锅炉燃烧系统的自学习模糊控制

The Autonomic Learning Fuzzy Control of the Combustion System of a Boiler

由于燃烧组织的不合理,以及操作、管理水平的偏低,很多工业锅炉存在着效率偏低以及污染严重的问题。本文设计了一套锅炉燃烧系统自学习模糊控制系统。该系统从对象的数据样本通过基于神经网络的模糊辨识获得虚拟对象模型,然后在虚拟对象上对虚拟模糊控制器进行训练,并将经过训练优化后的参数和控制策略应用于实际控制器。同时,本文提出了一种变论域方法,极大提高了模糊控制器的稳态精度。仿真和现场试验表明,该系统具有良好的控制效果,明显提高了锅炉的效率。

Because of the irrationality of the combustion organization and low level of operation and supervision, many industrial boilers have the problems of low efficiency and high pollution. It introduced an autonomic learning fuzzy control system of the combustion system of a boiler. From the data samples of the object, the system used a fuzzy identification method based on neural networks to acquire a dummy model of the object, then trained a dummy fuzzy controller on the dummy model, and finally applied the optimized parameters and control strategies to the practical controller. At the same time, an adaptive discourse of universe idea was put forwarder, which greatly increased the steady precision. Simulations and experiments showed that the system had an excellent control effect and obviously increased the efficiency of the boiler.