SOFC的支持向量机(SVM)辨识建模

Study of Support Vector Machine Model for SOFC

为了便于固体氧化物燃料电池(SOFC)的性能预测和控制方案设计,提出一种基于支持向量机(SVM)的建模方法,用具有RBF核函数的SVM建立了SOFC电池堆的非线性模型。应用仿真对所建SVM模型的有效性和精度进行了检验,并与BPNN模型的辨识效果进行了比较。仿真结果证明与BPNN模型相比,SVM模型具有较高的建模精度。该SVM辨识模型的建立,对SOFC系统的控制策略研究具有一定的实用价值。

For the purpose of perforrnance prediction and control scheme design, a nonlinear model based on support vector machine （SVM） was proposed for a family of complex systems with strong nonlinearity such as solid oxide fuel cell （SOFC）. The nonlinear model was built by SFM with RBF kernel The validity and accuracy of the SVM model were tested by simulation. At the same time, the simulation result comparisons between the SFM model and the BPNN model demonstrate that the SVM model is superior to the BPNN in modeling the nonlinear properties of the SOFC. It has applied value to establish this SVM model for studying the control strategies of SOFC system.