锅炉燃烧系统的建模研究

Research of the Modeling of Boiler Combustion System

针对传统BP算法存在易陷入局部极值点、收敛速度慢、泛化能力差等缺点,运用一种由遗传算法和BP神经网络相结合的方法对锅炉燃烧系统进行建模分析。通过与单纯用BP算法建模方法比较,这种方法能很好地克服用BP神经网络建模时的诸多缺点。模型已经由现场采集的锅炉数据验证,能够得到理想的效果.

As the traditional BP algorithm is easy to fall into local extreme point, slow con- vergence and poor generalization ability shortcomings, this paper models and analyzes the boiler combustion system by using BP neural network and genetic algorithm combination method. By comparison with simple BP algorithm modeling method, it is found that this method can overcome the shortcomings of the model well, which uses BP neural network to model. Models have been validated by scene boiler data and can obtain the ideal effect.