Construction And Analysis Of Performance Evaluation Model Of Electric Supply Business Office

Power supply business office is the most basic organizational unit of State Grid Corporation of China.It provides the first-line power supply services for the majority of urban and rural residents.Its work quality not only affects the overall operating income and future development of the company,but also shows the social image of State Grid Corporation of China.A reasonable performance evaluation model can effectively measure the work quality of power supply business,its evaluation results can not only correctly guide the management behavior of the superior managers,improve the level of decision-making management,but also help each power supply business to find out the leakage and make up the deficiency,and improve the efficiency of the organizational unit.Therefore,the performance evaluation model of power supply business office must adapt to the new environmental requirements and competition needs,innovate and change.First of all,this paper takes the performance evaluation system of the power supply business office of C City as the research object,analyzes the shortcomings of the existing evaluation index system,and puts forward a more comprehensive performance evaluation index system.The index system can not only effectively measure the short-term performance of power supply business,but also reflect its future development ability.In order to eliminate the influence of managers’ subjective factors on the results of performance appraisal and make the results of performance appraisal as real and effective as possible,this paper proposes a combined evaluation model for performance appraisal of power supply business,which consists of three stages: first,use BP neural network to train the short-term performance index model of power supply business;then use TS fuzzy neural network to establish the development of power supply business Finally,using the learning results of the above two models as the input of the combined model,a combined evaluation model based on TS fuzzy neural network for performance appraisal of power supply business is established.In order to carry out model simulation training,the performance evaluation indexes of 100 power supply business centers in City C are collected and sorted out,and the performance grades of each power supply business are obtained,and the dimensional influence among sample indexes is eliminated through standardized processing.In view of the large number of indicators involved in the short-term performance indicator system,in order to overcome the high similarity between some indicators and eliminate redundant information,the principal component analysis method in factor analysis is used to reduce the dimension of short-term performance indicators,so as to reduce the number of network model input variables,simplify the network model structure and save the network model learning time.The training and testing of short-term performance,development ability and combined assessment model are realized by programming.The test results show that the output of the combined evaluation model can accurately evaluate the performance level of the power supply business.In order to verify the superiority of the combined evaluation model of performance evaluation of power supply business place proposed in this paper,under the same experimental conditions,the performance evaluation model of power supply business place based on BP neural network is constructed and trained by using sample index data,and the output evaluation results of the model and the combined evaluation model are compared and analyzed.The results show that the output accuracy of the combined evaluation model is better than that of the performance evaluation model based on BP neural network model.The research results enrich and perfect the theory and method of performance evaluation of power supply business,and provide a new way for performance evaluation of power supply business.