APSO-WLSSVM算法在水质预测中的应用研究

Research on Water Quality Prediction Based on APSO-WLSSVM

针对参数选择是影响加权最小二乘支持向量机水质预测效果的关键,给出了基于自适应粒子群优化算法参数优选的WLSSVM回归预测的建模过程,以大伙房水库为例,预测了库区水质主要影响因素,并与未优选的WLSSVM预测结果进行对比。结果表明,该方法参数寻优更可靠、快速,预测精度高。

Getting better parameters values is one of the key factors for weighted least squares support vector machine（WLSSVM）,which affect the water quality prediction.In this paper,adaptive particle swarm optimization algorithm based on weighted least squares support vector machine regression modeling process is studied.Taking Dahuofang Reservoir for an example,the main factors influence of water quality in reservoir area is predicted.Compared with APSO-WLSSVM and non-optimized WLSSVM,the results show that the proposed method is more reliable,fast and has good prediction accuracy.