Research of the risk assessment of thermal power project investment based on Artificial Neural Networks

This paper according to the investment phase of thermal power projects analyzes the risk factors affecting project investment, on this basis, establishes risk assessment index system of thermal power project investment, and proposes a viable risk assessment model based on BP (Back Propagation) neural network. This model uses BP neural network´s self-learning feature, through amending the weight constantly in the training process, to make the network´s actual output vector be close to the expecting output value gradually. Through simulation of the example by MATLAB neural network tool, the author verifies the reliability of the model, the study of the example shows that this method provides an effective management tool for the risk assessment of thermal power project investment.