The Application Of Artificial Neural Network On Safety Monitoring Of Dams

It is an important means to analyze the observation material to mastery the working state of dams, the safety degree of dams also can be judged according to the observation material. So it is one of the important tasks to make timely, reasonable and effective analysis for dam observation data.There are many influence factors of dams which are complex. It is an urgent problem in dam safety monitoring how we can make full use of masterying information and evaluate the safety of dams objectively and truly. There are some shortages of traditional mathematical models in application. The improved BP neural network is studied with optimizing classical BP neural network sufficiently, then is used in the safety monitoring of dams. The main contents of this paper are as follows:1. The BP neural network arithmetic is studied. Uniting the study fruit before, Levenberg Marquardt arithmetic is introduced to improve classical BP neural network, the BP neural network based on Levenberg Marquardt arithmetic is brought forth, which solve the problem of easily relapsing into local minimization of classical BP neural network.2. The configuration of neural network is optimized. According to the characteristics and working manners of neural network, the golden section principle which is used in living abroad is introduced to confirm the node of hidden layer of network. This method not only can conquer the shortage of traditional methods, but also can find the finer node of hidden layer quickly.3. The self-correlation correction for initial weights of network is studied. The initial weights is corrected with the neural node of every layer and transfer function. This method is different from traditional method which get initial weights randomly, and the stability and convergence precision of network is improved.4. The improved BP neural network is put forward with colligating every improved methods which are studied. The improved BP neural network can effectively conquer the shortages of classical BP network.5. The improved BP neural network is used in setting up the factors monitoring model, three-dimension displacement model and whole model. The whole model considers the influence factors of dams sufficiently. Compared with one point model or local model, the whole model can show the state and developing trend of dams roundly and truly.6. Based on analyzing and summing up the development of safety evaluation solution for dams, the evaluation solution of improved BP neural network is put forward. The perambulation and inspection message of dams is thought about, which also is subdivided and expressed in symbol. With every influence factors, the evaluation solution of improved BP neural network is used to evaluate the safety of dams, and the evaluation result is analyzed and summed up.