摘要
支持向量机是一种非常有前景的学习机器,它的回归算法已经成功地用于解决非线性函数的逼近问题.但是,SVM参数的选择大多数是凭经验选取,这种方法依赖于使用者的水平,这样不仅不能获得最佳的函数逼近效果,而且采用人工的方法选择SVM参数比较浪费时间,这在很大程度上限制了它的应用.为了能够自动地获得最佳的SVM参数,提出了基于自适应遗传算法的SVM参数选取方法.该方法根据适应度值自动调整交叉概率和变异概率,减少了遗传算法的收敛时间并且提高了遗传算法的精度,从而确保了SVM参数选择的准确性.将该方法应用于船用锅炉汽包水位系统建模,仿真结果表明由该方法所得的SVM具有较简单的结构和较好的泛化能力,仿真精度高,具有一定的理论推广意义.
The support vector machine (SVM) is a promising artificial intelligence technique, in which the regression algorithm has already been used to solve the nonlinear function approach successfully. Unfortunately, most users select parameters for an SVM by rule of thumb, so they frequently fail to generate the optimal approaching effect for the function. This has restricted effective use of SVM to a great degree. In order to get optimal parameters automatically, a new approach based on an adaptive genetic algorithm (AGA) is presented, which automatically adjusts the parameters for SVM. This method selects crossover probability and mutation probability according to the fitness values of the object function, therefore reduces the convergence time and improves the precision of GA, insuring the accuracy of parameter selection. This method was applied to modeling of water level system of a ship boiler's drum. Simulation results reveal that this method of parameter selection for SVM has a simple structure and good generalization ability. Since simulation precision is high, this method possesses certain practical application significance.
出处
《哈尔滨工程大学学报》
EI
CAS
CSCD
北大核心
2007年第4期398-402,共5页
Journal of Harbin Engineering University
基金
黑龙江省自然科学基金资助项目(2004-19)
关键词
机器学习
支持向量机
支持向量机回归
自适应遗传算法
非线性系统辨识
machine learning
support vector machine
support vector machine regression
adaptive genetic algorithm
nonlinear system identification