基于SSA-XGBoost的火电厂引风机故障预警研究  

Study on Fault Warning of Induced Draft Fan Based on SSA-XGBoost in Thermal Power Plant

作　　者：朱越 ZHU Yue

机构地区：[1]国网能源哈密煤电有限公司,新疆哈密839000

出　　处：《青海电力》2022年第S01期37-41,49,共6页Qinghai Electric Power

摘　　要：为了提高火电厂引风机故障预警能力,提出了一种基于麻雀算法(SSA)优化XGBoost(eXtreme Gradient Boosting)的故障预警方法。针对XGBoost模型超参数优化的问题,引入SSA算法对XGBoost算法的超参数进行优化得到预警模型的最佳参数。实例表明:SSA优化的XGBoost预警模型可以准确、高效地对引风机进行故障预警,与XGBoost模型和支持向量机(SVM)模型相比,SSA-XGBoost模型精度和效率更高,且具有较强的泛化能力。In order to improve the fault warning ability of induced draft fans in thermal power plants,a fault warning method optimized XGBoost(eXtreme Gradient Boosting)based on Sparrow algorithm(SSA)was proposed.In order to optimize the hyperparameters of XGBoost model,SSA algorithm was introduced to optimize the hyperparameters of XGBoost algorithm to get the best parameters of the early warning model.The example shows that the SSA-optimized XGBoost early warning model can accurately and efficiently carry out fault early warning for induced draft fan.Compared with the XGBoost model and support vector machine(SVM)model,the SSA-XGBoost model has higher accuracy and efficiency,and has strong generalization ability.

关 键 词：引风机 故障预警 XGBoost 麻雀优化算法 

分 类 号：TP277[自动化与计算机技术—检测技术与自动化装置]