论文部分内容阅读
针对支持向量机中的参数通常用交叉验证来确定的状况,提出了遗传支持向量机算法,即使用遗传算法来优化支持向量机中的参数并应用在基于火焰图像特征参数的锅炉燃烧状态诊断中。从火焰图像中提取的5个特征量作为支持向量机的输入,3种燃烧状态作为输出,选用径向基核函数,使用遗传算法得到优化参数。实验结果表明,该方法能在较大范围内准确地找到相应的优化参数,并能有效地进行锅炉燃烧状态诊断。
Aiming at the situation that the parameters in SVM are usually determined by cross-validation, a genetic support vector machine (SVM) algorithm is proposed, which uses genetic algorithm to optimize the parameters in SVM and apply it to diagnose the combustion state of boiler based on flame image characteristic parameters . The five features extracted from the flame image are used as inputs to the support vector machine, and the three combustion states are taken as output. The radial basis function is chosen and the genetic algorithm is used to get the optimal parameters. The experimental results show that the proposed method can accurately find the corresponding optimization parameters in a wide range and can effectively diagnose the combustion state of the boiler.