论文部分内容阅读
为了提高软测量模型的泛化能力,提出一种基于AdaBoosting算法的组合支持向量机(SVM)模型.该方法在贝叶斯分析的基础上,利用样本概率初始化惩罚系数,依据回归过程中的损失函数更新惩罚系数权重,使得SVM训练模型有强、弱之分,突出一些重要样本的作用,以提高模型的估计精度和泛化能力.仿真结果表明,依据该方法建立的组合模型明显改善了软测量模型的估计能力和泛化能力.
In order to improve the generalization ability of soft-sensing model, a combined support vector machine (SVM) model based on AdaBoosting algorithm is proposed in this paper.Based on the Bayesian analysis, the penalty coefficient is initialized by the sample probability and according to the loss in the regression process The function updates the weights of the penalty coefficients so that the SVM training model has strong and weak points, highlighting the importance of some important samples in order to improve the estimation accuracy and generalization ability of the model. Simulation results show that the combination model established by this method significantly improves the soft Measuring model’s estimation ability and generalization ability.