子空间预测控制在固体氧化物燃料电池中的应用

Application of Subspace Predictive Control to Solid Oxide Fuel Cell

为使固体氧化物燃料电池(SOFC)输出满足一定约束的稳定电压、提高其输出功率跟踪性能,采用文献中给出的SOFC动态模型,设计了基于子空间矩阵的预测控制器。将子空间辨识和预测控制算法结合起来,直接从开环输入输出数据获得预测控制器的控制矩阵,而不须经过中间参数模型辨识的步骤。仿真结果验证了所提子空间预测控制算法(SPC)的实用性。

For enabling the SOFC to generate stable output voltage which satisfies certain constraint, and improving the tracking performance of the SOFC output power, the SOFC dynamic model in the existing literature is adopted and a predictive controller based on the subspace matrix is designed. It combines the subspace identification method with predictive control algorithm, and obtains the control matrices in predictive controller from the open-loop input/output data. Thus, it does not need any step to identify the intermediate parametric model. A simulation result shows practicality of the proposed subspace predictive control algorithm（SPC）.