Entropy weight-set pair analysis (SPA) for dam leakage detection

Abstract

The dam seepage flow contains many kinds of materials, forming the information flow, the information entropy of which can be used to determine the nature of the flow field; and the water chemistry and environmental isotope detection methods have been widely applied. Based on the knowledge of information entropy and set pair analysis, a new method consisting of water chemistry and environmental isotope detection, entropy weight-set pair analysis model, is introduced to ascertain the dam concentration leakage channels. Firstly, PH, conductance, Cl, δD and δO are chosen as five detection indexes, and the connection degree of each index value to the classification standard is calculated. Then, the weight of each index is determined by using the information entropy weight method based on the principle of the maximum entropy. After that, the comprehensive connection degree of all samples from the drilling is calculated, and thus all the samples are classified so as to determine the resource of water. Finally, this new method is applied to detect the dam leakage channels of one south reservoir and the results are tested and improved by the method of horizontal velocity and temperature detection. It is concluded that the results of the entropy weight-set pair analysis model match those of the horizontal velocity and temperature detection very well.