基于ABC-SVM模型的固体氧化物燃料电池预测控制仿真研究

为了满足直流负载电压的稳定性及固体氧化物燃料电池（SOFC）的耐用性和安全性的要求，设计了两个控制回路分别对SOFC的输出电压和燃料利用率进行控制。通过设计一个简单的控制回路来使燃料利用率保持在恒定值，并在此基础上开发了一个非线性模型预测控制器以控制SOFC的输出电压。该非线性模型预测控制器基于改进的支持向量机（SVM）预测模型，首先利用Lipschitz quotients准则确定SVM预测模型的结构，然后通过人工蜂群算法（ABC）优化SVM参数。仿真结果表明，所提的基于ABC-SVM模型的SOFC预测控制算法可以很好地跟踪电压设定值，证明了ABC-SVM模型在SOFC非线性动态建模中的有效性。

Predictive control simulation of solid oxide fuel cells based on an artificial bee colony-support vector machine (ABC-SVM) model

In order to meet the requirements of stable DC load voltage, durability and safety of solid oxide fuel cells (SOFCs), we have designed two control loops to control the output voltage and fuel utilization of SOFCs. A simple control loop is designed to maintain the fuel utilization at a constant value, and based on this, a nonlinear model predictive controller is developed in order to control the output voltage of the SOFC. This nonlinear model predictive controller is based on a modified support vector machine (SVM) predictive model. First, the SVM prediction model structure is determined using the Lipschitz quotients criterion, and then the SVM parameters are optimized by the artificial bee colony (ABC) algorithm. Simulation results show that the SOFC predictive control algorithm based on the ABC-SVM model proposed in this paper can effectively track the voltage setpoint, demonstrating the effectiveness of ABC-SVM in modeling the nonlinear dynamics of SOFCs.