Cired 2019 - 1482
Cired 2019 - 1482
Cired 2019 - 1482
Paper n° 1482
Paper n° 1482
times
• Loss of DSO revenue because of reactive Hardware
maintenance policies after faults and breakdowns
The SEED device is mounted on a din rail in the circuit
• Talent draining: Having a professional technician breaker marshalling box in the 33kV substation. As shown
conduct manual analysis of outage identification in the diagram below, the feeder voltage is monitored by
takes time away from other skilled activities such reading the 110V AC secondary output of substation
as providing active technical service feeder Voltage Transformer. 110V AC of all three phases
are given to the voltage measuring channels of SEED.
• Low accuracy: Non-availability of real time and According to the PT ratio, the line voltage is calculated by
accurate localised outage area, results in healthy measuring this PT voltage. On the secondary of line CT,
network areas also suffering from outage until high precision clamp type instrument CTs or Rogowski
fault location is identified and cleared coils are used to measure the line current. Three CTs are
• The substation transformers are typically not used for three phases. The current of this instrument CT is
monitored in traditional SCADA MDAS systems measured and based on the instrument CT ratio and line
and hence it is difficult to understand transformer CT ratio the actual current flowing through the line is
losses, hot spot temperatures and probability of measured.
failure of transformers
• The high cost and set up time of SCADA and
MDAS systems lead to a low penetration of such
systems and hence most substations still lack real
time monitoring
Paper n° 1482
The image below captures the key protocols used for Machine Learning algorithms are run on datasets which
establishing secure and reliable communications. are created by the time series database, focusing strongly
on the importance of time stamping and time synchronised
measurements to enable highly accurate auditing and
analytical data.
Data Visualisation
Finally, the real time data and analytics driven insights is
presented to the DSO on a web and mobile enabled front
end (www.orxa.io) that enables users to fully exploit the
capabilities of the hardware and GAP. Dashboards have
been developed to be intuitive and simplicity driven for a
clean, minimal user interface.
SYSTEM BENEFITS
Paper n° 1482
more detailed substation-level data and single line performing feeders and assets. Instant alerts are configured
diagrams of their feeders and transformers, while Division for different parameters set by the DSO.
Officers view inter-substation energy flow within the
entire divisional distribution area. SAIDI SAIFI Reliability Calculations
Based on the above interruption classifications, the IOT
Presenting personnel with curated dashboards enabled system also monitors the DSO’s network performance in
users to identify meaningful insights quickly and real time and updates the dashboards of the management
effectively. It was realised that daily peak loading on all officials. All unplanned and sustained interruptions are
transformers within the division was a key parameter to be considered for the SAIDI and SAIFI calculations which
monitored and displayed on dashboard. Real time loading indicate the healthiness of the system. The Reliability
of transformers and feeders is done using visual colouring indices are benchmarked against previous month and
within the front end. previous year performance and future indices are
predicted.
Current efforts are now to create Network Diagrams to
map extra high voltage incoming sources into the
divisional distribution network for even further visibility Asset Health Monitoring
into the electrical energy flow in real time. Not only does the IoT system provide real time monitoring
of electricity but it also provides real time insights into the
health of critical assets within the network, specifically
Energy Accounting substation power transformers and substation DC battery
The system is used for energy accounting through the supply. OTI and WTI signals from offline transformer
substation to identify probable locations of higher than dials are monitored by the system. When dials are not
acceptable technical losses. Energy accounting is done on present or not in working condition, then non-intrusive
hourly data of incoming and outgoing energy within the temperature sensors are used on transformer body
substation as a whole to find net substation loss. Similar combined with algorithms that calculate the oil
calculations are also being done at individual transformer temperature and winding (hot spot) temperature within the
level to identify transformer efficiency. OrxaGrid is also transformer.
implementing an intelligent energy accounting system to
identify separate electrical circuits based on circuit breaker The transformer temperatures are critical parameters to
positioning to calculate net losses within each circuit rather understand how much loading the transformer can sustain
than computing a single figure for the entire substation. All without problems. Real time loading on the transformer
above information is made available in the form of has also provided insights into the loading patterns of each
downloadable reports and configurable real time transformer. It was observed that many transformers were
dashboard widgets. Locations and assets with higher than unevenly loaded throughout the day resulting in one phase
acceptable technical losses will then be thoroughly tested energy delivery to be much higher than the other two
through technologies such as thermal imaging and hot spot phases. Upon consultation with the DSO, it was observed
detections. that load shedding policies to mitigate shortage of power
supply dictated single phase supply for eight hours of the
day on many of the feeders connected to the transformer.
Outage Classifications This caused the uneven loading on the transformer through
Due to the shortage of electricity availability in the entire the day. Ideas were brainstormed to balance out the single
country, distribution networks have pre planned load phasing on the transformer to switch from one phase to
shedding schedules for daily and monthly basis. Feeders another after certain intervals of time and thus reduce
are categorised as per type of connected consumers (rural transformer losses.
domestic, agricultural, industrial etc) along with setting
priority to feeders within the categories. One purpose of Realtime loading of transformers and feeders is shown
these classification is to provide guidelines during planned using visual colouring on the orxa.io frontend while alarms
and unplanned interruptions. and trips on power transformers is provided to concerned
DSO personnel through email or SMS notifications. Other
OrxaGrid configured its IoT system to collect real time key alerts that are provided are transformer oil and winding
data of the interruptions and also configured the system temperature rising beyond threshold limits, or when OLTC
with the DSO’s load shedding policies. The system is tap position is required to be changed due to unstable
designed to flag interruptions or outages outside of the nature of the connected load.
planned load shedding schedule. All interruptions are
classified as planned or unplanned, along with interruption For health check of substation DC battery supply, DC
type such as fault due to overcurrent or earth fault, or voltage and current monitoring sensors are installed within
manual tripping for single phasing loads. A monthly report the substation and real time data is reported to the Grid
is generated at the end of every month that provides a Analytics Platform. It was observed that certain substation
detailed summary of all interruptions out of the scheduled battery chargers were unreliable. Since the system
outages along with key insights such as identifying poor monitoring began, OrxaGrid has collected numerous
Paper n° 1482
CONCLUSION
OrxaGrid’s IoT based solution provides DSOs with a
viable alternative for not only monitoring their entire
substations but also finding key efficiency improvement
insights. The system is completely scalable by adding
more devices to monitor other substations in other
divisions of the DSO. The system is also interoperable
where existing data sources from MDAS or SCADA can
be fed into the system for deriving insights.
REFERENCES