Application of Parameter Optimized SVM Model in Building Deformation Prediction
When the support vector machine (SVM) model is used for deformation prediction,the selection of model parameters di-rectly affects the deformation prediction results. In order to obtain the optimal parameters,particle swarm optimization (PSO) algo-rithm and genetic algorithm (GA) are introduced into the SVM model to build a combined prediction model. The new combined pre-diction model can effectively improve the global search ability,the prediction accuracy and the data adaptability of the model. Taking the measured building deformation data as an example,the results show that compared with other prediction models,the combined prediction model proposed in this paper has higher accuracy and stability,and is suitable for the deformation data prediction of actual engineering projects.
support vector machineparticle swarm optimizationgenetic algorithmdeformation predictionaccuracy analysis
万方数据
维普
