1.1 Introduction To The Project
1.1 Introduction To The Project
1.1 Introduction To The Project
INTRODUCTION
For scheduling load management, an effective optimisation method has been developed. It is
based on an analytic model of the load under control which gives it the advantage of
allowing any length for the control periods and any cycle rates. The method can be used to
minimise different objectives including peak load and production cost. The method is
possible in case of two-way connection or by the virtue of Internet of things which help in
direct control of the appliances in all three microgrids i.e. residential, commercial and
industrial
The introduction of DSM can bring following advantages to the electricity market.
It can promote efficient operation of the market and effectively restrain market
power.
It can realize instant information exchange about supply and demand, produce more
reasonable and transparent transactions, and speed up and improve the formation of
an electricity price mechanism.
It can effectively relieve demand congestion during peak hours and improve the
reliability of the power system.
It can effectively alleviate investment pressure on power generation, transmission,
and distribution.
1.1.2 Microgrid
A Micro-grid refers to a load and DGs connected together in an area that can be operated in
a controlled, coordinated way. They can be connected to the main power grid, operate in
“islanded” mode or be completely off-grid. The primary purpose of the microgrid is to
ensure local, reliable, and affordable energy security for urban and rural communities, while
also providing solutions for commercial, industrial, and residential consumers. Benefits that
extend to utilities and the community at large include lowering greenhouse gas (GHG)
emissions and lowering stress on the transmission and distribution system.
In many respects, microgrids are smaller versions of the traditional power grid. Similar to
the present electrical grids, they consist of power generation, distribution and controls such
as voltage regulation and switchgears. However, microgrids differ from traditional electrical
grids by providing a closer proximity between power generation and power use, resulting in
efficiency increases and transmission reductions. Microgrids also integrate with renewable
energy sources such as solar, wind power, small hydro, geothermal, waste-to-energy and
combined heat and power (CHP) systems.
With the increased number of blackout and load on the conventional grid the need for
microgrids emerged and with the introduction of microgrid there comes a problem of best
utilisation of the electric energy in a microgrid to minimise the cost of bills paid by the
consumers and moreover unbundling of electric utility services has become a major interest
in the industry.
The objective of the project is to minimise the cost of the bill by reducing the peak load in
the microgrids by moving and assigning the time changes to it based on the proposed
architecture of Demand Side Management which presents a day ahead DSM strategy for the
future microgrids ( using load shifting technique controlled by the central controller ).
The vision of the future power system is represented by ‘Smart Grid’ integrating advanced
sensing technologies, control methodologies and communication technologies at
transmission and distribution levels in order to supply electricity in a smart and user friendly
way. According to the U.S. Department of Energy’s modern grid initiative report, the main
characteristics [1] of a smart grid are consumer friendliness, hack proof self-healing,
resistance for attack, ability to accommodate all types of generation and storage options,
electricity market based efficient operation, high power quality, and optimal assets. This
modern grid is prompted by several economic, political, environmental, social, and technical
factors. Many utilities have the capability to control the loads of water heaters, air
conditioners and other devices in their service area. For the control of residential appliances,
this has been accomplished by installing remote control switches at the appliances to be
controlled. Radio, power line carrier and telephone communications have been used to
control the switches.
Utilities have different objectives for load control including objectives related to operating
cost, system security and load research. The 2 common objectives of load control are:
Minimisation of Production Cost
Minimisation of Peak Load
The minimization of peak load may be desired, by some utilities, to increase system security
by improving operating reserve margins. Other utilities buy most or all of their power with
the cost of energy dependent on their contribution to their supplier’s peak load. For these
utilities, the reduction of the supplier’s peak, using their load control capability, can result in
large savings. For utilities that generate most or all of their energy, reduction of production
cost through load management may result in significant cost savings. In the era of load
growth and increasing constraints on new and existing generation capacity, Demand Side
Management (DSM) options are being considered all over the world as possible bridges
between these two apparently conflicting requirements. The high variability of load from
one day to another, and from one hour to the next, may provide significant opportunities for
demand side management. DSM provides a workable solution to some of the major
problems confronting the electric utility today. There is a great deal of uncertainty in future
demand, fuel prices, construction cost, availability and cost of power from other utilities,
independent power producers, and the regulatory environment. Controlling and influencing
energy demand can reduce the overall peak load demand, reshape the demand profile, and
increase the grid sustainability by reducing the overall cost and carbon emission levels.
Efficient demand side management can potentially avoid the construction of an under-
utilised electrical infrastructure in terms of generation capacity, transmission lines and
distribution networks in a microgrid.
DSM originated from the energy crises. The first energy crisis (also called the “first oil
shock”) happened in October 1973. During the 1973 Arab-Israeli War, to beat Israel and its
allies, the Organization of Arab Petroleum Exporting Countries (OAPEC) announced an oil
embargo and export suspension, causing a rise in oil prices. Crude oil prices increased by
almost four times from $3 per barrel to nearly $12 per barrel, which caused recessions in
Western developed countries. This situation brought energy management into the public
consciousness. In response to that, the US Congress legislated the National Energy Act of
1978. As part of it, the National Energy Conservation Policy Act and the Power Plant and
the Industrial Fuel Use Act ware enacted, which took DSM into consideration.
The second energy crisis in 1979 and the third energy crisis in 1990 sped up the
development of DSM. The outbreak of the Iranian revolution and the Iran-Iraq War caused a
sharp drop in crude oil production. The crude oil price increased dramatically from about
$15 per barrel in 1979 to $39 per barrel in 1981. Then the Gulf War in 1990 stimulated the
international market. To deal with this, the Energy Policy Act of 1992 was passed. It
addressed the importance of energy efficiency, energy conservation, and energy
management, and also prompted the use of renewable energy.
DSM became well known to the public in the 1980s, popularized by the Electric Power
Research Institute.
“Demand response” mainly refers to the actions taken on the customer side that use the
market price to influence the level and time of electricity demand.
In general, the introduction of demand response into the power market requires a
precondition: the electricity market must achieve tentative liberalization or full
liberalization, which means some kind of real-time market prices and effective market
mechanism mist exist in the electricity market. Meanwhile, demand response will accelerate
the formation of the real-time market pricing mechanism. And with the high penetration of
demand response into the market, it can provide economic incentives to promote other
projects such as energy efficiency and energy storage in DSM. But DSM does not need this
mechanism. Even without it, DSM can realize some of its projects. At the same time, DSM
can fully boost and amplify the economic effectiveness of demand response.
The term “negawatt power” is a derivative term from DSM and demand response. It is
coined for the way of supplying additional electrical energy to consumers without adding
generation capacity. Even though it is a theoretical unit for electrical energy, it can still be
traded as a commodity in the electricity market
Typically, demand response can provide five services to the system: (1) peak clipping, (2)
valley filling, (3) load shifting, (4) strategic conservation, and (5) strategic load growth. The
first three can be grouped as load management, and the last two can be grouped as load-
shape change. Load management is normally related to deliberate behaviours enforced by
utilities. In contrast, the load-shape change can be both natural behaviours of customers and
deliberate behaviours enforced by utilities
Peak Clipping
When the demand approaches the threshold of the supply capacity or the transmission
system approaches the threshold of the thermal requirements, this peak load demand must be
reduced. This can be realized by the direct load control in the residential sector (e.g., turning
down the thermostat of heaters and increasing the temperature of refrigerators). This can
also be achieved by interruption in the industrial and commercial sectors. Fig 2.1 shows a
peak reduction from 12MW to 10MW during the period from 18:00 to 20:00. This service
can help to release the stress on the system during the peak period. However, because it
curtails the consumption of certain loads, it can cause customer dissatisfaction.
Valley Filling
When the demand is manifestly low at off-peak time, which is also not favorable for system
stability, the demand should be increased. The commonest method is to add storage devices
(e.g., thermal storage for heaters and plug-in electrical vehicles). Fig2.2 shows a valley
filling from 4 MW to 6 MW during the period from 2:00 to 6:00. This service increases the
total power consumption of customers but may not necessarily increase the bill.
Load Shifting
When the load is apparently higher than the average level in a certain period, a certain
amount of load must be moved from that period to other periods. This primarily relies on the
deferrable appliances, which can justify the time of use (e.g., washing machines). In the
short term, load shifting can be achieved on a daily basis from peak time to off-peak time. In
the long term, load shifting can be achieved on a seasonal basis. Fig 2.3 shows a daily load
shifting in which part of the peak demand is shifted from 18:00–20:00 to 2:00–6:00. It does
not reduce the total consumption but only changes the time of use. Therefore, this service
does not cause customers substantial inconvenience.
Strategic Conservation
When the overall load exceeds the supply level, customers are encouraged to reduce their
overall consumption. One basic method is to improve energy efficiency. This can be done
on a small scale by the replacement of traditional devices with energy-efficient devices (e.g.,
changing filament lamps to fluorescent lamps). It also can be done on a large scale (e.g.,
weatherization program). Besides the technical improvements, the informa- tion support is
also important. In general, providing consumption and cost details to customers can
facilitate power reduction. Fig 2.4 shows strategic conservation from a high-power level to a
low level.
When the demand falls below the normal level of supply, customers are encouraged to
increase their overall consumption. The electrification technology has the potential to
achieve this (e.g., the popularization of electrical vehicles). Fig2.5 shows strategic load
growth from a low power level to a high level.
2.1.3 Microgrid
Microgrid is a localised group of electricity generation sources and loads that normally
operates connected to and synchronous with the traditional centralised electrical
grid(Macrogrid), but can also disconnect to "island mode" — and function autonomously as
physical and/or economic conditions dictate.
In this way, a microgrid can effectively integrate various sources of Distributed generation
(DGs), especially Renewable Energy Sources (RES), and can supply emergency power,
changing between island and connected modes. Control and protection are major challenges
to microgrids.
The U.S. Department of Energy Microgrid Exchange Group defines a microgrid as a group
of interconnected loads and distributed energy resources (DERs) within clearly defined
electrical boundaries that acts as a single controllable entity with respect to the grid. A
microgrid can connect and disconnect from the grid to enable it to operate in both connected
or island-mode.
The EU research project describes a microgrid as comprising Low-Voltage (LV) distribution
systems with distributed energy resources (DERs) (microturbines, fuel cells, PhotoVoltaics
(PV), etc.), storage devices (flywheels, batteries) energy storage system and flexible loads.
Such systems can operate either connected or disconnected from the main grid.
These microgrids never connect to the Macrogrid or main grid and instead operate in
an island mode at all times because of economic issue or geographical position.
Typically, an "off-grid" microgrid is built in areas that are far distant from any
transmission and distribution infrastructure and, therefore, have no connection to the
utility grid.
These microgrids are being actively deployed with focus on both physical and cyber
security for military facilities in order to assure reliable power without relying on the
Macrogrid.
These types of microgrids are maturing quickly in North America and Asia Pacific;
however, the lack of well –known standards for these types of microgrids limits them
globally. Main reasons for the installation of an industrial microgrid are power
supply security and its reliability. There are many manufacturing processes in which
an interruption of the power supply may cause high revenue losses and long start-up
time.
2.1.3.2 Basic Components in a microgrid:
a) Local Generation:
It presents various types of generation source that feed electricity to user. These
sources are divided into two major groups – thermal energy sources (ex. Diesel
generators) and renewable generation sources (e.g. wind turbines, solar).
b) Consumption:
It simply refers to elements that consume electricity which range from single devices
to lighting, heating system of buildings, commercial centers, etc. In the case
of controllable loads, the electricity consumption can be modified in demand of the
network.
c) Energy Storage:
Advantages
A microgrid is capable of operating in grid-connected and stand-alone modes and of
handling the transition between the two. In the grid-connected mode, ancillary services can
be provided by trading activity between the microgrid and the main grid. In the islanded
mode, the real and reactive power generated within the microgrid, including that provided
by the energy storage system, should be in balance with the demand of local loads.
A microgrid may transition between these two modes because of scheduled maintenance,
degraded power quality or a shortage in the host grid, faults in the local grid, or for
economic reasons. By means of modifying energy flow through microgrid components,
microgrids facilitate the integration of renewable energy generation such as photovoltaic,
wind and fuel cell generations without requiring re-design of the national distribution
system. Modern optimization methods can also be incorporated into the microgrid energy
management system to improve efficiency, economics, and resiliency.
Challenges
Microgrids, and integration of DER units in general, introduce a number of operational
challenges that need to be addressed in the design of control and protection systems in order
to ensure that the present levels of reliability are not significantly affected and the potential
benefits of Distributed Generation (DG) units are fully harnessed. Some of these challenges
arise from invalid assumptions typically applied to conventional distribution systems, while
others are the result of stability issues formerly observed only at a transmission system
level.[13] The most relevant challenges in microgrid protection and control include:
In regards to the architecture of microgrid control, or any control problem, there are two
different approaches that can be identified: centralised and decentralized. A fully centralized
control relies on a large amount of information transmittance between involving units and
then the decision is made at a single point. Hence, it will present a big problem in
implementation since interconnected power systems usually cover extended geographic
locations and involves an enormous number of units. The fully centralized control is
currently considered infeasible. On the other hand, in a fully decentralized control, each unit
is controlled by its local controller without knowing the situation of others. The fully
decentralized control is also irrelevant in this context due to strong coupling between the
operations of various units in the system. A compromise between those two extreme control
schemes can be achieved by means of a hierarchical control scheme consisting of three
control levels: primary, secondary, and tertiary.
a) Primary Control:
b) Secondary Control:
Secondary control has typically seconds to minutes sampling time (i.e. slower
than the previous one) which justifies the decoupled dynamics of the primary and
the secondary control loops and facilitates their individual designs. Set point of
primary control is given by secondary control in which as a centralized
controller, it restores the microgrid voltage and frequency and compensates for
the deviations caused by the primary control. The secondary control can also be
designed to satisfy the power quality requirements, e.g., voltage balancing at
critical buses.
c) Tertiary Control:
Tertiary control is the last (and the slowest) control level which consider
economical concerns in the optimal operation of the microgrid (sampling time is
from minutes to hours), and manages the power flow between microgrid and
main grid. This level often involves the prediction of weather, grid tariff, and
loads in the next hours or day to design a generator dispatch plan that achieves
economic savings. In case of emergency like blackouts, Tertiary control could be
utilized to manage a group of interconnected microgrids to form what is called
"microgrid clustering" that could act as a virtual power plant and keep supplying
at least the critical loads. During this situation the central controller should select
one of the microgrid to be the slack (i.e. master) and the rest as PV and load
buses according to a predefined algorithm and the existing conditions of the
system (i.e. Demand and generation), in this case, the control should be real time
or at least high sampling rate.
1. industrial sector
2. residential sector
3. commercial sector
4. transportation sector
As for demand response, the industrial, residential, and commercial sectors are mainly
concerned.
The usage patterns in the residential sector are more complicated than in the other sectors.
Firstly, the number of customers is much higher. The distribution of customers is wide and
scattered. Secondly, the types of appliances used by customers are diverse. Even for the
same type of appliances, power consumption of different brands can differ. Thirdly, every
customer has his or her own personal usage preference. That means each customer needs to
be treated specifically rather than identically.
1. Long-range customers: their elasticity of electricity use is relatively high.
They are
able to modify their use over a wide time range.
2. Real-world postponing customers: they consider the current and future
electricity
prices, and give certain responses to utilities,
3. Real-world advancing customers: they focus on the past and future
electricity prices,
and also give certain responses to utilities,
4. Real-world mixed customers: they are a combination of both postpon-
ing customers
and advancing customers.
5. Short-range customers: they pay attention only to the current electricity
price. They
are not willing to change their consumption pattern.
On the basis of the operational characteristics of appliances, the loads can be classified by
two standards:
whether the occupied time duration of appliances can be modified or not;
whether the total electricity consumption of appliances can be modified or not.
For the first standard, loads can be divided into deferrable loads and non-deferrable loads.
For the second standard, loads can be divided into adjustable loads and nonadjustable loads
The activation of deferrable loads can be stopped, restarted, or shifted to other time slots
(e.g., washing machine and electrical vehicles). Generally, most of the wet loads belong to
the deferrable loads. These loads can be scheduled by a demand response program. On the
basis of the electricity price or the monetary incentive, they can be shifted from peak hour to
off-peak hours, therefore reducing the peak load demand. In contrast, the nondeferrable
loads need to finish the schedule in a specified time, (e.g., lighting systems and kitchen
systems). These loads do not allow a time shift and interruption. As such, these loads are not
suitable for the demand response program.
For the adjustable loads, the consumption can be adjusted to a lower level (e.g., in winter,
heaters can be set at 23◦C rather than 25◦C). Normally, most of the thermal loads are part of
the adjustable loads. These loads can be involved in the demand response program. The total
consumption can be reduced on the basis of the electricity price or the monetary incentive.
However, reducing the consumption can affect customers’ comfort as described by the
quality of experience (QoE). In contrast, for the nonadjustable loads, the total consumption
is fixed (e.g., TVs and computers). Same like nondeferrable loads, nonadjustable loads
cannot be scheduled by a demand response program.
In these methods, incentives are offered to customers depending on their behavior in the
demand response programs. Normally, customers change their consumption voluntarily.
Direct Load Control:
According to the advanced agreement between customers and utilities, utilities can
remotely control some customers’ appliances (e.g., air conditioners and water
heaters). The notices for the operation are normally announced a short time ahead.
To participate in this method, customers need to be equipped with a remote control
switch system so that utilities can reschedule, turn on, or turn off the appliances.
Interruptible/curtailable service:
Compared with direct load con- trol, this method is normally applied to the industrial
sector and large- scale commercial sector. When the system is congested, customers
are asked to reduce some loads to a certain level. By participating in this, customers
can receive a rate discount or bill discount.
Demand Billing:
Instead of being asked by the utilities to take part in demand response programs,
customers can make decisions by themselves in this method. On the basis of the
generation and demand situation, utilities announce the total amount of electricity
that must be curtailed. Customers can bid for the amount on the basis of their own
situation and the wholesale market.
When the system is short of reserve, customers are required to reduce their
predefined consumption. The announcement is normally released one day ahead.
These curtailments are treated as system capacity to replace the conventional
generation and delivery resources. By proving their ability for curtailment, customers
can receive a reservation payment.
Similarly, to demand bidding, customers also bid for electricity curtailments. These
bids are offered to an inde- pendent system operator/regional transmission
organization. These curtailments are used as operational reservation. If the bid was
accepted, customers need to abide by a standby standard.
In these methods, electricity prices vary according to the cost of generation and the demand
for electricity. On the basis of these prices and other information, customers can decide on
their consumption. Generally, there are four types of pricing schemes:
a) Flat pricing
b) Time-of-use (ToU) pricing
c) Critical peak pricing
d) Real-time pricing
a) Flat Pricing:
This is the most traditional and widely used pricing scheme. The electricity price is
constant all the time. In this situation, the only way to reduce the bill is to reduce the
total consumption. The prices can be set seasonally.
b) ToU Pricing:
This is an improvement on flat pricing. The prices are different in different time
slots. Within each slot, a flat price is applied. Usually the prices are predefined for 1
day. In this scheme customers tend to shift their demand to a lower-price period. In
this way, the ability to reduce the total electricity demand is narrowed.
This scheme is derived from the ToU pricing scheme. The extreme peak demand
period is picked out. During this period, a much higher electricity price is
announced. This scheme can effectively bring down the peak demand. The critical
peak price can be set on the basis of the demand level or the time of the day. Three
types of pricing have been considered: fixed-period critical peak pricing, variable-
period critical peak pricing, and variable critical peak pricing.
d) Real-Time Pricing:
The electricity price fluctuates frequently, nor- mally by hours. The change of price
can indicate the relationship between supply and demand in the wholesale market. It
requires effective two-way communication between utilities and customers.
Sometimes, market aggregators also participate in this scheme to deal with data
collection and increase the efficiency. Customers are involved mostly in this scheme
and are notified of these prices in a day-ahead manner, hour-ahead manner or 15-
min-ahead manner. On the basis of the price and their own situation, customers can
decide on their consumption pattern.
2.1.7 Demand-Side Management Methods:
To effectively design DSM programs, many technologies have been used; for example, a
multiobjective optimization method and the linear matrix inequality (LMI) approach.
This method can take several objectives into consideration and obtain a fair decision that
does not favor any specific objective. It has been used in multiple microgrid system design
for a market operator and distribution network operator. Three participants are considered:
microgrids, a power grid, and an ISO. The relationship among these three participations is
shown in Fig. 7.14, in which three microgrids (N = 3) and two energy storage systems (Ns =
2) are involved. Table 7.2 presents our notation.
For the power grid the objective is to maximize the overall utility: the net gain for providing
power to microgrids. The utility function can be defined as
For microgrids the objective is to maximize the overall utility: the net revenue for
consuming power, which is provided by the power grid.
The LMI approach has been used for many situations. Because of the convex property, the
associated problem can be solved efficiently. It has been used to design a storage system in
smart grid networks.The basic idea is to charge the batteries when the price is lower than the
threshold and discharge the batteries when the price is higher than the threshold. Price
signals, system uncertainties, and physical constraints are three important factors that must
be considered in this design.
Fig. 7.18 shows an example of the system model. It consists of one conventional power grid
and five users. These users are assumed to be smart houses and have storage devices. All
users are connected to the power grid so that the power can flow between the grid and all
users. Some users are connected to others, and hence the power can also flow among users.
This assumption is based on two-way power flow and two-way communication.
Chapter:3
METHODOLOGY
This project presents a generalized day-ahead demand side management (DSM) strategy for
the future smart grid. It uses load shifting as the primary technique that can be utilized by
the central controller of the smart grid. Objective of the demand side management could be
maximizing the use of renewable energy resources, maximizing the economic benefit,
minimizing the power imported from the main distribution grid, or reducing the peak load
demand. Smart grid manager designs an objective load curve according to the objective of
the demand side management. The proposed optimization algorithm aims to bring the final
load curve as close to the objective load curve as possible such that the desired objective of
the DSM strategy is achieved. For example, if the objective of the demand side management
is to reduce the utility bill, an objective load curve will be chosen such that it is inversely
propositional to electricity market prices. Fig.2 shows the proposed architecture for the day-
ahead demand side management strategy. According to the proposed architecture, the
demand side management system receives the objective load curve as an input, and
calculates the required load control actions in order to fulfill the desired load consumption.
Therefore, the proposed algorithm is flexible in that it is completely independent from the
criteria used to generate the objective load curve. The demand side management is carried
out at the beginning of a predefined control period which is typically a day. Then, the
control actions are executed in real-time based on the results. Fig. 2 shows the exchange of
information between the demand side management controller and each appliance during the
real-time operation. This takes advantage of the communication capability of the smart grid.
When a customer presses ON button of an appliance, the connection request is sent to the
demand side management controller. The demand side management controller replies based
on the results of demand side management technique applied, that was carried out in
advance. The reply is either the connection permitted or a new connection time.
The proposed demand side management strategy schedules the connection moments of each
shift-able device in the system in a way that brings the load consumption curve as close as to
the objective load consumption curve. Proposed load shifting technique is mathematically
formulated as follows.
The increment in the load at time t due to the connection times of devices shifted to time ,
and the increment in the load at time t, due to the device connections scheduled for times
that precede t. The 𝐶𝑜𝑛𝑛𝑒𝑐𝑡(𝑡) is given by the following equation:
Where 𝑋𝑘𝑖𝑡 is the number of devices of type k that are shifted from time step i to t, D is the
number of device types, Pk and P(1+l)k are the power consumptions at time steps 1 and (1+l)
respectively for device type k , and j is the total duration of consumption for device of type k
. Fig. 3 illustrates 𝐶𝑜𝑛𝑛𝑒𝑐𝑡 (𝑡), where Load A is shifted from tA to (t-1) , and 𝐿𝑜𝑎𝑑 𝐵 is
shifted from tB to t . Similarly, 𝐷𝑖𝑠𝑐𝑜𝑛𝑛𝑒𝑐𝑡(𝑡) also consists of two parts: the decrement in
the load due to delay in connection times of devices that were originally supposed to begin
their consumption at time step t , and the decrement in the load due to delay in connection
times of devices that were expected to start their consumption at time steps that precede t.
The 𝐷𝑖𝑠𝑐𝑜𝑛𝑛𝑒𝑐𝑡(𝑡) is given by the following equation:
where the number of devices of type that are delayed from time step to , is the maximum
allowable delay.
Fig. 4 illustrates , where is shifted from to ,and is shifted from to . This minimization
problem is subject to the following constraints:
Thenumberofdevicesshiftedcannotbeanegativevalue.
The number of devices shifted away from a time step cannot be more than the number of
devices available for control at the time step.
Type equation here.
To demonstrate the effectiveness of the proposed approach, the proposed demand side
management strategy is tested on three different areas of a smart grid, each with different
types of customers; namely residential, commercial and industrial customers. Electrical
network diagram of the smart grid is shown in Fig. 6. The entire network operates at a
voltage of 410 V. Each interconnection link including the link between the smart grid and
the main grid has a resistance of 0.003 pu, a reactance of 0.01 pu, and maximum power
transfer limit of 500 kVA. Length of the links in the residential microgrid is 2 km, while
those in the commercial and industrial microgrids are 3 km and 5 km respectively. This
network information is used for scheduling of resources without any congestion.
Residential Area
The devices subjected to control in the residential area have small power consumption ratings and short
durations of operation. Table II shows device types that are subjected to load control and their consumption
patterns. There are over 2600 controllable devices available in this area from 14 different types of devices.
Commercial Area
The devices subjected to load control in the commercial area have consumption ratings which are slightly
higher than those in the residential area. The consumption patterns of the loads under the control are given in
Table III. There are over 800 controllable devices available for control in this area from 8 different types of
devices.
Industrial Area
The number of devices available for control in the industrial area is the smallest among all three areas;
however, the devices have largest consumption ratings and longest consumption periods. The consumption
patterns of the devices in this area are given in Table IV. The reason for a small number of devices available
for control can be attributed to the fact that most of the industrial loads are critical and cannot be subjected to
load control. The control periods of the devices are similar to those in the other two areas. There are over 100
controllable devices belonging to 6 different types.
4.1
Simulation results show that the proposed demand side management strategy has managed to bring the final
consumption close to the objective load curve in all three cases. The proposed algorithm has efficiently
handled the large number of controllable loads of several types, and adopts all heuristics in the smart grid. The
simulation results obtained for the residential area are given in Fig. 7. The utility bill of the residential area for
the day reduces from $2302.90 to $2188.30 with demand side management strategy, resulting in about 5.0%
reduction in the operating cost. The results obtained for the commercial area are given in Fig. 8. The utility bill
of the commercial area for the day reduces from $3636.60 to $3424.30 with demand side management strategy,
which results in approximately 5.8% reduction in the operating cost. Fig. 7. DSM results of the residential area.
Fig. 8. DSM results of the commercial area. Fig. 9. DSM results of the industrial area. The results obtained for
the industrial area are given in Fig. 9. The utility bill of the industrial area without demand side management
strategy is $5712.00 for the day; whereas it is $5141.60 with demand side management strategy, resulting in
10% reduction in the operating cost. Table V summarizes the simulation results from the proposed demand
side management strategy for these three areas of the smart grid. The approach has successfully managed to
achieve the objective in all three areas, with considerable savings in the utility bills. Typically, demand side
management results are better when the number of devices available for control increases [3]–[15]. However,
this may not be true in this case study because complexities of the devices under control pose restrictions. In
this case study, even though the number of devices available for control is the least in the industrial area, the
percentage reduction in the operating cost is the highest among all areas. On the other hand, the residential area
has the highest number of devices available for control in terms of quantity and variety, but the percentage
reduction in the operating cost is not much as expected. This can be attributed to the fact that high power
consumption of devices in the industrial area compared to much lower consumption of devices in the
residential area. Additionally, even small load shifting of high power devices results in huge savings for the
customers. In a smart grid, it is expected that utilities will bundle up services and provide several contract
options for customers. The amount of reimbursement taken by customers through such schemes will depend on
how much inconvenience the customer is willing to undergo. In case of customers in residential area, level of
tolerance is high for most of the devices. This means that generally the customers do not have strong
preference regarding the time when the loads have to be consumed. On the other hand, customers in
commercial and industrial areas are less willing to change their consumption patterns. Effective demand side
management provides benefits not only to the end users but also to the utilities. One of the main advantages is
the reduction in peak load demands. Table VI shows the peak load demands with and without the proposed
demand side management strategy for the three areas. It can be observed that the proposed demand side
management strategy reduces the peak load demand for each area. Reduction in the peak load demand
improves grid sustainability by reducing the overall cost and carbon emission levels. Furthermore, this will
lead to the avoidance of the construction of an under-utilized electrical infrastructure in terms of generation
capacity, transmission lines and distribution networks. Generation companies also stand to benefit from
demand side management, as the reduction in peak load demand results in substantial cost savings since costly
generators that are typically turned on to provide power during the peak load demand are no longer needed.
When system peak load demand reduces, the operating cost of generators will therefore be reduced
substantially. This would also result in increasing reserve generation capacity of the system. In order to
determine the cost saving for generation companies, the entire smart grid comprising of three areas is
scheduled for the day with and without proposed demand side management strategy. Problem formulation and
methodology of the generation scheduling are described in [21], [22]. Figs. 10–12 show the generation
scheduling of residential, commercial, and industrial areas without proposed demand side Fig. 10. Residential
area without DSM. Fig. 11. Commercial area without DSM. Fig. 12. Industrial area without DSM.
management strategy respectively. Fig. 13 shows power exchange among the areas. In this case, the industrial
area spends $695.51 with 346.34 kWh of load curtailment, and the commercial area spends $551.33, whereas
the residential area saves $225.48 for the scheduling day. As a whole, the smart grid spends only $53.70 with
346.34 kWh of load curtailment during the scheduling day.
The simulation results show that demand side management is indeed beneficial to both consumers and the
utility companies. In the test case, customers achieve 5% to 10% cost savings, and generation companies
archive a substantial saving with optimized generation scheduling. In addition, transmission companies achieve
between 15% to 20% reduction in the network congestion. The above test cases are only isolated examples.
The actual cost savings and peak reduction will be much higher when the proposed strategy is used for a large
smart grid. The above test cases are just examples to show how the proposed demand side management
strategy could be applied for smart grid operation. The proposed demanded side management strategy is a
generalized technique based on a day-ahead load shifting, and can be applied for large smart grids of the future.
The inputs to the problem are: control period (i.e., number of time steps), discrete model of devices’ load
consumption pattern, power consumption at each time step and various uncertainties. The proposed
evolutionary algorithm converges well, and it takes about 6 h for the test case comprising three areas. The
convergence characteristic for the case study is shown in Fig. 18. Although the computation time appears large,
it is acceptable as it solves a day-ahead demand side management strategy which is to be carried out 24 h
before the time of operation. Even though the proposed strategy provides good results, further work is needed
for real time implementation. A real-time demand side management strategy, together with this proposed day-
ahead demand side management strategy and demand side management techniques such as load curtailment,
load shedding and load shifting with small time span can be developed for real-time operation of smart gird.
The nature of real-time demand side management problem shows that it is possible to develop only on a
distributed operational platform.
Chapter 5
Conclusion
Demand side management has potential to provide many benefits to the entire smart grid, particularly at
distribution network level. This paper presents a demand side management strategy that can be employed in the
future smart grid. The proposed strategy is a generalized technique based on load shifting, which has been
mathematically formulated as a minimization problem. A heuristic based evolutionary algorithm is developed
for solving the problem. Simulations were carried out on a smart grid which contains three different kinds of
customers’ areas. The simulation outcomes show that the proposed algorithm is able to handle a large number
of controllable devices of several types, and achieves substantial savings while reducing the peak load demand
of the smart grid.