Research On Risk Management Of Water Conservancy Project Based On BP Neural Network

With the vigorous development of China’s water industry, the increasing of all kinds of water conservancy project, various risks have occurred. Our country is a large shipping country, do a good job of water conservancy project risk management accurately and effectively plays an important role in the safety of hydraulic engineering. Effective risk management, is bound to reduce the loss to the country and the people injury.This paper is mainly directed against the existing process in water conservancy project construction risks, risk risk identification, establish the model of appropriate evaluation indicators, and using the BP neural network modeling method, which using the established model to evaluate the risk of water conservancy projects and draw the corresponding conclusions and proposals, risk control and supervision, the formation of a set integrated risk management model.Firstly, through the relevant information on the basic theory of risk and risk management of water conservancy project is expounded, and each of the risk management process is analyzed, then the principle of BP neural network are introduced.In fact, the use of a questionnaire survey on water conservancy project risks hidden investigation analysis, to identify the main risk factors of water conservancy project, and establishes the corresponding evaluation index system, and then using expert scoring method to quantify 10 sample selection. Then in order to simplify the input of the network, improve the convergence rate of the network. Therefore, when the input data dimension is large, the use of SPSS software, using principal component analysis to reduce the dimensions of the sample data, and data discretization, which simplifies the structure of BP network, improve the training rate.Again, the dimensionality reduction processing of hydraulic risk evaluation index system, using the MATLAB tool, the former 8 samples as the training data in BP neural network, after 2 samples as test data. The test result shows that the risk evaluation model is effective.Finally, carries on the risk management of the project in the construction of drainage pumping station, using principal component and BP neural network combination of the risk assessment, and analyzed the influence of the key risk factors of drainage station according to the coefficient of the evaluation results and main composition of the formula, take targeted measures to reduce the adverse impact on the project of these risk factors.