Logo image
Back
Intelligent Coordination and Control of Distributed Energy Resources, OLTC and SCB in Smart Grid Using X-DNN
Journal article   Open access   Peer reviewed

Intelligent Coordination and Control of Distributed Energy Resources, OLTC and SCB in Smart Grid Using X-DNN

Tolulope David Makanju, Ali N Hasan and Thokozani Shongwe
IEEE open access journal of power and energy, v 13, pp 321-332
21 Apr 2026
DOI: https://doi.org/10.1109/OAJPE.2026.3686384
Featured in Collection :   Drexel's Newest Publications
url
https://doi.org/10.1109/OAJPE.2026.3686384View
Published, Version of Record (VoR) Open

Abstract

Communication systems deep neural network Distributed energy resources grid-connected inverter Millimeter wave integrated circuits MIMICs Monolithic integrated circuits Network topology optimal control Radio access networks Regional area networks Voltage multipliers voltage regulating devices Capacitors Telecommunications
The introduction of distributed energy resources (DERs) into distribution networks has caused a lot of challenges, such as power flow imbalance, voltage fluctuation, and operation control of voltage regulation devices. The optimization techniques used in power systems to address these issues have disadvantages, such as slow computational speed. To this end, we developed a two-stage control strategy for modern power system networks, wherein an Explainable-Deep neural network (X-DNN) controller was developed and trained using the output responses of centralized multi-objective AC optimal power Flows. The first stage involved the use of offline optimization methods by embedding a controller to generate optimal set points of the grid-connected inverter (GCI) and voltage-regulating devices' responses under the uncertainty of Photovoltaic (PV) power output and loading conditions, forming the training dataset for the X-DNN model. The trained X-DNN controller was deployed for real-time control, and the simulation of this approach was tested on the IEEE 69-bus system with a high penetration of PV grid-connected inverters (GCI), and compared with particle swarm optimization (PSO) and the genetic algorithim (GA). The results confirm that the X-DNN controller offers robust and efficient regulation, capable of maintaining system voltage stability by reducing the average voltage deviation by 34.7% and 56.69% compared to GA and PSO, respectively, and by 7.47% and 14.52% in active power losses for GA and PSO, respectively. Moreover, the X-DNN controller offers faster computational performance, making it a promising framework for smart power networks.

Metrics

1 Record Views

Details

Logo image