The Study On The Evaluation Model Of Clean Production Based On BP Neural Network

The auditing of cleaner production is an important link in cleaner production. Its main role is to diagnose unclean factors in enterprise, to bring up solutions, to obtain the purposes of saving energy, reducing consumption and pollution and additional benefits. However, in the auditing of cleaner production, to determine ranks of cleaner production in enterprise and the size of potential for cleaner production in enterprise, the improve production direction and how to carry out the implementation of works in cleaner production, the qualitative elements are more than the quantitative elements. Generally, the method of single indication is used in quantitative research, and these experts' subjectivity is high during the operation process, thus the cleaner production level evaluation in enterprise lacks scientific and accuracy. Therefore, the research of scientific and effective methods for the auditing of cleaner production has an important significance on accurate assessment the level of clean production in enterprises and the implementation of sustainable cleaner production.In view of above problems, this paper carries out a new exploratory study to the methods of cleaner production in enterprise. The approach of neural network are used in the evaluation research of cleaner production, and calculation, analysis and evaluation the level of clean production in enterprises, improve efficiency the auditing of cleaner production, then confirm the standarad and compile the programs, using computer to share the information, technology and experts, to reduce the human labor.The paper describe the research status of cleaner production, based to the comparison and analysis of model theories and calculation methods in various types of neural network, the BP model is employed in the evaluation research of cleaner production. Firstly, the model parameters are determined, and then the comprehensive evaluation model of multi-indicators and information database of cleaner production are established, the relevant softwares are developed, And the model is tested through examples simulation. Finally, the model is applied to the auditing of cleaner production in the industry of tyre production. The results show that the method unify the qualitative indicators and quantitative indicators, give a clear quantitative result, and could transfer the relatively complicated and uncertain multi-factors problem into reasonable model parameters, and provide scientific evaluation methods to the quantitative assessment and indicators management of cleaner production in enterprise. Through the application in actual case, the evaluation has proven that the evaluate method of BP model has high capability of mapping nonlinear function and ability of self-adaptation, self-learning. And the evaluate method could effectively overcome the defects of strong subjectivity in the traditional evaluation method, and obtain the results too short, reduce people's labor strenghth, weaken the influence of index weight factors which are determined by man-made in traditional evaluation methods and has high precision.