Developing, Testing, and Implementing T&D Analysis Software for the Future Grid | SEPA Skip to content

Developing, Testing, and Implementing T&D Analysis Software for the Future Grid

I recently had the opportunity to moderate a session at IEEE’s Innovative Smart Grid Technologies (ISGT) conference in Washington, DC. I engaged with Dr. Robert Broadwater of Electrical Distribution Design (EDD) and project partners from Virginia Tech and Pepco at the conference to learn more about the team’s approach to an increasingly complex electric grid.

Project Background

Through a three-year, U.S. Department of Energy-funded project, EDD and its project partners have been building, testing, and implementing a model-based real-time analysis software architecture. This work represents a major step forward in grid monitoring and control. The project, which also includes utility partner Pepco Holdings, and technology/research providers Clean Power Research, Florida State University, SEPA, University of Delaware, Dominion Voltage Inc, and KITU Systems, is set to close at the end of October 2023. It will include published results from software lab and field testing through a customer pilot program. Additional details on the project are available in this SEPApower blog from October, 2020.

An Integrated System Model (ISM) is at the core of the software architecture. The ISM works to integrate previously disparate transmission, substation, primary distribution, and secondary distribution models into a unified framework. This model can break down silos between different utility organizations by offering utilities and other electric grid operators a single information source. The ISM software has the power to correlate data from grid measurement systems to detect abnormalities in real-time, uncover the source of abnormalities, and employ multi-mode, coordinated control to mitigate abnormalities (e.g., voltage instabilities or cyber attacks), increase energy savings, and increase PV hosting capacity.

The Graph Trace Analysis (GTA) is unique to the ISM model and allows for matrix-free calculations. A GTA power flow algorithm has been demonstrated to run up to 24 times faster than traditional, matrix-based power flow analysis approaches. The GTA power flow algorithm can solve transmission, radial distribution, lightly meshed distribution, and heavily meshed distribution all in the same model.

The GTA power flow-based, coordinated control algorithm is projected to save approximately $180,000 per year worth of energy on a feeder to be field tested at Pepco Holdings while reducing carbon emissions by approximately 746 US tons. The projected energy savings will be validated during field testing using an approach to weather-normalization of loads being employed at Pepco.

As illustrated in the figure below, weather, PV forecasts, utility bellwether AMI meter measurements, utility SCADA measurements, and customer smart inverter measurements are inputs to the software to be tested, referred to as the Model-Measurement Integrator for Ensuring Grid Security (M2IEGS).

M2IEGS: The Four Core Analysis Modules
M2IEGS synchronizes weather forecasts, PV generation forecasts, SCADA measurements, AMI voltage measurements, and PV inverter measurements. The software conducts analysis using four core modules as outlined below. These modules condense real-time and forecasted data into power flow forecasts, instability and abnormality warnings, and optimal set points for controllable devices.

Faster-Than-Real-Time (FTRT) Power Flow Analysis
On the half-hour in M2IEGS, a one-minute step size, 30-minute into the future PV forecast occurs. On the hour in M2IEGS a one-hour step size, 24-hour into the future PV forecast occurs. Using the weather-dependent load modeling employed at Pepco, corresponding load forecasts are also performed. Three types of power flow analyses are performed: power flow is run for every SCADA, AMI meter, and inverter measurement sample; a one-minute step size, time-series, power flow forecast is run every 30 minutes for 30 minutes into the future; a one-hour step-size, time-series, power flow forecast is run every hour for 24 hours into the future. If an event occurs, such as a switching operation or a detected abnormality, all power flow analyses are immediately performed.

Voltage Stability Analysis
The voltage stability analysis forecasts the voltage stability of lines and buses, providing notifications regarding low voltage stability margins or events that could lead to a voltage collapse, such as a loss of renewable generation below a transmission system bus. In the event of transmission system voltage instabilities, the coordinated control module switches to voltage stability mode and provides support to the transmission system.

Abnormality Detection Analysis
The Abnormality Detection analysis employs historical error statistics between power flow calculations and measurements (e.g., comparing power flow calculations of customer voltages with AMI meter measurements of the customer voltages), and uses error statistics between time series, power flow analysis (from the FTRT simulator) and field measurements (i.e. SCADA, AMI, and smart inverter measurements) to detect and flag abnormalities in real-time. Abnormalities can occur for a number of reasons, including cyber and physical attacks, failed controllers or instruments, and unknown field device operations.

Coordinated Control
The Coordinated Control module determines a 30-minute control schedule for smart inverters, and a 24-hour, hourly schedule for all control devices – Load Tap Changers (LTCs), voltage regulators, switched capacitor banks, and smart inverters. The Coordinated Control seeks to achieve a desired voltage profile by determining setpoints for devices under coordinated control that correspond to voltages from a time-series, optimal power flow solution. The desired voltage profile changes with the mode of control (e.g., maximize energy efficiency mode, voltage stability mode, or smart inverters not responding mode).

Success in the Lab
EDD collaborated with Florida State University and the University of Delaware to test M2IEGS under lab conditions. Together, they conducted three studies that validate the software’s availability to detect cyber-attacks on utility equipment and inverters, achieve energy savings with coordinated control, and respond to transmission system low voltage while cyber-attacks occur on PV inverters.

In the first study, the team evaluated the ability of the abnormality detection module to flag cyber-attacks on utility equipment and inverters using software simulations. The team selected 25 scenarios that investigated block, modify, or delay attacks on DER and inverter controls, SCADA controls, and meters that could change any set of measurement and/or control signals. Of the 25 scenarios, the Abnormality Detection successfully detected 22 attacks and partially detected 3 attacks. The team also tested the Abnormality Detection using two physical inverters in a hardware-in-the-simulation-loop. In this case, the team performed 23 denial of service, intermittent power, and modification-of-the-inverter-controller attacks on one or both of the inverters. The Abnormality Detection successfully detected 21 attacks.

In a second study, the team examined seasonal opportunities for carbon reduction, energy savings, and annual dollar savings by leveraging Coordinated Control on a two-transformer, open bus, single feeder serving 2,040 residential, commercial, and industrial customers. The team found that, compared to the existing control strategy, the use of coordinated control provided a 3.43% energy savings, corresponding to annual savings of $90.86 per customer, and an annual carbon reduction of 746 U.S. tons. The team also noted that the control devices on the feeder used in the study were not designed for optimal operations. Previous work has indicated that energy savings could be significantly increased through a redesign of the placement and size of the utility control devices over the time-varying load.

In the third study, the team paired a low-voltage event with a cyber-attack. They used two physical inverters in hardware-in-the-simulation loop experiments to test M2IEGS’s ability to simultaneously stabilize voltage using the Stability Analysis and Coordinated Control modules and detect cyber-attacks using the Abnormality Detection module. The team simulated a two-hour transmission system voltage drop, where the voltage dropped by about 10% of its nominal level. This voltage drop triggered the Coordinated Control to switch to Voltage Stability mode. The team then launched an intermittent real power attack, an intermittent reactive attack, and a standby mode attack to evaluate attack detection on different inverter control variables. The Abnormality Detection module detected all three attacks, although the Coordinated Control lost inverter voltage control capability during the reactive and standby mode attacks.

Through the three studies, the team found an overall abnormality detection success rate of 90.2%. The team presumes that the software can improve situational awareness for grid operators, detect and mitigate abnormal operations, reduce costs, and improve voltage stability margins in upcoming field tests at Pepco.

Preparation and Implementation in the Field
Nicholas Cincotti, Senior Project Manager, Smart Grid Project Execution at Exelon also spoke on the ISGT panel. He discussed Pepco’s coordination with EDD to implement M2IEGS in the field and validate the lab simulation and hardware-in-the-simulation loop studies, working with approximately 50 smart inverters, including one very large inverter, on two circuits. Pepco is working with EDD to enable the integration of their existing grid metering and monitoring systems, such as SCADA, AMI, inverter, load forecasting and PV forecasting to better monitor, manage, and control DERs on the grid. In the future, the utility sees a cost-effective opportunity to increase PV hosting capacity limits, especially in areas of their service territory that are currently at PV hosting capacity limits.

In preparation for field testing, beginning during the summer of 2023, Pepco is establishing communication to the M2IEGS software, which includes building, enabling, and testing communication paths, coordinating with internal and external stakeholders on change management, and developing a high-level project plan and schedule. The project will be implemented through a customer pilot, set to begin before the end of the year. Residential, commercial, and industrial customers with existing solar generation will be given smart inverter equipment that can send PV generation data to Pepco.

Pepco and the University of Delaware are working to upgrade and install communications infrastructure on the two test circuits and setting up a test-bench configuration to conduct smart inverter testing in coordination. Before the field testing and pilot go live, Pepco’s final steps will include building out and testing IT infrastructure, troubleshooting and testing field data sources, and deploying and testing smart inverters for participating customers.

Together, EDD and Pepco are taking the first step to implement and test the next generation of grid monitoring and control software in the field. Their collaboration, alongside project partners from academia, corporate solution providers, and industry organizations is making important strides towards building a secure and sustainable electric grid of the future.