Slope Displacement Prediction Based On Genetic Ant Colony Combination Algorithm For Optimizing SVM Model

The displacement of the slope is the key factor of its stability.Therefore,the monitoring and prediction of the slope displacement is indispensable in the construction of the slope.By establishing an effective prediction model to analyze the law and trend of the slope displacement,the slope is determined.Stability is the ultimate goal of slope displacement prediction.In view of the commonly used prediction models,there are certain problems in model establishment and prediction accuracy.Based on the theory of support vector machine regression machine,this paper proposes combining genetic algorithm and ant colony algorithm and improving the pheromone of ant colony algorithm.The generation strategy is used to optimize the support vector machine parameters.Finally,the support vector machine prediction model based on this combination algorithm is applied to the slope displacement prediction.This paper first expounds the research background and significance of slope displacement prediction,and the current research status of slope deformation prediction and support vector machine,and introduces the basic genetic algorithm and ant colony algorithm and related theoretical knowledge of support vector machine.Aiming at the problem of selecting kernel function in the support vector machine slope prediction model,the research work is carried out.The two commonly used kernel functions are selected to establish the prediction model and the prediction experiment is carried out.The results show that the Gaussian radial basis kernel function is applied.The prediction model has a better prediction effect.Therefore,the Gaussian radial basis kernel function is chosen to carry out the research work of building a support vector machine slope prediction model based on the combination algorithm.Secondly,in order to solve the problem of support vector machine parameter selection,the genetic algorithm and ant colony algorithm are combined,and the pheromone generation strategy in ant colony algorithm is improved,which is applied to the parameter optimization of support vector machine.In order to verify the advantages of the combination algorithm,the experimental research on the shortest path of the business travel is carried out.The results show that the optimization performance of the combined algorithm is higher than the optimization performance of the two algorithms.Therefore,this combination algorithm isapplied to the optimization of support vector machine parameters,and a support vector machine slope prediction model based on combination optimization algorithm is constructed.Finally,this model is applied to two engineering examples,and the Matlab program is written to experiment with it.The accuracy is compared with the support vector machine prediction model established by genetic algorithm and ant colony algorithm respectively.The average relative error is given.Evaluation.The final experimental results show that the average relative error of the support vector machine slope prediction model based on genetic algorithm optimization is 3.07% and 3.21%.The average relative error of the prediction model based on ant colony algorithm is 1.82% and 1.63.%,and the prediction model constructed based on the combination optimization algorithm has an average relative error of1.02% and 0.97%.It can be seen that the slope prediction model based on the combination algorithm to optimize the parameters of the support vector machine has a better prediction effect and can be applied to practical engineering.