改进蚁群PID-神经元解耦的CFB锅炉燃烧系统控制

床温和主汽压都是循环流化床锅炉生产运行中的重要参数,会直接影响机组的安全性和经济性。但由于这两者对象存在非线性、大时延、强耦合等特点,其现场控制效果一直不太理想。本文首先采用自适应神经元将床温和主汽压解耦,再利用具有分工特征的蚁群算法优化参数的PID控制器对两者进行独立控制。采用该算法优化常规PID控制参数,能够实现控制参数的快速寻优。该方案应用于循环流化床锅炉燃烧系统仿真,结果表明能有效实现系统解耦,且具有响应快、超调量小等优点,有效地提高了控制品质。

The application of neuron decoupling - improved ant colony algorithm PID controller in the control of CFB boiler bed temperature and main steam pressure

Bed temperature and main steam pressure are important parameters in the operation of CFB boiler, which will directly affect the safety and economy of the unit. But as a result of the exist of the characteristics of nonlinear, large time delay and strong coupling between them two, the effect of field control has not been very good. At first, this paper adopts the adaptive neuron decoupling bed temperature and main steam pressure. Then use the PID controller of which parameters is optimized by ant colony algorithm with the division characteristics to control respectively. Using this algorithm optimization of conventional PID control parameters can find out the best parameters rapidly. This scheme was applied to CFB boiler combustion system simulation, the result shows that it can effectively decoupling the system with the advantages of fast response, less overshoot, this improved the quality of control effectively.