摘要
锅炉沾污因数用于定量表征锅炉的积灰程度,在积灰机理尚不明确的前提下,计算锅炉沾污因数存在一定困难。通过对锅炉历史监测数据进行特征重构,研究锅炉状态数据和锅炉沾污因数之间的潜在关系,基于随机森林回归算法实现对锅炉沾污因数进行预测。案例表明,基于重构特征建立随机森林回归模型,可以有效预测锅炉沾污因数,并且能够实时反映锅炉特定受热面的积灰程度。
The pollution factor of the boiler is used to quantitatively characterize the degree of pollution of the boiler.Under the premise that the pollution mechanism of the boiler is not clear,it is difficult to calculate the pollution factor of the boiler.By reconstructing the characteristics of historical monitoring data of the boiler,the potential relationship between the state data and the pollution factor of the boiler was studied,and the pollution factor of the boiler was predicted based on the random forest regression algorithm.The case shows that the random forest regression model established based on the reconstructed characteristics can effectively predict the pollution factor of the boiler,and can reflect the ash accumulation degree of the specific heating surface of the boiler in real time.
出处
《上海电气技术》
2022年第1期29-32,共4页
Journal of Shanghai Electric Technology
关键词
锅炉
沾污因数
随机森林回归算法
预测
Boiler
Pollution Factor
Random Forest Regression Algorithm
Predication
作者简介
第一作者:谢春(1988—),女,硕士,工程师,主要从事数据挖掘与数据可视化工作
