The model and application of the financial risk forecast in electric power enterprises based on improved BP neural network algorithm
Zhibin Liu, Shaomei Yang · 2007
For the particularity of electric power enterprises themselves, the commonly methods used to forecast their financial risk is limited and inadequate. To forecast the financial risk of the power enterprises scientifically and accurately, this paper proposes the improved BP neural network imports the adjustable activation function and Levenberg -Marquardt optimization algorithm. The improved model not only simulate the expert in forecasting the financial risk and avoiding the subjective mistakes in the evaluation process, but also enhance the learning accuracy and the algorithm convergence speed greatly. The financial risk forecast of 12 power enterprises in National Power Company shows that the improved model is stable and reliable, and this method to forecast the financial risk of the power enterprises is feasible.