摘要
针对混合核函数支持向量机(SVM)在建模中的重要参数值选择问题,提出利用具有较强全局搜索能力的改进粒子群优化算法,对混合核函数SVM建模过程中的重要参数进行优化调整,给出应用该方法的具体步骤,通过仿真实验验证该算法的有效性。该方法用于谷氨酸发酵过程的建模研究,取得了较高建模精度。
In order to overcome the disadvantage that it is difficult to get better parameter value in modeling of Support Vector Machine with mixture kernels,a new method is put forward to find the better parameter value by using improved particle swarm optimization which has better global search ability. The concrete step of the method was shown. The simulated experiment proved the effectiveness of the above algorithm. The method for the research of modeling in the Glutamic acid fermentation process obtains higher accuracy.
出处
《江南大学学报（自然科学版）》
CAS
2009年第6期631-634,共4页
Joural of Jiangnan University (Natural Science Edition)
基金
国家863计划项目(2006AA020301)
关键词
建模
混合核函数
支持向量机
粒子群优化
modeling, mixture kernels, support vector machine, particle swarm optimization
