Decarbonized DSO: The Grid Observability Challenge

The #energytranstion effect on the future of the electricity distribution network, designed to incorporate intermittent generation sources and involving millions of advanced technology users, is going to introduce a significant level of uncertainty and variability into the system. The transition can affect the security of the electricity supply and must be carefully planned to avoid risks to the reliability and quality of #electricpower supply.

Distribution grid management is and will be challenging due to the uncertainty associated with decarbonized #powergrid. Continuous increase in massive data generation and flow, is also a rising issue, as the grid control paradigm becomes more distributed and adaptive. The new challenges mostly appear in controllability and observability of the distribution grid.

Modern distribution grid is a complex network of electrical lines, transformers, switchgear, energy resources, storages. loads and other equipment that requires monitoring and management to ensure balanced, reliable and efficient operation.

Distribution grid observability refers to the ability to monitor, measure and analyze the performance of the electrical distribution system in a predictive as well as real-time manner, with a high degree of accuracy and granularity. Distribution grid observability becomes even more critical in the presence of massive distributed energy resources (DERs) which can inject energy into the grid at various locations, making it difficult to predict and manage the flow of electricity.

Distribution grid observability refers to the ability to monitor, measure and analyze the performance of the electrical distribution system in a predictive as well as real-time manner, with a high degree of accuracy and granularity.

Advanced analytics and machine learning algorithms can process the vast amounts of data generated by these sensors to identify patterns and anomalies in the system's behavior. These tools can help operators to identify potential problems before they occur, such as voltage fluctuations, overloads, or outages.

Grid observability is the key to reliability, resilience, and operational excellence in the decarbonized distribution grids.

One of the key challenges for distribution grid observability is the sheer scale of the system. The distribution grid includes thousands of components spread out over a large geographic area, which makes it difficult to collect and manage data effectively.

Controllability and Observability: The Synergy

Controllability refers to the ability to control the operation of the grid, the ability to change the grid's behavior, while observability is the ability to measure and understand its behavior, the ability to observe its state.

Without adequate observability, it is difficult to control the grid effectively, as operators may not have the information they need to make informed decisions. Conversely, without adequate controllability, even the most advanced monitoring systems may not be able to prevent outages and other disruptions.

Distribution grid must be both controllable and observable in order to operate efficiently and reliably.

How Observable

Observability also can be defined as chronological, geospatial, and topological awareness of all grid status.

Observability for distribution grids is basically a complicated issue. Complicating factors include feeder branches and laterals, unbalanced circuits, poor documentation, enormous numbers devices feeders with inter-ties, time-varying circuit topology, etc.

Distribution grid characteristics are:

·      Operating in a time-varying unbalanced mode

·      Actual connectivity can be poorly estimated

·      Topology can change in between the time of a state estimate and the time that actions based on that estimate are taken.

For the decarbonized power distribution grid the level of observability and awareness should be high enough to cover the requirements of grid stability and reliability in presence of intermittent generation and load.

Observability in the High Voltage

The HV network in normally fully observable but it might be requires to expand the traditional SCADA system with modules capable of advanced calculations, providing more predictive and dynamic results for line loading , congestion management and development of optimal system configurations.

TSO-DSO collaboration and coordination is vital concerning mitigating the system instabilities, and optimizing the use of renewable resources.

Observability in the Medium Voltage

MV is part of the grid that is most susceptible to faults and failures due to its proximity to customers and the large number of components in the network.

Smart grid development can bring a great deal of observability to the medium voltage networks. A network of optimized number of sensors together with advanced data analytics and machine learning techniques is necessary to predict future failures, identify the root cause of problems, and optimize the operation of the network. This can help to improve the reliability and efficiency of the network, reduce downtime and outages, and ultimately provide better service to customers.

Observability at low voltage

LV grids present special problems in terms of topological state. Such state information is crucial because it is the context in which grid data, events and control commands must be interpreted.

In addition, distribution grid topology can be dynamic, such as cases where feeders are partially meshed or are tied to other feeders for reliability reasons.

AMI systems could be used to operate the LV networks as it allows the monitoring of measurements in the secondary substations. Various smart grid applications can make use of the AMI data either offline or close to real-time mode to assess the grid voltage conditions and estimate losses in the lines/cables.

Low voltage observability is essential in managing and optimizing the performance of the distribution grid. It is important because it allows utilities to detect and diagnose problems in the grid, such as voltage fluctuations or power quality issues, before they escalate and cause disruptions to customers. Moreover, by collecting data on the grid's operation, utilities can use advanced analytics and algorithms to identify opportunities for improving efficiency, reducing energy losses, and optimizing the placement and sizing of assets, such as transformers or capacitors.

Steps to more observable distribution grid

To achieve observability, the first step would be developing an observability strategy based on planning and operation targets. Then, classify data and measurement types and characteristics, determine the sensor mix and build a sensor location plan. The use of advanced machine learning and analytics technologies can also help in reducing the number of sensors required to reach to certain level of observability.

As there are no indicators defining the level of grid observability, especially at the LV, the strategy should take into account those elements determining the observability that will allow the achievement of the developed targets.

The use of advanced machine learning and analytics technologies can help in reducing the number of sensors required for reaching to a certain level of observability, and reducing the costs

The very next step would be the implementation of a set of features and elements in the grid management system that enables an appropriate level of observability:

• Grid Model and Digital Twins
• Smart Grids
• AMI integration
• ADMS SCADA and grid calculation outcome
• FLISR/OMS using the data collected through fault indicators

More on AI, analytics and machine learning

More on enhanced HV observability applications

More on OMS

Popular Posts

Decarbonization News

Events