基于神经网络误差修正的灰色广义预测控制

提出一种基于神经网络误差修正的多步灰色广义预测控制算法,采用神经网络对灰色系统的建模误差进行预测,同时根据新信息优先的原则,用最新预测误差不断更新神经网络建模数据,对新出现的误差具有动态补偿动力,最终预测值为模型预测值与预测误差之和,有效地抑制模型误差的影响,增强灰色广义预测的鲁棒性.仿真结果验证了该算法的有效性.

Grey Generalized Predictive Control Based on Error Correction Using Neural Network

A long-range grey generalized predictive control algorithm based on error correction using neural network is presented. The neural network error prediction model is introduced according to the past error data. The newest predictive error is used to renew the data of the neural network. The model prediction and predicted error are combined to form the final prediction. This algorithm effectively reduces the influence of the predictive error, and also strengthens the robustness of the grey generalized predictive control. Simulation indicates the effectiveness of the algorithm.