Research on risk evaluation of power marketing security based on improved BP neural network

With the development of power market, the power company's targets and emphases are changing. Therefore, strengthening the risks' management and control of power marketing security, in order to reduce the company's losses, becomes more and more important. This paper first built an evaluation index system of marketing security risk, proposed an improved BP neural network model, combining principal component analysis (PCA), BP neural network and expert scoring method, then calculated the level of marketing security risk based on the treated data. At last, a test is made to demonstrate the effectiveness of this improved model. The result shows this method could accurately evaluate the security risks of marketing, with strong practical significance for avoiding the marketing security risks of power companies.