摘要
群体智能是基于生物群体行为规律的智能计算技术,常用以解决参数寻优等问题;作为群体智能的两种典型算法,蚁群算法和粒子群算法应用极为广泛;文章分析了标准蚁群算法和粒子群算法的不足,分别采用改进的蚁群算法和粒子群算法对支持向量机回归模型参数进行优化,并以钕铁硼吸氢阶段合金氢含量预测为例,通过MATLAB对改进后的预测模型进行了仿真验证,最终给出了两种方法优化后,模型的预测效果及性能对比;仿真结果表明,改进的群体智能算法对工艺优化控制有着重要的意义。
Asa intelligent computing technology based on biological laws of group behavior, Swarm intelligence is widely used to solve the problems of parameter optimization. Ant colony algorithm and particle swarm optimization are widely used as two typical algorithms of swarm intelligence. In this paper we did analyze the insignificance of the standard ant colony algorithm and particle swarm optimization, then the improved ant colony algorithm and particle swarm optimization were respectively used to optimize the parameters of the regression model of support vector machine, and the hydrogen content of NdFeB alloy in the hydrogen absorption stage is taken as an example to simulates and verifies the improved model by MATLAB. The contrast of prediction performance of the regression model between two algorithms was given at last. The results of simulation indicate that the improved swarm intelligence algorithm has important significance on optimizing the process control.
出处
《计算机测量与控制》
北大核心
2014年第9期2890-2892,共3页
Computer Measurement &Control
基金
国家自然科学基金(61064001)
关键词
改进蚁群算法
改进粒子群算法
支持向量机回归模型
参数寻优
收敛速度
相对误差
improved ant colony algorithm
improved particle swarm optimization
regress model of support vector machine
parameter optimization
convergent rate
relative error
