摘要
本文指出超临界CFB机组给水控制系统因其非线性强、耦合度高且滞后大等特点,使得给水系统给水流量的控制难度大,控制效果并不十分理想。针对这一问题,为了对给水流量进行有效的预测,本文采集并分析研究了某电厂CFB机组的现场运行数据,选取了机组运行时影响给水流量的主要变量,建立了基于随机森林算法的给水流量预测模型。预测结果显示:该预测模型的预测效果较好,误差较小,可为现场控制系统提供参考,有助于提升给水系统的控制品质和给水系统给水流量的控制效果,具有一定的实用意义。
This paper points out that the feedwater control system of supercritical CFB unit is difficult to control the feedwater flow of the feedwater system because of its strong nonlinearity,high coupling and large lag,and the control effect is not very ideal.To solve this problem,in order to effectively predict the feedwater flow,this paper collected and analyzed the on-site operation data of a CFB unit in a power plant,selected the main variables that affect the feedwater flow during the unit operation,and established the feedwater flow prediction model based on random forest algorithm.The prediction results show that the prediction effect of the prediction model is good and the error is small,which can provide reference for the on-site control system,and help to improve the control quality of the water supply system and the control effect of the water supply flow of the water supply system,and has certain practical significance.
作者
韩丹
HAN Dan(National Energy Group Material Co.,Ltd.,Beijing 100070,China)
出处
《科技创新与生产力》
2023年第2期124-126,共3页
Sci-tech Innovation and Productivity
