基于LS-SVM的SOFC系统自适应逆控制

Adaptive Inverse Control Based on Least Square Support Vector Machines for Solid Oxide Fuel Cell

关于固体氧化物燃料电池(Solid Oxide Fuel Cell,SOFC)性能的优化问题,其中工作温度和电压是关键参数。针对固体氧化物燃料具有较强的非线性且常规成熟线性理论不适用的特点,提出了一种最小二乘支持向量机(LS-SVM)的自适应逆控制策略。首先建立了SOFC的机理模型,然后采用LS-SVM方法建立了SOFC系统的逆动力学模型。在获得逆动力学模型的基础上,设计了一种逆动力学递推最小二乘支持向量机的控制方法。在自适应逆控制下,逆模型通过RLS算法更新,控制器依据ε-滤波进行在线调整。SOFC系统辨识和仿真结果表明,改进方法的可信性,辨识出的逆动力模型具有较高的精度,所设计的控制器能获得较好的控制性能。仿真结果可以为SOFC的实用化和产业化提供一定的理论依据。

The operating temperature and voltage are the key parameters affecting the performance of Solid Oxide Fuel Cell（ SOFC ）. Aiming at the strong nonlinear of SOFC and conventional mature linear theory does not apply, the paper adopted an adaptive inverse control strategy based on least square support vector machine. First, the principle dynamic model of SOFC was, constructed. Then, the LS-SVM method was used to establish the inverse dynamics mod- el of SOFC. Based on the inverse dynamic model acquired, a control algorithm based on recursive least squares sup- port vector machine（RLS-SVM） of inverse dynamics was designed. In this adaptive inverse control mechanism, the inverse dynamic model was updated by RLS algorithm. The parameters of controller were adjusted on-line with e-fil- tering. The simulations of SOFC system identification and control show that the method is credibility, the inverse dynamic model identified has high precision and the designed controller has good control performance. The simulation results can provide certain theoretical basis for the practical and industrialization of SOFC.