论文部分内容阅读
针对工业烧结炉具有大时滞、非线性的问题,提出了一种基于RBF神经网络的动态矩阵预测控制方法。该方法将RBF神经网络与动态矩阵预测控制相结合,既保留了常规预测控制的优点,又克服了复杂对象难以精确建模的缺点,有效地解决了不确定时滞对烧结炉温度控制性能的影响。仿真结果表明,该方法比常规预测控制有更好的控制效果。
Aiming at the problem of large time delay and nonlinearity in industrial sintering furnace, a dynamic matrix predictive control method based on RBF neural network is proposed. This method combines RBF neural network and dynamic matrix predictive control, which not only retains the advantages of conventional predictive control, but also overcomes the shortcomings of complex objects that are difficult to accurately model. It effectively solves the problem of uncertain temperature delay on sintering furnace temperature control performance influences. Simulation results show that this method has better control effect than conventional predictive control.