Research On Slope Deformation Prediction By Support Vector Machine Model Based On Improved Cuckoo Search Algorithm

In recent years,due to the repeated occurrence of landslide incidents,it has brought huge disasters and impacts to the people.Therefore,it is necessary to carry out safety monitoring of slopes and predict their deformation trends.Aiming at the shortcomings of traditional prediction methods,this paper proposes an improved cuckoo search algorithm(ICS)and optimizes the parameters of support vector machine(SVM)to construct a support vector machine combined prediction model based on improved cuckoo search algorithm(ICS-SVM).)and applied to the actual slope deformation prediction.Firstly,the damage caused by landslide is expounded,and the current research status of slope deformation prediction is analyzed.The SVM is used to predict the deformation of slope.For the problem that the SVM parameters are difficult to select,the CS algorithm is introduced to optimize the parameters of the SVM.Secondly,the ICS algorithm is proposed for the traditional CS algorithm,such as low optimization accuracy and slow convergence speed.Seven benchmark functions are used to test the optimization performance of the ICS algorithm and compare it with the CS algorithm.The experimental results verify that the ICS algorithm has obtained More accurate optimization results and faster convergence.Then the ICS algorithm is used to optimize the SVM parameters to construct the ICS-SVM slope deformation prediction model.Finally,the ICS-SVM model is applied to the Madiwan slope and Huapingzi slope examples,and compared with the CS-SVM model.The mean square error(MSE)and the mean ralatived error(MRE)are used to evaluate Prediction accuracy.The results show that for the Madiwan slope,the MSE and MRE of the CS-SVM model are 1.05 and 3.06%,respectively,and the MSE and MRE of the ICS-SVM model are 0.07 and 0.76%,respectively.For the Huapingzi slope,CS-The MSE and MRE obtained by the SVM model are 1.05 and 5.72%,respectively,and the MSE and MRE obtained by the ICS-SVM model are 0.05 and 1.09%,respectively.It can be seen that the MSE and MRE obtained by the ICS-SVM model are smaller than the CS-SVM model,which verifies that the ICS-SVM model has higher prediction accuracy and can better reflect the actual deformation of the slope,which has certain reference value.