Application of High Density Electrical Method Based on BP Neural Network in Detection of Desilting Cofferdam in Reservoir
High density electrical method has the characteristics of large data collection,high efficiency and abundant inversion information,and has been widely used in the field of reservoir dam disease detection.At present,the inversion based on least square method is easy to be affected by the local extreme value of geo-electric data,which makes the detec-ted location and scale of the disease inaccurate.In view of this situation,this paper establishes the forward modeling mod-el of abnormal body with different parameter values,shape size and location distribution,and takes the model data as training samples to build the high-density electrical inversion model based on BP neural network.The trained inversion model was applied to the analysis of high-density electrical detection results of the cofferdam in the back row of the dred-ging and expansion dam of Reservoir.The results show that the proposed method can reduce the shielding effect caused by local current extremum,narrow the scope of hidden trouble detection,improve the accuracy of hidden trouble resolu-tion and inversion accuracy of high-density electrical method under the influence of high resistance shielding,and make more accurate interpretation of geophysical data.
high density electrical methodBP neural networkinversion modeldesilting cofferdamhazard detection
NETL
NSTL
万方数据
