Reasonable Range Prediction and Loss Reduction Governance Model of Line Loss in Station Area Based on GRNN

With the continuous expansion of the grid connected scale of distributed power generation, it is urgent to accurately calculate the line loss rate of the station area with distributed power generation. However, at present, there are technical bottlenecks in the acquisition and calculation accuracy of operating parameters in low-voltage active stations, so it is difficult to realize the reasonable range prediction of line loss and high loss identification. This paper fully considers the basic operation attribute of the station area and the grid connection attribute of distributed energy, and proposes a calculation method of GRNN station area line loss rate based on feature extraction. Firstly, the line loss influence factors of the station area with photovoltaic distributed power supply are extracted. Secondly, each station area cluster is trained by GRNN, and the error between the training results and the real line loss is analyzed; finally, the ability of the proposed algorithm in reasonable interval prediction and high line loss rate station area recognition is tested.