针对超临界燃煤锅炉调峰运行引起的过热汽温波动大、控制难度增加的问题,提出了基于中间点焓值预测的超临界燃煤锅炉过热蒸汽温度控制优化策略。首先,以历史运行数据为驱动,结合灰狼优化与极端梯度提升组合算法构建中间点焓值的预测模型；接着,采用自回归模型辨识给水流量扰动试验数据,获得不同工况下的给水流量动态响应特性与阶跃响应序列；然后,采用模糊加权方法设计多模型动态矩阵控制器,将中间点焓值的预测值作为前馈信号引入来优化过热汽温控制策略；最后,以某在役600MW超临界锅炉为研究对象进行分析,结果表明,提出的中间点焓值的预测模型误差在(?1%, 1%)之间,基于中间点焓值的预测的过热蒸汽温度控制优化策略可使过热汽温动态偏差控制在±5 ℃以内,提高了过热汽温的控制品质。

One of the pressing concerns for the supercritical coal-fired power plant is the challenge of controlling the variation in superheated steam temperature during peak shaving operations. This paper presents a new approach for optimizing the control of superheated steam temperature through the prediction of enthalpy at the intermediate point. Firstly, a hybrid model based on Gray Wolf Optimization and Extreme Gradient Boosting is proposed to predict the enthalpy at the intermediate point. Secondly, autoregressive exogenous is employed to discriminate the dynamic characteristics of feed-water control system. Subsequently, a dynamic matrix control strategy is designed employing fuzzy control algorithm, incorporating the predicted enthalpy into the superheat steam control system as a feed-forward signal. Finally, the proposed control strategy is applied to an on-duty 600 MW super-critical boiler. Results show that the relative error of the proposed enthalpy predicted model is within (-1%, 1%). The optimization control strategy designed for superheated steam temperature ensures that the dynamic deviation of superheated steam temperature remains under ±5℃, suggesting that the strategy can enhance the control quality of the superheated steam.

山西省科技重大专项（202201090301001); 国网山西省电力公司科技项目（520530230026）