介绍了一种模糊支持向量机控制器,它将模糊控制与支持向量机相结合,融合了两者的优点,既不依赖被控对象模型又有泛化能力强等特点。利用遗传算法优化控制器的参数,以使其控制性能能适应对象的变化而达到最优。火电厂主汽温控制的仿真结果表明了这种控制器具有良好的控制性能。 A fuzzy support vector machine controller is introduced, which combines fuzzy control with support vector machine, and combines the advantages of both. It not only depends on the controlled object model but also has the ability of generalization. The genetic algorithm is used to optimize the parameters of the controller so that its control performance can be adapted to the change of the object to achieve the optimal result. Simulation results of main steam temperature control in thermal power station show that the controller has good control performance.