基于神经网络和遗传算法的智能控制方法

根据油品调合的特点，应用人工神经网络(Neural Network，NN)对油品在线优化调合
系统进行建模， 并采用量化正交遗传算法(Orthogonal Genetic Algorithm with Quantization，OGA/Q)对神经网络模型进行学习。提出了用于油品在线优化调合系统的人工神经网络适宜的拓扑结构，给出了量化正交遗传算法优化的具体执行过程，实现了系统的主要功能，该系统已经广泛地应用于生产实践中。
关键词：神经网络；在线优化；油品调合；量化正交遗传算法
A Novel Intelligent Control Approach Based on NN and OGA/Q Zhang changnian1, 2
(1 School of Information Engineering, University of Science and Technology Beijing, Beijing, 100083, China) (2 Department of Computer, Guilin College of Aerospace Technology, Guilin 541004, China) Abstract: According to the characteristics of oil blending, develop the on-line optimization of oil blending system using the artificial neural network, and then train the neural network model by the orthogonal genetic algorithm with quantization. The artificial neural network topology for the on-line optimization of oil blending system is proposed in this work. The implementation process of orthogonal genetic algorithm with quantization is introduced in this paper too. The major function of oil blending system is achieved, and this system is wildly applied to the practice.
Keywords: neural network; on-line optimization; oil blend; orthogonal genetic algorithm with
quantization