关于循环流化床锅炉燃烧温度系统的建模控制

Modeling and Control of Combustion Temperature System of Circulating Fluidized Bed Boiler

针对蒙西发电厂2×300MW循环流化床锅炉,实现以排渣量为主控手段的自动控制首先需建立被控对象模型,而机理建模难以运用到实际,传统算法辨识建模易发生收敛速度慢、陷入局部最优等问题导致模型与实际系统过程不匹配。为获得控制系统动态模型,提出基于遗传算法的参数辨识方法,结合运行经验,处理运行数据,建立现场50%负荷与100%负荷两种典型工况下反映给煤量、排渣量与床温、主汽压动态特性的传递函数阵。拟合结果表明,模型误差在可允许范围内,所建立的传递函数模型能够完全表征该工况下系统的动态特性,证明了遗传算法建模的有效性,为后续循环流化床锅炉以排渣量为主控手段的燃烧温度系统控制策略设计和控制器优化提供依据。

Aiming at the 2×300MW circulating fluidized bed boiler in Mengxi Power Plant,it is necessary to set up the controlled object model to realize the automatic control of the main control method. And itis difficult to apply the mechanism modeling to reality. The traditional algorithm to identify the modeling is prone to slow convergence,falling into the local optimal and other issues which lead to the model and the actual system process does not match. In order to obtain the dynamic model of the control system,a parameter identification method based on genetic algorithm was proposed,which combined operational experience,processing operation data,and 50% load and 100% load on the site. The two kinds of typical operating conditions were used to control the system design. The control system was designed to reflect the amount of coal,the amount of slag and the bed temperature,and the main steam pressure dynamic characteristics of the transfer function array. The fitting results show that the model error can be used to characterize the dynamic characteristics of the system under the allowable range. The validity of the algorithm was proved,which provides the basis for the control strategy design and controller optimization of the combustion temperature system for the subsequent circulating fluidized bed boiler with the slag amount as the main control means.