论文部分内容阅读
由于气象环境复杂多变且具有动态的不确定特性,选取太原市2014年至2015年的空气污染物监测数据,将模拟退火算法(SA)与粒子群算法(PSO)相结合并对其进行改进,优化支持向量机(SVM)完成参数寻优,并运用偏最小二乘法(PLS)分析各污染物因子间的相互作用,构造出一种新的空气质量评价模型.实验结果表明,改进的SAPSO-SVM与PSO-SVM和SVM相比,模型运行时间短、等级分类精度高,具有良好的评价性能,为空气质量评价提供了新思路.
Due to the complicated and fluctuating meteorological environment and its dynamic and uncertain characteristics, the data of air pollutants monitoring in Taiyuan from 2014 to 2015 are selected and the SA algorithm is combined with the particle swarm optimization (PSO) to improve it (SVM) to optimize the parameters, and PLS was used to analyze the interaction between various pollutants to construct a new air quality evaluation model.The experimental results show that the improved SAPSO Compared with PSO-SVM and SVM, SVM has short run time, high classification accuracy and good evaluation performance, which provides a new idea for air quality evaluation.