Study On Optimization Of Landslide Treatment Plan Based On Multiple Attribute Decision Making And Landslide Monitoring Analysis

Landslide control is a hot spot in disaster control engineering. It is the important foundation of landslide control project to choose economic, reasonable and effective landslide control scheme. But in the actual landslide prevention and control, the result of the control is not good due to the unreasonable selection of the control scheme and great waste of resources has been caused because of the emergency management and the change of the plan in the later period. Therefore, it is very important to choose the optimal treatment plan before the landslide control project is carried out. In this paper, based on four governance schemes proposed at the initial stage of a landslide in Yanan,AOWEA operator and risk attitude factor are used to optimize the four alternatives respectively and sensitivity analysis of risk attitude factors ? of the second methods was also carried out. Then, according to the monitoring data of surface deformation, the effectiveness of the actual implementation of the governance program is evaluated. Finally, based on the Elman neural network, the deformation of the landslide is predicted according to the ground deformation monitoring data.The main results of this paper are as follows:(1) The optimization model of landslide treatment scheme is established based on AOWEA operator and the optimizing sorting results of landslide treatment scheme is S1(?)S3(?)S4(?)S2.(2) The optimization model of landslide treatment scheme is established based on risk attitude factor and the optimizing sorting results of landslide treatment scheme is that when??[-0.20, 0.50], S1(?)S2(?)S3(?)S4; when ??[-0.50, -0.20), the composite attribute utility value of S4 exceeds S3, S2 and S1 in turn, and when ? = -0.5,S4(?)S3(?)S2(?)S1.(3) Sensitivity analysis of program ranking under the influence of risk attitude and the result is that the sensitive interval of S3 and S4 is [-0.21, -0.20], the sensitive interval of SZ and S4 is [-0.34, -0.33], and the sensitive interval of S1 and S4 is [-0.48, -0.47].(4) According to the surface deformation monitoring and analysis, the effectiveness of the implementation of the governance program is evaluated, and the results show that the accumulative displacement, velocity and acceleration of ground surface deformation in the middle landslide area are large, and the effect of engineering treatment is poor; the accumulative displacement, velocity and acceleration of the ground surface deformation in the southern landslide area are small, and the effect of engineering control is remarkable; the accumulative displacement, velocity and acceleration of the trailing edge of the landslide in the north area are smaller, but the accumulative displacement,velocity and acceleration of the deformation in the middle of the landslide are larger, and the effect of the project is general.(5) Based on Elman neural network, the dynamic prediction of landslide deformation at monitoring points J24 and J29 is taken as an example. And the results show that the deformation prediction curve of J24 and J29 is basically consistent with the measured value curve, the maximum error of the predicted value and the measured value is respectively 8.00% and 5.84%, and the average error is respectively 1.72% and 1.07%.