一种模糊神经网络自适应预测控制方案的研究

在过程控制中，由于被控对象常具有非线性、不确定性及参数时变等复杂因素，难以建立精确的数学模型，从而直接影响了控制效果，提出了一种模糊神经网络自适应预测控制议案，对学习公式进行了理论指导，并结合误差补偿以提高预测控制的精度，仿真实验表明，该算法可实现模糊控制和神经网络的优势互补，对非线性复杂系统具有良好的控制性能。

INVESTIGATION ON AN ADAPTIVE PREDICTIVE CONTROL PROGRAM BASED ON A FUZZY NEURAL NETWORK

In the process control,due to some complex factors of the controlled object with nonlinearity,uncertainty and time varying parameters,it is difficult to construct an accurate mathematical modal,and therefore the control effect is affected.In this paper an adaptive predictive control algorithm based on fuzzy neural network has been developed,and a new learning algorithm was derived theoretically.Based on the error compensation method,the accuracy of the algorithm was improved.Experiment results show that the algorithm can obtain the advantages of the fuzzy system and the neural network can have better performance in controlling the nonlinear and complex system.