基于深度长短记忆网络的汽轮机数据清洗  

Turbine data cleaning based on deep LSTM

作　　者：许小刚[1,2,3] 王志香 王惠杰 XU Xiaogang;WANG Zhixiang;WANG Huijie(Department of Power Engineering,North China Electric Power University,Baoding 071003,China;Key Laboratory of Low Carbon and Efficient Power Generation Technology of Hebei,North China Electric Power University,Baoding 071003,China;Baoding Key Laboratory of Low Carbon and Efficient Power Generation Technology,North China Electric Power University,Baoding 071003,China)

机构地区：[1]华北电力大学动力工程系,河北保定071003 [2]华北电力大学河北省低碳高效发电技术重点实验室,河北保定071003 [3]华北电力大学保定市低碳高效发电技术重点实验室,河北保定071003

出　　处：《热力发电》2023年第8期179-187,共9页Thermal Power Generation

基　　金：中央高校基本科研业务费专项资金资助(2019MS094)。

摘　　要：汽轮机运行过程会产生多样且大量数据。为适应大数据驱动及仿真建模对高质量数据的要求,高效的数据清洗十分必要。利用长短记忆层对于时序数据出色的非线性拟合能力搭建了汽轮机半监督数据清洗模型。模型选取机组的3个边界条件作为输入,对待清洗数据进行预测,根据预测值与实际值的残差完成异常值剔除,之后选用模型的预测值进行数据填充,保证数据的完整性。利用模型对某电厂650 MW机组进行数据清洗,并且为克服样本失衡给清洗模型指标选取带来的问题,对准确率进行了改进并将其作为清洗效果的衡量指标。结果表明:深度长短记忆网络的数据清洗模型改进准确率高于其他3种常见清洗方法,可有效识别数据是否异常,且可利用预测值进行数据填充,保证清洗前后数据量一致。A large amount of data is generated during steam turbine operation.In order to meet the requirements of high quality data driven by big data and simulation modeling,efficient data cleaning is very necessary.The semi-supervised data cleaning model of steam turbine is built by using the excellent nonlinear fitting ability of long and short memory layer for time series data.The model selects three boundary conditions of the unit as input to predict the cleaning data.Outliers are eliminated according to the residual difference between the predicted value and the actual value.After that,the predicted value of the model is used to fill the data to ensure the integrity of the data.The model is used to clean the data of a 650 MW unit in a power plant.In order to overcome the problems caused by sample imbalance in the selection of cleaning model indicators,the accuracy rate is improved and taken as the measurement index of cleaning effect.The results show that the improved accuracy of the data cleaning model of the deep long and short memory network is higher than that of the other three common cleaning methods,which can effectively identify whether the data is abnormal,and can use the predicted value to fill the data to ensure the consistency of data before and after cleaning.

关 键 词：长短记忆网络 深度学习 数据清洗 异常值 汽轮机 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] TM621[自动化与计算机技术—控制科学与工程]