基于人工鱼群和分形维数融合SVM的空气质量预测方法

发布时间：2018-01-14 05:02

本文关键词：基于人工鱼群和分形维数融合SVM的空气质量预测方法　出处：《模式识别与人工智能》2016年12期 　论文类型：期刊论文

【摘要】：为了克服现有方法在空气质量预测上存在的缺点,文中通过采用改进的离散型人工鱼群算法,并结合分形维数,提出基于人工鱼群和分形维数融合SVM的空气质量预测方法.首先对人工鱼群算法聚群、觅食行为及移动方式进行离散化改进,引入跳出局部最优策略和并行机制.然后将改进的离散型人工鱼群算法结合分形维数,约简空气质量数据集.最后采用基于高斯核SVM建立空气质量预测模型.在北京、上海和广州近2年的空气质量数据上的实验表明,文中方法预测性能较优,具有较高的稳定性和可信性.
[Abstract]:In order to overcome the shortcomings of the existing methods in air quality prediction, an improved discrete artificial fish swarm algorithm is adopted in this paper, combined with fractal dimension. An air quality prediction method based on artificial fish swarm and fractal dimension fusion SVM is proposed. Firstly, the algorithm of artificial fish swarm clustering, foraging behavior and moving mode are discretized. Then the improved discrete artificial fish swarm algorithm is combined with fractal dimension. Finally, the air quality prediction model based on Gao Si nuclear SVM is established. The experiments on the air quality data of Beijing, Shanghai and Guangzhou over the past two years show that the performance of the proposed method is better. It has high stability and credibility.
【作者单位】：合肥工业大学管理学院;合肥工业大学过程优化与智能决策教育部重点实验室;
【基金】：国家863云制造主题项目(No.2015AA042101) 国家自然科学基金重大研究计划培育项目(No.91546108);国家自然科学基金重点项目(No.71490725);国家自然科学基金项目(No.71271071);国家自然科学基金青年基金项目(No.71301041)资助~~
【分类号】：X51;TP18
【正文快照】： NI Zhiwei,ZHU Xuhui,CHENG Meiying(School of Management,Hefei University of Technology,Hefei 230009)(Key Laboratory of Process Optimization and Intelligent Decision-Making of Ministry of Education,Hefei University of Technology,Hefei 230009)relatively hig

【相似文献】