fault isolation and restoration of supply.
 Voltage and reactive power: Voltage quality is impacted by the electrical installations of connected
network users. Thus the task of the DSO in ensuring voltage quality must account also for the
actions of network users, adding complexity and the need for both real-time measurement and
mitigating resources (i.e. on-load voltage control) and strict network connection criteria. European
standard EN 50160 specifies that the maximum and minimum voltage at each service connection
point must allow an undisturbed operation of all connected devices. Voltage at each connection
should thus be in the range of ± 10% of the rated voltage under normal operating conditions. In
some countries, compliance with these or other specified national voltage quality requirements
that can be even more restrictive represents part of DSOs’ contractual obligations and quality
regulation. In some countries, network operators are required to compensate customers in case
the overall voltage quality limits are breached.10
1.2.3. Traditional Design of Distribution Networks
The fundamental topological design of traditional distribution grids has not changed much over the past
decades. Up until recently, DSOs have distributed energy and designed their grids on a top-down basis.
Under the paradigm “networks follow demand”, their primary role was to deliver energy flowing in one
direction, from the transmission substation down to end users. This approach makes use of very few
monitoring tools and is suitable for distribution networks with predictable flows.
Because of the different development of electrification, distribution networks characteristics differ from
country to country. Voltage rate levels are usually distinguished as LV, MV or HV.11
As illustrated in Figure 8
the level of supervision, control and simulation in HV distribution networks is close to that of TSOs in their
networks. MV and LV networks are mostly rather passive – here DSOs lack network visibility and control.
The lower the monitoring level, the lower the operational flexibility. For more detailed information on
automation and control see results of a survey conducted among EURELECTRIC members in Annex 3.A.
Figure 8 Current DSO networks
10
CEER Benchmarking report, 2011.
11
CENELEC standard EN 50160 (2010) uses following the classification: LV < 1 kV, MV 1 kV – 36 kV, HV 36 - 150 kV.

10
BOX 2: Traditional distribution networks
Traditional distribution networks have different characteristics in topology (meshed or radial), operation
type (meshed or radial), number of assets and customers, operational flexibility and monitoring level:
 HV networks (also called sub-transmission) are quite similar to transmission networks. The topological
design of the grid is meshed and may be operated as radial or meshed depending on the situation. HV
networks are operated in a similar way all around Europe: N-1 or N-2 contingency criteria12
are usually
in place for rural and urban areas, respectively. The monitoring level at HV is very high. DSOs operating
HV grids are able to supervise and control the HV network from the control room centres.
 MV distribution networks significantly differ in their characteristics with respect to urban and rural
grids. Mostly, meshed topology is used that can be operated either as meshed (closed loop) or radial
(open loop). In some countries or depending on the network type in a region, MV operation may
always be radial. A high density of loads and relatively high demand typically causes high equipment
load factors (transformers, cables) for urban areas. Rural areas are characterised by larger geographical
coverage and lower load density and thus longer lines, high network impedances and lower equipment
load factors. The proportion of European MV networks with remote monitoring, control and
automated protection/fault sectionalisation is currently low but increasing by necessity.
 LV networks are usually radially operated. Similar to MV networks, urban and rural LV networks have
different characteristics. The proportion of LV monitoring and control is typically even lower than in
MV. Measurements usually rely on aggregated information from substations and are only available
with a significant time lag. Profile information will not be available.
2. Active Distribution Networks
Once DG in distribution networks surpasses a particular level, building distribution networks able to supply
all load & DG within the existing quality of service requirements will frequently be too expensive and
inefficient. For example, in many places the network would only be constrained for few hours per year. In
addition, the security of supply and quality of service problems will no longer be limited to specific
situations.
Integrating the high amount of existing and projected DG and, later, other DER will require new ICT
solutions and an evolution of the regulatory framework for both network operators and users. Network
planning and operation methodologies need to be revised to take the new solutions into account.
2.1. Key Building Blocks
There is no one-size-fits-all solution because distribution networks are rather heterogeneous in terms of
grid equipment and DG density at different voltage levels. Every distribution network should be assessed
individually in terms of its network structure (e.g. customers and connected generators) and public
infrastructures (e.g. load and population density). Nevertheless, the needed development towards future
distribution systems which meet the needs of all customers can be described in the three schematic steps
pictured below: from (1) passive network via (2) reactive network integration to (3) active system
management.
12
Rule according to which elements remaining in operation after a fault of a distribution system element must be
capable of accommodating the new operational situation without violating the operational limits.

11
Figure 9 Three-Step Evolution of Distribution Systems
(1) Passive distribution networks make use of the so-called ‘fit and forget’ approach. This approach
implies resolving all issues at the planning stage, which may lead to an oversized network. DSOs
provide firm capacity (firm grid connection and access) that may not be fully used anymore due to
local consumption of the electricity produced by DG. This approach has the advantage of requiring
low flexibility, control and supervision, but is only possible for a network with very low DER
penetration. Once DER penetration rises, the system cannot be designed to cater for all
contingencies without very significant investment in basic network infrastructure, making this
approach less economical.
(2) Reactive network integration is often characterised by the ‘only operation’ approach. This
approach is used today in some countries with a high share of DG. The regulation requires
connecting as much DG as possible with no restrictions. Congestions (or other grid problems) are
solved at the operation stage by restricting both load and generation. This solution could restrict
DG injections during many hours per year and lead to negative business case for DG if they are not
remunerated for the restrictions. Already today, some ‘front-runner’ countries with high DG
penetration levels can be considered as having reached the interim ‘reactive network integration’
stage at which DSOs solve problems once they occur (largely only in operation).
(3) The active approach would allow for interaction between planning, access & connection and
operational timeframes. Different levels of connection firmness and real-time flexibility can
reduce investment needs. The existing hosting capacity of the distribution network can be used
more optimally if other options including ICT, connection & operational requirements guaranteeing
adequate performance of DER towards the system (i.e. via grid codes) and market-based
procurement of ancillary services from DER are considered. Operational planning of distribution
networks (similar to that at transmission level) would be in place in networks with high DER shares
in order to incentivise dispatch in a way that is compatible with the network. Improved network
capacity planning and congestion management at distribution level at different times and locations
will be required to maximise the level of generation which is injected in the most economical way
for all parties, while maintaining network stability. DSOs must have tools for overseeing
maintenance of network standards. Additionally they should have the possibility to buy flexibility
from DG and load in order to optimise network availability in the most economic manner or to
manage network conditions which are beyond the contracted connection of the customers. DSOs
should have the possibility to buy flexibility from DG and load on so-called ‘flexibility platforms’ in
order to solve grid constraints. The network reinforcement could be deferred until the moment
when it becomes more cost-effective than the on-going cost of procuring services from DER.
Interactions between DSOs, TSOs and market actors at different stages are depicted in
(4) Figure 10.

12
Figure 10 DSO interactions with markets & TSO at different time frames
Using the active system management approach would allow maximal integration of DER, making the
most of the existing grid while enabling DSOs to fulfil the security standards and enabling DER to find the
right conditions for their business plan in the most cost-effective way.
Table 1 highlights the key features of each phase, broken down into the different ‘layers’: development &
planning, operations, information exchange and technical development. The subsequent sections of the
report discuss the active system management approach within the individual layers in detail.

13
Networkdevelopment
(planning,connection&
access)
Fit and forget approach:
everything ‘solved’ at the
planning stage
Only operation approach:
connection with no
restrictions and solutions at
the operations stage
Or
Fit and forget approach
Combined planning and
operational solutions:
Active capacity and loss
management through
commercial interaction with
market actors selling
flexibility services
Connection and access
criteria combined with
operation tools to manage
DER
Flexibility support from DSO
to TSO and from TSO to DSO
when required
New system services for
DSOs arranged via
commercial ancillary
services and grid codes.
Networkoperation
Low monitoring & control of
DG RES, often only by the
TSO
Missing rules & services for
DG contribution to quality of
service, security of supply &
firmness
Emergency generation
curtailment by DSO
Active voltage control
distribution networks. Grid
codes for DG to meet
connection criteria and be
able of voltage based
control and reactive power
contribution
Information
exchange
Little information exchange
from TSOs/DER to DSOs
(small DER do not send
information)
High-level information
exchange from TSOs/DER
to DSOs
Structured and organised
off-line and where needed
real-time information
exchange (standardised
interfaces with DER
required)
Technicaldevelopment
Network
Limited monitoring & control
capabilities (usually only HV)
Conventional SCADA for HV
network and DMS/OMS for
MV and LV
Increased monitoring and
control at HV & MV via
telecommunications
SCADA/DMS/OMS with the
measurement of certain
new DG
Increased monitoring,
simulation and control down
to LV via
telecommunications
Advanced Distribution
Management Systems13
for
DSOs/
SCADA and Distribution
Management System (DMS)
DER
DG often not prepared for
power factor control
Storage & EV not developed
Enhanced DG protection
systems/ inverters enabling
voltage & reactive power
control
Configurable settings: e.g.
protection / fault ride
through settings, voltage
droop
Presence of storage & EVs
Table 1 Three-Step Evolution of Distribution Systems in detail
13
New SCADA and DMS/OMS could be integrated in a single system called Advance Distribution Management System
(ADMS).
Layer
Passive
Distribution network
Re-active distribution
network integration
Active distribution system
management

14
2.2. Distribution Network Development, Planning, Access & Connection
2.2.1. Coordinated Network Development
DSOs should be able to plan their grids well in advance to prevent bottlenecks in the most cost-effective
way. Data acquired within distribution network monitoring and information exchange with TSOs and
distributed energy resources (see sections 3.4 & 3.5) could be very beneficial in this respect.
In addition, every connection request should be analysed and considered in the planning process in order
to make the best of the existing network. According to the traditional regulatory approach to connection
requests analysis currently applied in most countries, the network operator performs an individual analysis
and provides an individual solution to each connection. The first connections may make use of the available
capacity of the existent network. But once there is an increased demand for new DG connections in the
same area and the available network capacity is limited this approach is not always optimal from the
overall cost and network development perspective.14
One way to tackle this issue is to allow for coordination of all relevant actors, including network
operators, investors and local authorities in the analysis of connection requests (see Box 3 for Spanish
example).
BOX 3: Coordination for efficient RES and network development: Spanish “Evacuation Boards” approach
To rationalise the RES expansion and optimise the available energy resources, some Spanish regions
created so-called "Evacuation Boards". They are characterised by a coordinated grid connection request
process. RES installation plans are deployed and coordinated between the administration, RES investors
and transmission and distribution system operators. In these evacuation boards the TSO or DSO do not
receive individual requests; they are collected by the Regional Administration and after a validation process
submitted for an aggregated analysis to be made together by the DSO & TSO. The positive impact of the
new networks for consumption (extra capacity for consumers) is also considered. In addition to the cost-
sharing mechanism (proportionally to the capacity assigned to each RES project), the covenants for the
development of such infrastructures contain the necessary guarantees, payment and execution terms.
Benefits of this approach include overall minimised network development and project cost, reduction of
project risks thanks to the possibility to correctly analyse both the costs and timetables needed for the
different RES penetration scenarios, and reduced time for acquiring all necessary administrative permits.
Coordination between TSOs and DSOs is likely to play a particular role. Whilst in some cases modifications
required by DSOs from TSOs or vice versa do not considerably affect the capabilities of one or the other to
maintain their network performance, the impact may be substantial in other cases. For example, when the
HV or UHV (ultra high voltage) network is saturated, connection of generation to the MV network cannot
be planned without taking into account the conditions at HV network. An optimal network development is
also key to minimise losses in the electrical system. Transmission or distribution network conditions which
require regular (or conditional) exchange of information between TSOs and DSOs should be defined.
Standard reciprocal data exchange arrangements about the expected development of generation/load
at the different voltage levels and about the network reinforcements needed at the TSO level, not
directly related to the lower voltage planning activities, should be put in place.
14
First requests for connection get the available capacity of the existent network at a low cost, but as they increase
they require more complex and expensive network development solutions. In countries where generators bear grid
connection as well as grid reinforcement/extension costs (deep connection charges), this may make the individual
projects economically unviable. In countries where generators bear grid connection cost but not the grid
reinforcement cost (shallow connection charges) and do not pay any use of system charge, those reinforcement
network costs are socialized.

15
Coordinated planning as per the Irish “Gate” process (see box below), batch planning of DG connection
applications at TSO and DSO level as appropriate, may see the most economical net cost of connection.
Coordination between the TSO and DSO in Schleswig-Holstein (northern Germany), where 9,000 MW of
wind should be connected by 2015, are another example in this respect.
BOX 4: Coordination in planning between TSO & DSO - The Group Processing Approach in Ireland
Up to 2003, applications to system operators for connection of wind were processed sequentially. Since
2004, following a moratorium on wind connections, DG is planned in ‘Groups’ based on their network
location and capacity and processed in batches known as ‘Gates’. This system was introduced to deal with
the massive number of applications. The queue for connection comprises both DSO and TSO applicants.
Applications are processed by the system operator (TSO or DSO) most suited to their group connection
irrespective of which system operator they originally applied to. The connection method is determined by
the Least Cost Technically Acceptable (LCTA) method for the defined sub-group. The cost of the network
reinforcement for the group is shared by the different wind farms proportionally to their capacity.
Connections below 500 MW have a different planning process and are treated as individuals with a quicker
process, reflecting their connection generally being less onerous and costly. The regulator (with input from
both system operators and wind industry) determines criteria for eligibility for inclusion of “Gate”. This
method facilitates large quantities of RES being processed in a structured manner and rationalises
allocation of scarce capacity. Transmission and distribution systems are developed in co-ordinated fashion.
2.2.2. Network Capacity Management
Network capacity management would encompass optimisation of network capacity via improved
consideration of DERs in network planning. The DSO ability to identify areas with possible overload
problems well in advance as described within the active approach is a precondition for this.
The options that should be further investigated include:
 New network access options such as variable network access contracts; and
 Alternatives involving close to real-time operation solution such as DSO tendering for a flexibility
service.
Variable access contracts
DG developers could have a possibility to select firm or variable access contracts based on their own
business plan. Variable network access rights could be offered as a discounted connection contract for
generation customers, with pre-defined mechanisms for DG to reduce their output to a predefined limit in
infrequent situations, expected only for few hours per year. If only several hours of re-dispatching per year
are needed to limit peaks of production and use network capacity more efficiently, those would be more
than offset by an additional DG output in all other hours due to a higher installed DG capacity up to a
certain point where the cost of net losses and curtailed generation become relevant to justify network
reinforcement.
A model example in Figure 11 illustrates that limiting peak power in-feed of e.g. 5% of the time would give
an opportunity to connect 220% more DG. In these cases, generation operators should be incentivised to
choose this option, e.g. by reduced connection cost (cheaper but with limited guarantee of injection within
a clearly defined framework) or other form of compensation. This could be executed either via direct
contracts between DSOs and generators/load or indirectly between DSOs and aggregators who would pay a
yearly option premium to DG/load and then offer flexibility to the DSO.

16
Figure 11 Variable access approach (Source: EWE Netz)
Today, variable access is precluded by obligation to compensate generators for any energy they are not
allowed deliver in many jurisdictions. As outlined above, curtailment is often possible only to deal with
short duration constraints; it is only temporary and automatically triggers grid adaptations such as
reinforcement if they are deemed economically justifiable by the connected parties or DSO. A form of
variable network access for DG/RES exists e.g. in the UK (known as non-firm access). The conditions where
a DSO can issue a curtailment instruction are set out in a connection contract agreed at the time of
connection in return for a lower cost of connection. For example, the feed-in management rules in
Germany (see box below) define a flexibility obligation in form of a capped connection.
BOX 5: Mandatory integration of feed-in management system in Germany
The flexibility obligations, including the technical and regulatory possibility to be curtailed are spread over
all production facilities above a certain capacity. The PV installation owner can choose to install a technical
receiver device allowing for feed-in reduction by the network operator or to reduce the feed-in power to
70% of the nominal power (installed power). In practice, this measure leads to the loss of about 5% of
energy feed-in from PV but allows connecting more DG to the network (and thus overall increased DG
production). The producers are compensated for lost production. This curtailment regime enables an
optimised use of the existing network, without jeopardising the business cases of new producers. It
applies until the relevant grid development is made.
Practical modalities of such arrangements may differ from one country to another. In any case, the
following aspects would have to be considered before implementation of a variable access scheme:
 It might be necessary to review the value of the discount from time to time, or it might be possible
to have an “auction” giving a kind of merit order list (i.e. those who pay most for the grid use would
be curtailed less often). A risk of changing conditions over time could be mitigated, e.g. by
participation via aggregators.
 DG developers should be provided with information on expected curtailment so that it can be
included in the risk analysis and economic viability analysis of projects prior to investing in these
projects. This should be based on the best available information and analytical techniques, to
ensure that generation developers can base their decision on an informed business plan.
 Establishment of procedures for assessing when the investment deferral is more cost-effective. As
expectations change depending on the amount of RES already connected and the amount of RES
that will be connected later on, it might also be necessary to have insight under what conditions
the DSO will undertake reinforcements, because additional RES might be added at each
reinforcement (depending on the “geographical” possibilities). If RES developers have sufficient

17
transparency, they will take possible interruptions into account, as well as the value they need to
earn back on a flexibility platform.
 The manner in which to curtail generators (pro rata, market based etc.) and the operational rules
for the constraint location where more than one generator contribute to the constraint would have
to be specified.
‘Firm DG production and load off take management’ via ahead tendering for commercial services
Assessment of firm capacity is important for network operators to foresee how much generation can
contribute to peak consumption. A mechanism incentivising DG to generate or the consumption to stop
consuming when the peak on the network takes place would enable more efficient use of the existing
distribution assets and deferral of grid reinforcement, as illustrated in Figure 12. Provision of certain firm
capacity would be rewarded as an extra service for the system. Remuneration for this service could be
determined through an open commercial tendering process organised by DSOs, e.g. via a local auction for
necessary services (DG to bid certain amount of capacity reliably available). In any such cases, the party (DG
or other commercial operator) deemed reliable and contracted to deliver certain firm capacity must be
held financially responsible for its delivery.
Figure 12 Extension of network capacity for peak load versus a load management solution (Source: EWE Netz)
All in all, due to various generation mixes, distribution of generation over different voltage levels and
geographic distribution of the resources, a one-size-fits-all approach may not be adequate from the social
welfare point of view. The option of giving network customers a choice between firm injection and higher
connection charges and non-firm injection and considerably lower connection charges should be assessed
against other solutions.
2.2.3. Connection
For proper integration into the network, distributed generation needs to fulfil minimum technical criteria:
the equipment and its protective relays must be able to resist voltage dips and prevent islanding and there
should be separate metering for production and consumption. DG should also bear the same costs as other
generators, including adequate connection fees. DSOs must know what is on-line when they are working to
prevent accidents. Therefore, they should have a possibility to verify compliance with requirements.
Distributed generation should thus ‘be registered’ with the DSO, and remote disconnection by the DSO to
prevent damage to facilities of other clients while manoeuvring should be technically possible under
conditions defined by regulation. To secure safe operation of the distribution grid, DSOs should also be able
to define control schemes and settings for generators connected to their grids, in coordination with the
TSO where necessary in order to ensure compliance with overall system requirements.

18
Recommendations:
 In order to make the best of the existing network, all relevant actors should be involved in the
analysis of connection requests. Network development coordination between TSOs and DSOs should
be enhanced. Practical modalities of these arrangements may differ from one country to another.
 Future production on DSO level should be taken into consideration when developing the ENTSO-E Ten
Year Network Development Plan and the relevant national plans. DSOs should be consulted
accordingly.
 Market-based network capacity management options such as open commercial tendering processes
or optional variable access contracts should be further investigated as an initial tool for
accommodating large amount of DG in situations when it is proven to be more cost-effective than
waiting for provision of connection and access to the grid until any curtailment is ruled out.
 Distributed generation needs to fulfil minimum connection criteria.
2.3. Active Distribution Network Operation
Distributed generation should be incentivised to sell their production into the electricity market15
.
Aggregation of DG in form of so-called Virtual Power Plants (VPPs) or of distributed generation, flexible
loads and possibly decentralised storage is expected to play an important role in facilitating access of small
customers to the market and addressing the uncertainty of availability and providing enhanced capability to
manage the risk of not being able to meet the contracted scheduled output. The aggregator role could also
be taken up by electricity service companies (ESCO) or suppliers. The aggregator would provide an interface
between DER and other market actors and system operators. In addition, DG should be obliged to meet
scheduling, nomination and balancing obligations as other power generators do, including payment of
balancing charges. DG should also be responsible for their imbalances on equal terms with other Balancing
Responsible Parties. It is highly beneficial for system stability and cost reduction if variable RES technologies
are incentivised to reduce forecast errors and to minimise imbalances in the market and take up necessary
responsibilities towards the system as other generation technologies do.16
The network plays an important service role of supporting the market. Operational barriers may arise,
characterised by one or more violations of the physical, operational, or policy constraints under which the
grid operates in the normal state or under contingency cases. They are transient – associated with a
specified point in time. As such, they may be detected before or during the day-ahead, the hour-ahead
markets or during real-time system operation. In order to facilitate secure network operation and smooth
functioning of the future market with high DER penetration, DSOs need to become active operators of
their networks. This means that they need adequate tools to operate their networks. In addition, network
users need to actively participate in network usage optimisation. In this way, the possible abovementioned
violations can be eliminated.
15
Today, small DG typically sells their electricity output at fixed prices to the TSO, an electricity supplier or other third
party market participant. Different economic support schemes for the production of electricity from RES and
combined heat and power (CHP) have been implemented at national level. Where fixed feed-in tariffs are in place,
generators are exempted from market signals & prices. Support schemes shall be made more market-oriented, based
on production rather than investment. Support schemes which expose RES generators to market prices are more
compatible with well-functioning of electricity markets, thus providing the correct price signals.
16
These obligations can mean new costs (which should be taken into account in adjustment of the support
mechanism), but can also mean new sources of revenues. See EURELECTRIC report RES Integration and Market
Design: Are Remuneration Mechanisms Needed to Ensure Generation Adequacy? May 2011.

19
2.3.1. DSO System Services & ‘Traffic Lights’ Approach
Active DSOs should be allowed to coordinate the offering of new system services, as required by the new
Energy Efficiency Directive (Art 15.1 of 2012/27/EC) while ensuring the security, integrity and quality of
supply in their networks. The DSO is best placed to facilitate this mechanism as the data need to be
gathered at substation level and in-depth knowledge of the grid layout and its behaviour is required.
Moreover, the DSO has a legal responsibility to ensure that such technical constraints are mitigated.
The blue area in Figure 13 schematically outlines the stages in the electricity market in which the network
operators do not interfere, but act as mere behind-the-scene facilitators. The green area depicts the system
services that are to be administrated by network operators (for both transmission and distribution system
needs). Such system services could be defined in grid codes (voltage and reactive power contribution) or
procured as ancillary services from DER within a transparent and non-discriminatory regulatory
framework.17
Table 1 provides a more detailed overview of such system services including their possible
form of delivery.
Today, ancillary services are procured by the TSO, largely from large power producers, to manage the
system as whole. In future, flexibility platforms where flexibility is offered (usually via aggregators) to
DSOs for relieving congestion in their networks but also to TSOs to provide balancing and redispatching
in transmission grids will play an important role, in particular for close to real-time flexibility.
Figure 13 Market and network operations
17
For definitions see Glossary.

System service Purpose Provision by which type
of DER
Information exchanges Form of delivery
Firm capacity
management (long-term)
DSO planning purposes;
optimize of network capacity
utilisation DG with a view to
using assets most efficiently
CHP, small hydro,
stochastic RES with
integrated storage
Aggregated demand
providers, DSM
 DG outage programs and availabilities information (DG->DSO)
 Real time generation output (DG->DSO)
 Real time demand flexibility information (DER->DSO)
 Firmness periods (DSO->DER)
Commercial
Losses compensation Increased system efficiency DSO, DG, demand
customers
 Real time load and network voltage or fault conditions (DSO->DSO)
 V, Q, pf setpoints (DSO->DG)
 Demand reduction signals (DSO->Aggregators)
Commercial
Security congestion
management (short-
term)
Operate the grid within the
security standards
RES, CHP, distributed
storage, DSM
 Real time load and network voltage or fault conditions (DSO -> DSO)
 Real time generation output & load flexibility (DG -> DSO)
 Reduced setpoint/ reduction signal (DSO -> DG)
 DG outage programs and availabilities information (DG->DSO)
Mandatory with compensation or by
commercial arrangement (non-firm
access contracts)
Anti-islanding operation Avoid unsafe, unbalanced and
poor quality distribution
electric islands
DG, storage, DSO (local
network controls)
 Local automatic signal generated in case of fault or triggering conditions -
> all local DG, storage, network control points
 Local signal generator -> DSO SCADA or central control (and local /
regional control depot), notification signal by DSO
Mandatory without compensation (grid
connection rules defined in grid codes)
20
Frequency control Under extreme situations of
system strain, TSO to call upon
DSO to deliver support – DSO
as a conduit but needs to see
what is happening
DG & load to TSO via DSO  Real time active and reactive power flows information exchange at the
T/D interface (DG & load -> DSO ->TSO)
 Load/generation to adapt (TSO-> DSO -> DG & load)
Commercial (at TSO level)
Islanding operation Improve continuity of supply
when higher voltage source is
unavailable
DG, storage, DSO (local
network controls), DSM
 Real time active and reactive power flows information exchange (DER ->
DSO)
 V, P, Q setpoints (DSO->DER)
Mandatory with compensation
DSO Voltage control Local supply quality security
and increasing amount of DG
power that could be injected in
the grid
PV, Wind power, CHP,
distributed storage, DSM
 Reactive requirement (amount and electrical or geographical delivery
location) (TSO -> DSO)
 Real time load and network voltage or fault conditions (DSO -> DSO)
 Real time generation output (DG -> DSO)
 V, Q, pf setpoints (DSO -> DG)
Mandatory without compensation to
maintain defined limits for distribution
system stability. Commercial for
purposes beyond maintenance of
network stability or outside the scope
of the customer’s own connection
Information exchange Optimise DSO and TSO control
supervision, and scheduling
DER, TSO  TSO Real time and off-line measurements and topology information
(TSO -> DSO)
 DG and TSO outage programs and availabilities information (TSO->DSO).
 DG generation forecasting (DG-> DSO-> TSO)
Mandatory with compensation
Table 2 New System Services at Distribution Level

21
To manage the operation of distribution systems, similarly to the usual practice in transmission networks,
basic system states should be defined, e.g. within security standards or grid codes. A “traffic light scheme”
could be used to distinguish between system states and selecting appropriate type of actions (see Figure 14
and the box below):
 The ‘secure operating region’ (1) results from overlaying the stable operating points on an
illustrative power-voltage curve for abstracted two-bus system. It represents an area where the
network is secure and the power can flow either towards the consumer (because demand is
greater than generation) or back to the system (when generation exceeds demand). If the system is
at the secure operating point 1, the market operates and the DSO has clear visibility of this. The
boundaries of this secure operating region depend on the physical characteristics of the network
and system dynamics.
 ‘Unsecure operating points 2 and 4’ illustrate the situations hampering secure operation of the
distribution system: voltage increase and congestion (the plane has been rotated to align it with a
time-varying load profile).
 ‘Operating points 3’ indicate alert network states in which congestion management should be
conducted in order to keep the grid from entering the red emergency state. When possible the DSO
actively engages with the market (DG or load) to procure flexibility to relieve grid constraints.
All in all, market principles should work in the first place when the system is in the ‘green’ and ‘yellow’
states. That should be the case most of the time. DSOs should override control (directly or indirectly) of
supply or demand only in strictly defined emergency cases (when the system enters into the red state).
Actions in this state must be specific and well defined by e.g. regulation or contracts and be temporary in
nature. Transparency is needed (each time of occurrence and reasons); and analysis should be made if
indeed DG/load were not able to offer sufficient services.
Figure 14 Instability in the distribution system

22
The Traffic Light Concept
1. Green: Normal operating state
2. Yellow: Alert state – DSO has an emerging congestion. DSO actively engages with the market (DG
or load) to procure flexibility to relieve grid constraints:
a) Via flexibility platform: a methodology has to be developed which links the offers of
aggregators with identification for the location. For example, aggregators could divide their
Virtual Power Network capacity into local pools. A DSO can contract an aggregator for
delivering local generated electricity or load from the local capacity pool. This can be
facilitated by the DSO data hub.
b) Directly: A DSO could contract the customer (including DER) in order to maximise the
utilisation of distribution assets in planning and operational timescales where locally
needed with ensuring transparency and non-discrimination for example via tenders.
3. Red: Unsecure operation – Emergency cases
In strictly defined emergency cases, the DSO would be able to manage distributed renewable
generation, to implement grid efficiency improvement measures, and to control the isolation and
restoration of outages. This is the last measure taken by the DSO when every other option has
failed to restore system integrity.
18
Distribution congestion
Security congestion management would be used to solve the technical constraints at distribution level
close to real-time in the alert state (‘yellow light’). In these defined cases, the DSO could deviate from the
merit order, but ex-post justification and compensation should be provided. This could be executed by pre-
agreed contracts for instance. The DSO should then pay the “up regulating” cost elsewhere in the system,
and should remunerate the downward cost to the local generator that is constrained off in his grid
(including the missed income from the support scheme and the costs of keeping balanced position).
18
For definitions of CLS and DMS see glossary.

23
Emergencies
Emergency tools including direct load management (load shedding) and emergency DG curtailment
should be used only in well-defined emergency states/once the contracted options are exhausted. When
the grid stability is at risk, DSOs should be able to act physically to control and constrain off both local
consumption and production (as is already the case in some countries – e.g. Germany or Sweden). Priority
access rules should not restrict network operators’ ability to flexibly respond to emergency situations.
Any action on distribution network users requested by the TSO should be agreed with the DSO as system
operators. TSOs should not act on any individual DER connected to the distribution system. Any direct
order from the TSO to DER embedded in distribution networks targeted to safeguard operation of the
system will be executed by the DSO, not the TSO.
2.3.2. Information Exchange
Today, DSOs have no systems installed for acquiring data from DG of smaller size in particular. In some
cases, the TSO receives information from DG in real time while DSOs do not have real-time access to this
information. There is not usually an operational exchange between the TSO and the DSO.19
In the future, a well-structured and organised information exchange between relevant actors will be
necessary to operate the distribution network with high DG penetration in real-time or close to real-
time. Such information exchange will be used for DSO planning and asset management purposes (to
optimise network capacity and availability in the most economic manner) while ensuring that all
customers feel the absolute minimum impact of DG on power quality and continuity. It is the first and
potentially most effective step which can be taken to reduce the cost of DG connection and integration.
At the transmission level, generators already send schedules to the TSO for system balance purposes and to
gurantee that their realisation is technically possible. In systems with high DG penetration, the DSO will
need information about DG forecast, schedules and active dispatch to improve their visibility and to assist
with real-time or close to real-time management of the distribution network including local network
constraints. DSOs should have managed access to communication and monitoring assets of DG to collect
information that will be necessary for operation of their networks. The granularity of the data exchange will
depend on the size of the generating unit. The necessary information should be then exchanged between
the DSO and the TSO (in both directions). DSOs should provide the TSO with information on active power
that the TSO needs to facilitate secure system operation. DSOs operating sub-transmission networks may
also require TSO information in real time.
Figure 15 Information Exchange Today and in the Future
19
Exchange between TSOs and DSOs operating HV networks exists in some countries.

24
In addition the fact that the participation of DG and load connected to the distribution grid in TSO
balancing markets could create constraints in the distribution grid should be taken into account.
Generators and flexible loads connected to the distribution grid should be able to also provide these
ancillary services for system balancing and transmission system congestion management purposes.
Aggregators may be able to achieve certain modifications in the demand and in the generation of the
consumers and producers in their portfolio, in order to offer services to system operators. However, the
issue that an area managed by an aggregator may not correspond to the distribution network responsibility
area (forecasts and schedules are delivered at a portfolio level for a bidding zone) needs to be addressed. In
these situations, the DSO needs to have adequate visibility to ensure that this activation does not impede
security of supply in its own network. Similarly, actions by the DSO to solve constraints could affect the
system balance.
Appropriate methodologies and processes should be defined in order to ensure that market schedules
are not in conflict with network operation or that transmission and distribution network operation are
not in conflict with one another (e.g. TSO asking for a modification which would violate distribution system
security standards). Enhanced monitoring and control strategies for distribution networks will need to be
deployed (see section 2.4).
2.3.3. Voltage Control
As outlined earlier, voltage in the system needs to be maintained within a range defined by security
standards. Voltage control is a system service managed by network operators in order to maintain voltage
in their networks within limits and to minimise the reactive power flows and consequently, technical losses.
While generation/load balance is carried out at system level by the TSO, voltage control of the distribution
grid requires the involvement of the DSO.
The traditional approach to voltage control includes reinforcing the grid or installing preventative
measures. Voltage control has been traditionally done by transformers (using on and off load tap changers
moving reactive power) and capacitor banks that inject reactive power into the grid (see Figure 16). The
DSO fixes a setpoint and prepares scenarios/ranges for different voltages within which the voltage must be
maintained.
Figure 16 An illustrative example of a traditional approach to voltage control

25
HV  Usually power transformers with on-load tap changers
 Capacitors frequently used to control the voltage
MV  No analogue values in secondary substations obtained in real time (typically analogue
values only at MV feeders)
 MV networks connected to HV through a power transformer with on-load tap changer
Capacitors can be commonly found also in these substations to improve the power factor
 Voltage setpoints specified at MV substation busbars
LV  MV/LV transformers may be fixed ratio (i.e. have not tap changers) or have off-load tap
changers, manually controlled. Taps selected to compensate the effect of MV voltage
drops at LV levels (passive approach).
 These tap-changers operate only a few times during transformers lifetime.
 Sometimes capacitors are installed in consumer facilities to meet power factor
regulations
 New controllable MV/LV transformers are emerging but expensive
Table 3 Voltage control in current distribution networks
As the penetration of DG in networks increases, it is no longer possible to ensure sustained system
security without some dynamic resources, including reactive power compensation and active voltage
control. The DSO’s role in ensuring not only that adequate network capacity is available for all connected
customers (demand and load) under all conditions, but also in managing voltage and reactive power
flows on distribution networks is becoming more complex.
Two technical considerations for voltage control in distribution networks include:
 With high DG penetration, active power becomes a significant driver for voltages changes in MV and
LV networks (the kW-V effect is more significant than the kVAr-V effect). The lower the distribution
voltage level, the higher this effect.
 In MV and LV networks, the active power effect may not be always neutralised by the reactive
power injections/withdrawals available in the system. To regulate voltage, DSO should be able to
control reactive power and in some cases also active power.
In high-voltage networks, DSOs will be able to maintain the voltage within the security standards if they
have the means to manage reactive power flows.20
Active power will not cause voltage deviation under
normal operation. On the other hand, at medium and low voltage levels, active power changes due to DG
feed-in cause voltage rises (especially in cables).21
Distributed generation changes voltage triangle design
and as a result, both scenarios with and without DG have to be considered. Reactive power may help
compensate the active power effect but may not be sufficient to neutralise it.
As reactive power cannot be transported over long distances and as many regions with high DG penetration
(thus voltage control challenges) have no conventional sources of reactive power, managing voltage on a
local or sub-regional basis could be the most economically viable solution for the entire electricity system.
For these purposes, the DSO should be able to sectionalise its networks, to actively interact with DG to
request supply voltage control and reactive dispatch or to exploit active demand services over the
distribution network to solve operational problems. Clear interface conditions and an agreed operational
framework at the interface between TSO and DSO are also necessary (including parameters for TSO-DSO
interface nodes).
20
Through issuing suitable voltage set points and operating power factors to DG and operating network reactive
power or voltage installed resources.
21
The R/X ratio (ratio between the resistance and the reactance of overhead lines or cables) explains the relationship
between active and reactive power and the voltage level. See Case study A in Annex 2.

26
Innovative approaches to voltage control should be explored. Experiences from several European
countries show that compensation at the TSO-DSO connection point may not be flexible enough in case of
emergency. The increasing DG penetration requires continuous adaption of the capacities at the
connection points. Voltage control requires a system approach that would include minimising system
losses, optimising network investments and coordinating the necessary operational windows and control
actions between the individual TSO and DSO, with particular attention to cases of emergency.
Solutions on the network side and on the network users’ side should be considered:
Network users’ contribution
Generators could contribute to voltage based control. This is already the case in some countries and is
being tested in others, as shown in case study B in Annex 2. The best generation technologies for voltage
control are those that are able to absorb and deliver reactive power. These include modern wind turbines,
small hydro and power electronics. Inverters originally supported neither active or reactive power but
today address reactive power and have also elements for frequency support. For example DERs would react
to voltage changes/disturbances or faults by altering real or reactive power based on pre-defined and
dynamically configurable set-points (similarly to how frequency control operates at TSO level today).
Network contribution
There are cases where HV voltage control cannot be achieved in a manner which maintains both
transmission and distribution security without additional reactive power resources on the network.
Investments in new analytical, control and monitoring expertise and facilities in MV and LV (like power
electronics, real time supervision) may be required. For example, satellite cables with large diameters could
drastically reduce voltage rise within the grid.22
Alternatively, voltage controlled MV/LV transformers can
be installed in order to decouple the voltage rise within the MV and the LV networks caused by RES feed-
in.23
These innovative network techniques would need to be considered and allowed within regulation.
Combination of both solutions
A combination of both solutions may be used. Studies show that in some situations even a contribution
by distributed generation will not be able to maintain the voltage within limits. Case study A in Annex 2
demonstrates that situations when the required reactive power exceeds the installed active power capacity
may occur. Both to minimise any impact (active power generation) on DG and to ensure that TSO-DSO
interface standards can be maintained for system stability, it may thus be necessary that additional reactive
resources are installed on distribution networks. The contribution of the system operators or DG to these
resources must be reflective of the particular network issue being addressed and the beneficiaries of the
installation. It may also be necessary in future for DG which will displace reactive power resources to also
contribute to the reactive power facilities on its hosting networks so as to maximise its potential to deliver
active power.24
This is already the case with larger (industrial) demand facilities that are required to meet
strict power factor regulations. Where this solution is deemed the most cost-effective, generators should
be required to comply with minimum connection & operational requirements necessary for managing
distribution network stability. On the other hand, DG contribution to loss minimisation may be
compensated on a commercial basis.
22
Werther, Becker, Schmiesing, Wehrmann: Voltage control in low voltage systems with controlled low voltage
transformer (CLVT); CIRED Workshop - Lisbon 29-30 May 2012.
23
Thies, Zdrallek, Schmiesing, Schneider: Future structure of rural medium-voltage grids for sustainable energy supply,
CIRED Workshop - Lisbon 29-30 May 2012.
24
Traditionally reactive power was supplied, rather than absorbed, by generation. As DG displaced conventional
generation, reactive power resources are reduced.

27
Reduction of active power that might be technically necessary to manage the network voltage and avoid
complete generation disconnection should be used only when other solutions have been exhausted.
Congestion management would apply in this case, with emergency DG curtailment only if necessary.
Recommendations:
 Active DSOs should be allowed to coordinate the offering of new system services. Such system
services could be procured as ancillary services from DER or defined in grid codes (voltage and
reactive power contribution).
 Flexibility platforms where flexibility is offered (either directly by larger DG/load or via
aggregators which group a large number of DG and loads) to DSOs (but also to TSOs to provide
balancing and redispatching in transmission grids) could play an important role for close to real-
time flexibility in particular.
 DSOs should be included in the information exchange about
o planned location / connection / clustering of contracted customers (aggregators)
o forecasts and schedules necessary for dealing with local grid constraints.
Appropriate methodologies and tools for these purposes should be developed.
 Voltage control requires a system approach considering transmission and distribution networks.
Coordination between DSO and TSO at the interface point should aim at reactive power
optimisation and minimising system investments and losses.
 DG contribution to voltage control, probably in combination with network solutions, is likely to
play an important role in keeping distribution networks stable. Where this solution proves to be
the most cost-effective one, generators connected to the distribution network above a certain
capacity should be required to have reactive power capabilities in line with those for transmission
connected RES, adapted to the connection level and capacity. Where it goes beyond maintaining
local system security as impacted by the generation connection, a market-based approach should
be adopted for procuring additional amounts of reactive power such as DG contribution to
minimise losses.
2.4. Technical Development: Towards Flexible Distribution Systems
With a rising share of DER, monitoring, control strategies and advanced protection systems in MV and LV
distribution networks will have to develop to enable DSOs to:
 Determine and forecast grid capacity and bottlenecks;
 Supervise and control power flows and voltage in their networks;
 Enable the new operations at the distribution level (including non-discrimination and effective
real-time grid capacity monitoring and management of injections/withdrawals);
 Enable market-based congestion management and voltage control in distribution networks;
 Enhance the distribution grid hosting capability.
This solution should not to be taken up in isolation from other options such as services from DERs to help
manage the distribution system mentioned above. The route to the future energy system will incorporate
both aspects. The most economically efficient solution will depend on the topology, load and generation
profiles within a given distribution system. A two-level approach might be adopted: larger DG (with a
defined installed capacity and level of connection) could be monitored and telecontrolled while other DER
may not be dispatchable should be forecast and monitored by the DSO on an aggregated basis at the
substation level.

28
Technology options for development of the abovementioned options are outlined in Table 4. For examples
of implementation of different solutions see case studies C, D & E in Annex 2. In addition, new design ideas
such as “satellite cables”25
and dynamic line rating (DLR) are being tested. With DLR, real-time information,
for example the temperature of the conductor or the wind speed, can be used to calculate a temporary
rating of a line, thus allowing more power to flow. This technology increases the utilisation of the line and
also allows more RES to be integrated than if static ratings were used. Smart meters will provide relevant
monitoring functionalities, such as real-time local voltage and load data that will be of high importance in
distribution network management processes and systems.
Function Medium Voltage Low Voltage
Monitoring Current, voltage, fault passage indicators
& other sensors
Centralization of information via secondary
substations
Control Remotely controllable reclosers, switches
automated protection / fault
sectionalisation – remote monitored
fault detectors
Voltage or power factor set points to DGs
Controllable MV/LV transformers26
(with
centralised or decentralised sensors)
Table 4 Technology options for distribution network development
3. Implications for Regulation and Market Design
The transition towards more active distribution networks requires the development of technology as well
as requirements for both network operators and network users to contribute to system security.
Appropriate commercial and regulatory frameworks need to be put in place.
As mentioned above, the new Energy Efficiency Directive provides a good baseline for setting up system
services at distribution level that will allow for an energy efficient use of infrastructure. Its appropriate
implementation is thus crucial.
In addition, regulation and a flexibility market model should be further developed to address the following
issues:
 Principles and methods for defining system states within the so-called ‘traffic lights concept’:
boundaries between the green, yellow (procurement of flexibility from markets) and red zones
(emergency state with non-commercial services) described in chapter 2.3.1 should be defined
depending on physical operating boundaries of a system, including methodologies of how they will be
monitored, calculated and audited.
 Key performance indicators and criteria for selecting most cost-effective solutions for different
situations: Inefficient investment deferral on one hand and oversized networks with long lead times for
connections on the other hand should be avoided. Flexibility operators such as aggregators should not
be incentivised to cause constraints that are not caused by reaction to market signals and subsequently
remunerated for solving them. It is key that a system approach taking into account benefits to the
society is adopted. Already existing methodologies should be used as a starting point.
26
Werther, Becker, Schmiesing, Wehrmann: Voltage control in low voltage systems with controlled low voltage
transformer (CLVT); CIRED Workshop - Lisbon 29-30 May 2012.

29
 Regulations for steering the most cost-efficient solutions: Regulatory mechanisms for steering when
the investment deferral is more cost-effective than active distribution approach should be elaborated,
for example in order to ensure that network development occurs when excursions to the yellow and
red operating states become too frequent (what is ‘frequent’ is also to be defined).27
Active system
management will affect the amount and structure of operational expenditure and would replace some
CAPEX with OPEX.28
DSOs should be able to look at the business case for both the investment solution
(CAPEX) and the service-based solution (OPEX), or a mixture of the two, and decide which is preferable.
Adoption of a regulatory mechanism is necessary to integrate this approach in the grid fee calculation.
DSOs need to be provided with adequate remuneration for the most adequate solution: investment or
active system management tools including procurement of flexibility services from network users.
 DG raises the need to review the suitability of the existing distribution network charging structures
(current network tariffs are largely volumetric). EURELECTRIC will present its proposals on this issue
later this year.
 New roles and relationships between different actors in the market: In particular, the role of flexibility
providers (to be taken up by aggregators/suppliers/ESCOs) and the relationship between these
flexibility providers and network operators, suppliers, the BRP (balance responsible party) and the local
customer/local producer should be addressed.
 Interaction between the DSO and the market: How will the DSO have visibility of what is happening on
the market side that may impact on the DSO network and cause constraints in the short or long term?
For example, a flexibility provider may have a contract to provide reserves to the TSO, but needs to use
the distribution system to which its resources are connected to deliver this service. However, a
bottleneck on the distribution system may prevent this delivery. These technical aspects of flexibility
markets require further investigation.
 Measurement/determination of “non-produced” and “non-consumed” energy is an important
related issue: unlike consumed and produced energy, saying how much a generator has “not
produced” when he reduced his production on demand of e.g. an aggregator is rather difficult, and
appropriate methodologies have to be developed and implemented. Knowledge of the initial schedule
is important in this respect.
 Revision of planning guidelines for distribution network development: Revision of these rules should
be reconsidered in order to account for ancillary services that the DSO can procure. Are deterministic
security standards such as the N-1 criterion appropriate or should they move to being more
probabilistic and based on reasonable risk of expected energy not served in order to maintain high
reliability while minimising over-investment in redundant capacity?
 Establishment of contingency assessment and outage management (organisation and coordination of
outages) rules for distribution networks and their users.
27
The intervention frequency on the customer side is registered – this parameter can be used as an indicator for the
necessary grid reinforcement or extension.
28
See EURELECTRIC report Regulation for Smart Grids, February 2011.

30
Key Conclusions
1. Secure and cost-effective integration of DER requires a rethink of how distribution grids are
planned and operated
Distribution networks are currently coping with the rapid increases in decentralised RES feed-in that Europe
is experiencing. Decentralised feed-in of RES has already started to outstrip local demand in some European
regions. This growth can be quite rapid, as the increased feed-in often occurs in rural areas with little
demand. Building sources of production close to consumption does not reduce distribution network cost.
The network still has to be designed to supply maximum demand for situations when there is no DG
production. In addition, operation of such a system becomes more complex. Once the share of
decentralised RES passes a certain point, it overburdens the local distribution grid. DSOs are therefore
increasingly facing voltage problems and grid congestions.
2. DSOs need tools that allow them to become real system operators
DSOs are responsible for developing their grids efficiently and providing quality of service for end
customers. But in order to satisfy these responsibilities in the changing context of DER/DG, DSOs need
adequate new tools. System services at the distribution level are key in this respect. Such services include
the participation of decentralised generators in voltage and reactive power management, distribution
network capacity management and congestion management, and information exchange between TSOs,
DSOs and DER.
Distribution network capacity management involves taking services from DER into account in the planning
stage to optimise investments and ensure that infrastructure is only extended when it is more cost-efficient
than procuring services from DER. To support congestion management, a ‘traffic light’ concept could be
introduced to denote different system states.
3. Optimising DER management through flexibility services should be explored
Traditional methods of grid expansion reinforce the network so that it may bear maximum feed-in from
distributed generation. However, such maximum capacity is only required for a short period of time each
year. While this approach will remain important, it may therefore not always be the most optimal and cost-
effective solution – not least because new lines often face problems of public acceptance.
New solutions must be developed to enhance the hosting capacity of the distribution grid, minimising
situations when DG feed-in has to be reduced or new connections have to be denied. Such solutions
include new network technologies and design concepts, the contribution of generation to system
performance, and access to flexibility services provided by DG operators, storage and demand response.
Their flexibility could be procured on a competitive basis via flexibility platforms and be offered to DSOs,
but also to TSOs as a way of managing redispatching issues in transmission grids. They should be used
whenever it is most efficient.
4. One-size-does-not-fit-all: different distribution networks require different solutions
While best practice sharing is desirable, the most economically efficient solution will depend on the
topology, demand and generation profiles within a given distribution system. The route to the future
energy system will probably incorporate aspects of each of the abovementioned solutions. The regulatory
framework should grant DSOs enough flexibility to determine the most appropriate solution for their
network. Methods of prioritising and selecting the most cost-effective solutions should be further
investigated.

2013 02-12 asm full report

More Related Content

2013 02-12 asm full report