The Roles of Edge Computing in Monitoring, Control, Management and Digitalization of Distribution Grids

By Omid Alizadeh-Mousavi, Joël Jaton

The design and operation of electrical grids have been receiving particular attention for using smart grid technologies. The transmission grids are critical infrastructure that have been designed and operated by deploying monitoring, control, and automation technologies ensuring their continuous, stable, and secure operation. The distribution grids, specifically in the low and medium voltage levels, have been traditionally operated almost without any monitoring devices, and the over-sizing of the grid infrastructure and the unidirectional power flows have allowed electric utilities to operate their grids in a safe manner.

The recent and emerging trends in the distribution grids utilization, as indicated below, have posed the need for new technologies for design and operation of the distribution grids.

• Trend 1: the growth of renewable energies, distributed energy resources, and prosumers are putting the low and medium voltage grids under heavy strain
• Trend 2: electricity regulators and consumers are becoming more and more stringent toward the electric utilities with regards to the electric power quality
• Trend 3: electric vehicle stock set to increase by 20-35 times by 2025, requiring the electric utilities to deliver more distributed and stochastic electricity from the grid to supply the e-mobility electricity consumption
• Trend 4: access to the flexibilities of distributed resources in the low and medium voltage grids for providing ancillary services and energy management services while respecting the impacts of these services on the distribution grid operational limits

An important challenge to implement the distribution grids conforming the abovementioned trends is the lack of monitoring and visibility in the low and medium voltage grids. The lack of monitoring and observability diminishes the ability to control and manage the distribution grids and distributed resources while complying with regulatory security requirements.

Nevertheless, these challenges can be turned into an opportunity by introducing smart grid technologies with intelligence at the edge of the gird. An edge computing solution for distribution grids is a decentralized technology aggregating measurement, analysis, and action functions, three essential features for efficient integration of distributed energy resources at the edge of the distribution grid. Such technology provides full digitalization of distribution grids and makes the grid observable and stable and also enables provision of flexibility services from distributed resources. The edge computation is a key feature for smart grid technologies since distribution grids are geographically widely spread and many distributed energy resources are being located at the grid edges, where communication and big data can be serious concerns for a centralized computing solution. The more centralized computing layers, such as fog computing and cloud computing, can be used for the coordination between the neighboring geographical locations.

Allocating the intelligence at the low voltage nodes (i.e. the edge of the grid) offers a bottom-up monitoring approach capable of capturing all information and events primarily in the low voltage grids and then demonstrating the grid status at the higher voltage levels. This bottom-up monitoring approach reduces or even removes the expenditures and difficulties for the installation of medium-voltage measurement devices providing cost-effective visibility for the both of low and medium voltage grids.

This edge computing technology can be also used for the optimal control of various distributed energy resources to ensure the technically secure and stable grid operation. The control algorithms are needed to be decentralized to avoid the issue of single point of failure and the control actions should be coordinated along the feeders of the higher voltage levels. For this purpose, the edge computing should be able to communicate with various controllable resources through different communication methods and protocols. It enables the provision of controllability and flexibility from distributed resources while considering the impact of these services on the grid operational limits.

Moreover, the edge computing solution requires, on one side, to make use of the available smart metering data for grid analysis, and on the other side, to interact with the existing SCADA/ADMS systems deployed in higher voltage levels.

From the grid integration point of view, the edge computing technology is easily installed as a retrofit solution for the existing infrastructure in distribution grids while avoiding numerous system integrator and the legacy systems and decreasing the system integration complexity. The requirements for electromagnetic compatibility and safety standards are also respected. The measurement sensors with required accuracies and time synchronization are integrated. The security of communication against possible cyber-attacks is guaranteed. Note that the edge computing solution should have sufficient decentralized processing power for performing various monitoring, control, and communication tasks.

Having access to the measurement data, the edge computing solution is thoroughly and perfectly positioned to provide further applications such as power quality analysis, fault identification, and assets failure prediction as well as to support data-driven and automated applications such as asset management and network planning. The data-driven approach is an important particularity of the suitable edge computing solution for distribution grids, specifically low voltage grids, in which accurate and up-to-date grid model is not always available and grid’s topology is changing frequently. In this approach, the measurement data is used to understand the important behavior and characteristics of the grid. It significantly decreases the required manpower for maintaining an up-to-date grid model.

The data-driven and automated applications provided by the described edge computing answer various needs of electric utilities in different time horizons, from real-time monitoring and control to steady-state analysis and management, as well as long-term planning and investment decision making.

Such edge intelligent solution facilitates the implementation of microgrids and energy communities.

The edge computing solution should be a combined hardware and software to provide a unified and comprehensive solution for monitoring, control, optimization, automation, management, and digitalization of distribution grids.