基于多Agent粒子群优化的多步SVR模型预测控制

提出一种基于多Agent粒子群优化支持向量回归机（support vector regression， SVR）参数的优化算法，并利用该算法建立多步预测控制模型，对非线性系统进行预测控制。通过预测控制的机理推导出满足滚动优化目标函数的多步预测输出的控制律。将该模型与基于遗传算法优化的RBF神经网络预测控制器、基于粒子群优化的多步SVR模型预测控制器和基于遗传算法优化的多步SVR模型预测控制器进行比较分析，仿真结果表明该预测控制模型优于其他控制器，具有良好的预测性能，可有效的对非线性系统进行预测控制。

Multi-step model predictive control based on SVR multi-Agent particle swarm optimization algorithm

An optimal selection approach of support vector regression (SVR) parameters is proposed based on the multi-Agent particle swarm optimization (MAPSO) algorithm. A multi-step predictive control model based on the SVR to predict nonlinear systems is established; and the optimal parameters of which is searched by MAPSO. With the objective function of rolling optimization, analytical solutions of multi-step predictive control laws are obtained by the predictive control mechanism. Comparing with the model predictive controllers based on SVR optimized by the particle swarm optimization algorithm (PSO-SVR), SVR optimized genetic algorithm (GA-SVR), and RBF neural network algorithm optimized genetic algorithm (GA-RBF), the simulation results show that the proposed method has better prediction results than others and is effective for nonlinear systems.