Research On Value Evaluation Of Photovoltaic Enterprises Based On BP Neural Network Model

With the introduction of "carbon peak" and "carbon neutral" policies,how to achieve deep decarbonization of power generation industry has become a key issue of national concern.New energy is gradually replacing traditional energy by virtue of its many advantages,and in these new energy field industries,photovoltaic power generation has a broader market development space and development prospects.Nowadays,with the continuous expansion of photovoltaic installed capacity,the transfer,investment and other businesses of photovoltaic enterprises are also gradually increasing.However,there are relatively few researches on the value assessment methods of photovoltaic enterprises at home and abroad,and no unified conclusion has been reached on the selection of the value assessment methods of photovoltaic enterprises.Based on the principle of market method,this thesis evaluates the value of photovoltaic enterprises through BP neural network model,and makes a judgment on the evaluation accuracy.The purpose is to provide reference for the transfer and investment of photovoltaic power generation enterprises and enrich the value assessment methods of photovoltaic enterprises.Firstly,on the basis of theoretical analysis,the applicability of BP neural network model in photovoltaic enterprise value assessment is expounded.This thesis analyzes the factors that affect the value of photovoltaic enterprises from the aspects of financial factors and non-financial factors,and finally sorts out 15 influencing indicators.Secondly,this thesis collected 15 indicators and enterprise value data of100 photovoltaic enterprises,and in order to improve the accuracy of model prediction,based on the principal component analysis method,dimensionality reduction processing was carried out on the impact indicators,and 7 principal component indicators were extracted according to the principle of cumulative contribution rate greater than 85%.Thirdly,L-M algorithm is introduced to optimize the BP neural network model.In 100 sample enterprises,90% photovoltaic enterprises are taken as training samples,and the remaining 10% photovoltaic enterprises are taken as test samples.The value of photovoltaic enterprises is evaluated,and the evaluation results before and after model optimization are compared.The evaluation accuracy of the improved BP neural network model was analyzed.Finally,the value added method is used to evaluate the value of Risen Energy Co.,Ltd.,one of the test enterprises,and the evaluation results are compared with the evaluation results obtained by the BP neural network model optimized based on L-M algorithm.It is found that the evaluation results of the optimized BP neural network model are closer to the real value of Risen Energy Co.,Ltd..Based on these studies,this thesis finally draws the conclusion that BP neural network model has good applicability in photovoltaic enterprise value assessment.This thesis has 12 figures,28 tables and 80 references.