Evaluation Method of Distributed PV Station Efficiency Based on Regression Optimization

Abstract

Comprehensive photovoltaic evaluation system can effectively analyze the efficiency of distributed photovoltaic power generation systems and is of great significance for improving corporate performance. This paper proposes an evaluation method of distributed by station efficiency based on regression optimization. It uses multiple linear regression algorithm to reduce the dimensionality of the evaluation indicators of photovoltaic power plants selected. By searching for the mainly impact indicators, the corresponding indicator data of plants are standardized, and the data envelopment analysis method is selected respectively. The input-oriented model and the output-oriented model are used to evaluate the overall efficiency, technical efficiency and scale efficiency of distributed PV stations and compare the experimental results. Finally, for the distributed PV station whose efficiency evaluation is not effective, finding the existing problems of photovoltaic power plants from the return on scale of plants and the analysis of main balance of indicators, and propose relevant suggestions at the same time.