基于改进自适应粒子群算法的混合核函数最小二乘支持向量机大坝变形预测

针对大坝变形影响因素的复杂性以及监测数据的非线性、随机波动大和预测难度大等问题,提出一种改进自适应粒子群(particle swarm,PSO)算法的混合核函数最小二乘支持向量机(least squares support vector machine,LSSVM)模型,实现了大坝水平变形的时间序列预测方法.基于Mercer理论,将多项式核函数和高斯核函数进行线性组合,构建混合核函数,作为LSSVM模型的核函数,并以特征因子与大坝变形间的相互联系为基础,采用动态自适应惯性权重的PSO算法,对混合核函数的LSSVM模型进行参数寻优,以确保建立最佳LSSVM预测模型.将模型应用于丰满大坝,并与传统多项式核函数和传统高斯核函数的LSSVM模型进行对比仿真实验,对所提方法的有效性和准确性进行验证评估.结果表明,该模型在预测精度上有了明显提高,预测性能尤佳.可见改进自适应粒子群的混合核函数LSSVM模型对大坝变形的时间序列预测有良好的实用价值.

Prediction of Dam Deformation Based on Hybrid Kernel Function LSSVM Based on Improved Adaptive Particle Swarm Optimization

Aiming at the complexity of dam deformation influencing factors, as well as the nonlinearity of monitoring data, large random fluctuations and large prediction difficulty. in this paper , a hybrid kernel function least squares support vector machine model with improved adaptive particle swarm optimization (PSO) algorithm(LSSVM) is proposed, a time series prediction method for horizontal deformation of dams. Based on the Mercer theory, the polynomial kernel function and the Gaussian kernel function are linearly combined to construct a hybrid kernel function as the kernel function of the least squares support vector machine model. Based on the correlation between the characteristic factor and the dam deformation, dynamics are adopted. The particle swarm optimization algorithm with adaptive inertia weights optimizes the parameters of the least squares support vector machine model of the hybrid kernel function to ensure the optimal LSSVM prediction model. The model is applied to the plump dam, and the simulation experiment is carried out with the traditional polynomial kernel function and the least squares support vector machine model of the traditional Gaussian kernel function. The validity and accuracy of the proposed method are verified and evaluated. The results show that the model has a significant improvement in prediction accuracy, and the prediction performance is particularly good. It can be seen that the improved kernel function least squares support vector machine model with improved adaptive particle swarm has good practical value for time series prediction of dam deformation.