燃煤机组智慧燃烧优化控制研究及应用

Research and Application of Intelligence Combustion Optimization System in a Coal-fired Boiler

锅炉燃烧是一个多变量、强耦合的非线性物理化学过程,随着入炉煤质、环境温度、设备老化等边界条件的变化,传统燃烧优化调整无法满足锅炉的最优工况常态化运行。为此,本文以五彩湾北二电厂为研究对象,基于建立的神经网络与函数混合的特性响应模型,综合考虑锅炉安全性、锅炉效率、NOX生成量等,建立锅炉输入参量与控制目标之间映射关系。在此基础上,通过遗传算法找出目标参数最优值下氧量给定、各层二次风门开度、燃尽风门开度和燃烧器摆角等可调输入参数的最佳组合。通过修改锅炉燃烧侧主要热工逻辑的组态及控制逻辑参数的整定,实现锅炉燃烧系统相关参数的闭环优化控制,降低了炉膛出口烟气温度,保障机组不发生大范围结焦,在满足机组安全运行和污染物排放需求的前提下,提高了锅炉效率。

Boiler combustion is a multivariable and strongly coupled nonlinear physicochemical process.With changes in boundary conditions such as coal quality,environmental temperature,and equipment aging,traditional combustion optimization adjustments cannot meet the optimal operating conditions of the boiler.Therefore,this article takes Wucaiwan North Second Power Plant as the research object,based on the established neural network and function mixed characteristic response model,comprehensively considers boiler safety,boiler efficiency,NOx generation,etc.,and establishes a mapping relationship between boiler input parameters and control objectives.On this basis,genetic algorithm is used to find the optimal combination of adjustable input parameters such as oxygen content,secondary air door opening of each layer,exhaust air door opening,and burner swing angle under the optimal target parameter value.By modifying the configuration of the main thermal logic on the combustion side of the boiler and adjusting the control logic parameters,closed-loop optimization control of the relevant parameters of the boiler combustion system is achieved,reducing the flue gas temperature at the furnace outlet,ensuring that the unit does not experience large-scale coking,and improving the boiler efficiency while meeting the requirements of safe operation and pollutant discharge of the unit.