Fuzzy Neural Network Based Optimal and Fair Real Power Management for Voltage Security in Distribution Networks with Hig
- PDF / 1,836,252 Bytes
- 8 Pages / 595.276 x 790.866 pts Page_size
- 42 Downloads / 215 Views
ORIGINAL ARTICLE
Fuzzy Neural Network Based Optimal and Fair Real Power Management for Voltage Security in Distribution Networks with High PV Penetration Jindong Yang1 · Junyuan Luo2 · Haitao Zhang2 Received: 12 December 2019 / Revised: 13 July 2020 / Accepted: 18 August 2020 © The Korean Institute of Electrical Engineers 2020
Abstract The high penetration of distributed generation (DG) sources in distribution networks (DN) can induce overvoltage issues. In this paper, an artificial intelligence based fair and optimal method for voltage regulation in DN with high photovoltaic (PV) penetration is proposed. Based on the forecasting of solar radiance and load profiles, the method optimally dispatches the generation of PVs to prevent overvoltage with the objective of minimizing the energy curtailment of PVs for a given long period. In addition, the RPCM can adaptively adjust the curtailment of PVs based on fuzzy neural network algorithm so that the PV systems in the DN could reach and keep similar accumulated curtailments during the period. Steady state simulation studies under various scenarios have been carried out on a 69-bus distribution feeder and an actual distribution network to demonstrate the effectiveness of the proposed method. Keywords Distributed generation · Distribution network · Optimal dispatch · Photovoltaic · Voltage regulation
1 Introduction Having more and more distributed generation (DG) sources including solar and wind energy in power systems has been viewed as an effective way to satisfy the high requirement of energy security and environmental protection. Therefore, the penetration level of DG systems is experiencing a steady increasing in distribution networks (DN) in the U.S. as mentioned in [1]. It was reported that the utility-scale solar power from both concentrating solar power (CSP) technology and photovoltaic (PV) panels reached to 27,897 MW by the first quarter of 2020, which took 20.7% of the demand of California ISO [2]. The high penetration of DG in DN can induce overvoltage issues. Especially under low load levels when the PV generation is higher than the load demand in the DN, the extra power is going to be fed back to the grid, which can cause overvoltage. Therefore, overvoltage is a common and * Jindong Yang [email protected] 1
Yunnan Power Grid Electric Power Research Institute, Kunming, Yunnan, China
Lincang Power Supply Company, Lincang, Yunnan, China
2
important issue for DN with high penetration of DG and has inspired a lot of studies. In [3] and [4], a straightforward method was proposed to regulate the voltage by reducing DG power production directly. According to [5] based on the dynamic Thevenin equivalent, a real-time prediction algorithm was proposed to calculate the active power limit as a reference of the generation of PVs. In [6], a network partition-based zonal method was proposed for the voltage regulation. This method used voltage sensitivity indexes to participate the DN in to different arears based on local reactive power balance, and the
Data Loading...
