The Research On Intelligent Monitoring Model And Software System Of Concrete Dam

Dam safety monitoring is based on the measured data of dam and applying some mathematical method to establish a monitoring model that can effectively reflect the relationship between effect set and load set for simulating the running state of dam and then comprehensively evaluating the health status of dam.It is the most commonly used method to ensure the dam operation safely and effectively.After many years of hard work of domestic and overseas people in water conservancy industry,the research and application of dam satety monitoring theory and method have made great progress,which play a huge role in ensuring dam safe running.However,there are still problems and shortcomings.In this paper,based on the study about the traditional statistical model of concrete dam,introducing the extreme learning machine(ELM)algorithm and establishing the ELM monitoring model for the shortcomings of the traditional statistical model,and studying its performance by using measured data of the Fengman dam.Study result shows,by contrasting with a variety of traditional monitoring models,the ELM monitoring model has great advantage in concrete dam safety monitoring.At the same time,in order to evaluate the running state of dam in a more comprehensive and accurate way,and to meet the needs of practical engineering applications,developing a dam data analysis software system containing different performance monitoring models based on the MATLAB language.This research work not only plays an important role in practical engineering,but also is significant to the research of dam safety monitoring in China.The main reseach contents of this paper are as follows:(1)Comprehensively summarizing the reseach results of dam safety monitoring statistical model at home and abroad,and studying various traditional statistical monitoring models including the multiple linear regreession model,stepwise regression model,BP neural network model with the measured data of Fengman dam.The shortcomings of traditional monitoring models are pointed out.(2)The extreme learning machine(ELM)algorithm is introduced in detail,and establishing the ELM monitoring model based on the measured data of Fengman dam.Compared with the multiple linear regreession model,stepwise regression model,BP neural network model in performance,the ELM monitoring model has the advantages of easy operation,fast calculation speed,high prediction precision etc,which makes up the lacks of traditional monitoring models in many aspects.(3)Based on the MATLAB language,the DDAS(dam safety monitoring data analysis software system)of dam safety monitoring is developed,and the simplicity and validity of the system are verified.