RBF神经网络用于系统的预测最优控制

Application of RBF-Neural Network Combined with the Optimal StableSolution for the Predicative Optimum Control of System

为了实现系统的预测最优化控制，问题的关键是如何准确而迅速地对未来一段时间内的系统状态进行预测，然后利用此预测结果及优化指标来控制有关系统变量，人们对此进行了许多研究，但尚有其不足之处。为此，提出压缩预测变量集规模等办法来增加ＲＢＦ神经网络有效预测时间长度，在此基础上利用稳态最优解和优化指标来控制有关量。此方法被用于某化工过程，结果能使系统运行更为平稳，并使有关量达到了预测的优化指标。

To realize the predicative optimum control of a system, the key of the problem is how to predicate the states of the system in the coming time space accurately and quickly and then use the predicated results and the optimal indexes to control the related system variables. A lots of study works in this aspect have been done. But there are still many short-comings. This paper proposed a method by which the scale of the predicted variables set could be reduced. By combining this method with other methods, the valid predication time length of the RBF-neural network was increased. Based on this, the optimum stable state of the system and the optimum indexes were used to control related variables. The methods presented in this paper have been applied to the simulation of a chemical process. The results showed that the system could reach the indexes as well as run in a stabler state.