Intelligent Implementation Technologies on Sensing Dam Safety Based on Neural Network

The equipments on sensing dam safety works usually under extremely bad working conditions. The reliability, stability and accuracy, etc, are very difficult to be guaranteed. The signal is sensitive to noise. The fault is often caused. Micro-electronics technology, computer science and artificial intelligence technologies provide strong technical support and security on improving the shortage of technologies on sensing dam safety, raising the level of automation, intelligent of dam safety monitoring. The Artificial Neural Network (ANN) has strong nonlinear fitting ability, learning function and parallel processing ability. Above excellent features of ANN are used to implement the adaptive suppression for noise and self-diagnosis for faults of sensors. The intelligent principle, method and realization way are presented. The constitution and training algorithm of an adaptive neural network filter are proposed. With this filter, the useful quantitative information can be extracted automatically from noise data. The information can describe the characteristics of detected objects. A fault diagnosis method of nonlinear observer is proposed. The nonlinear dynamic relation between input and output of the system can be obtained by use of the learning function of radial basis function neural network. The error can be calculated and the logical judgment can be made in real time with the proposed observer.