循环流化床床温的多变量建模

Multi-variable Modeling of Circulating Fluidized Bed Temperature

针对循环流化床锅炉床温对象,分析一次风对其的影响,提出了一种逆向响应模型结构,结合给煤量对床温的影响,建立了包含惯性权重因子的两入一出多变量模型。对现场历史数据的筛选和预处理,采用粒子群优化算法辨识多模型结构中的待定参数及惯性权重因子,并通过另外三段不同负荷段的历史数据完成模型验证。最后将多变量模型和单结构模型进行对比,多变量模型精度更高,适用性更广,为实现循环流化床锅炉床温的优化控制奠定了基础。

This paper analyzes the influence of primary air on the bed temperature of CFB boiler,and presents a reverse response model.A two-input-one-output multivariable model with inertia weight factor is established.Particle swarm optimization(PSO)algorithm was used to identify the undetermined parameters and inertia weight factors in the multi-model structure,and the model verification was completed through the historical data of the other three different load segments.Finally,the multi-variable model is compared with the single-structure model,and the multi-variable model has higher accuracy and wider applicability,which lays a foundation for the optimal control of bed temperature of CFB boiler.