基于Elman网络补偿模型的Smith预测控制

研究基于Elman网络补偿模型的Smith预测控制问题.采用互补建模方法对被控对象进行建模,其中机理模型反映被控对象的主要工作规律和运行趋势,但不可避免地存在一定的模型误差;通过Elman网络拟合机理模型的模拟误差,并对其进行补偿.实验结果表明,基于Elman网络补偿模型的Smith预测控制充分利用了神经网络的非线性拟合能力,只要对纯滞后环节精确建模,就可以完全抵消该环节对控制品质及系统稳定性的不利影响.该方法使得Smith预测控制可以用于模型不易精确确定的系统.

Smith Predictive Control Based on Elman Network Compensatory Model

Smith predictive control based on the Elman network compensatory model is studied. A mutually compensatory modeling method is employed, in which the mechanism model simulates the main performance of the controlled process. As the mechanism model inevitably involves modeling error more or less, an Elman network can be used to model the modeling error of the mechanism model, and to compensate it. The simulation is performed following these ideas. The results prove that the Smith predictive control algorithm based on Elman network compensatory model takes good advantage of the nonlinear modeling capability of the neural network, and that the harm from the time delay to the performance and stability of the system can be counteracted completely if only the time delay is precisely known. Accordingly, with the help of the Elman network compensatory model, Smith predictor can be advanced to control the system whose mathematical model is difficult to determine precisely.