The analysis of dam seepage monitor is an important means of behavior to master the dam running. It is also an important basis for judging the safety of the dam. Therefore, timely, reasonable and effective analysis of the material to the dam is important to the safety of the dam. Factors which affect the dam safety are various and complex. How to make full use of the information and make objective and accurate assessment of the safety of the dam is one of several problems to be solved in the dam monitor. This paper introduces the situation of present times in and out of China. At the same time, the basic theory and structure of stepwise regression analysis,BP neutral network and Generalized Regression neural network are also introduced. According to the monitoring materials of Hualingting Reservoir, the model of stepwise regression analysis, BP neutral networks and Generalized Regression neural networks are finished. Comparing the advantages and disadvantages of these three models, the separate characteristics and applicability of these models are researched. This can give reference and choice to the dam monitoring in the future.
|