Delay-Tolerant Predictive Power Compensation Control for Photovoltaic Voltage Regulation

• Published in: IEEE transactions on industrial informatics Vol. 17; no. 7; pp. 4545 - 4554
• Main Authors: Zhang, Zhanqiang, Mishra, Yateendra, Yue, Dong, Dou, Chunxia, Zhang, Bo, Tian, Yu-Chu
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 01.07.2021; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Artificial neural networks; Communication; Communication delay; Compensation; Delay; Delays; double neural network; Electric potential; Electric power distribution; Feeders; Inverters; Mutation; Neural networks; Photovoltaic cells; power compensation control (PCC); Predictive control; Reactive power; Sensitivity; Switches; Voltage; Voltage control; voltage regulation
• ISSN: 1551-3203; 1941-0050
• DOI: 10.1109/TII.2020.3024069

Voltage regulation is imperative for the successful operation of electricity distribution networks, especially with a high penetration level of photovoltaic (PV) systems. Power compensation control (PCC) that uses both reactive power compensation and active power curtailment has shown promising results in alleviating voltage rise problems. It crucially relies on real-time communications among distributed PV systems. However, the transmission of state measurements and control signals in PCC is hampered by inevitable communication delays. Therefore, it is important to not only estimate the maximum tolerable communication delay (MTCD) but also develop an alternative technique for PCC under abnormal communication delay (ACD) conditions. This article presents a delay-tolerant predictive PCC for voltage regulation in distribution feeders. After estimating the MTCD based on voltage and power mutation, it uses normal PCC for effective operation when communication delay is within MTCD, or switches to predictive PCC under ACD conditions. An accurate prediction is achieved using a double neural network with online adjustment of weights and samples. Simulations on a sample distribution network demonstrate the effectiveness of our presented approach.