基于压缩空气流量预测的火电机组空压机节能研究  

Thermal power unit air compressor energy saving research based on compressed air flow prediction

作　　者：王冠华 孙宇贞 彭道刚 汪皓然 WANG Guanhua;SUN Yuzhen;PENG Daogang;WANG Haoran(College of Automation Engineering,Shanghai University of Electric Power,Shanghai 200090,China)

机构地区：[1]上海电力大学自动化工程学院,上海200090

出　　处：《流体机械》2024年第1期96-104,共9页Fluid Machinery

基　　金：上海市科技创新行动计划高新技术领域项目(22511103800)。

摘　　要：针对火电机组内空压机布局分散和加卸载频繁造成电能浪费严重的问题,提出一种基于压缩空气流量预测结果进行空压机联网运行组合优化的方法。为提高预测模型的预测精度,提出鹈鹕优化算法(POA)与长短期记忆神经网络相结合的预测模型(POA-LSTM),通过相关性分析,选取空压机运行电流、储气罐压力、空压机排气压力和排气温度作为压缩空气流量的影响因素。结果表明,该模型预测误差率小于4%,能较准确地预测压缩空气流量的变化趋势;根据用气量预测结果进行杂用压缩空气联网改造测试,优化空压机的组合运行方式,可以少开杂用空压机1台,能耗降低13.01%;全厂空压机均采用基于压缩空气流量预测的联网供气方式可实现全年节约电能超过440万kW·h,节省电费268万余元,减少2 559 t二氧化碳排放。To address the problems of scattered air compressor layout and frequent loading and unloading in thermal power units,a method is proposed to optimize the combination of air compressor network operation based on compressed air flow prediction results.In order to improve the prediction accuracy of the prediction model,a prediction model(POA-LSTM)combining Pelican Optimization Algorithm(POA)and Long Short-Term Memory Neural Network is proposed.Through correlation analysis,the air compressor operating current,storage tank pressure,air compressor exhaust pressure and air compressor exhaust temperature are selected as the influencing factors of compressed air flow,and the prediction results show that the prediction error rate of the model is less than 4%,which can more The prediction results show that the prediction error rate of the model is less than 4%,which can predict the change trend of compressed air flow more accurately.Based on the prediction results of air consumption,we carried out the renovation test of miscellaneous compressed air network and optimized the combination operation of air compressors,which can reduce one miscellaneous air compressor and reduce energy consumption by 13.01%.The networked air supply method based on compressed air flow prediction can save more than 4.4 million kilowatt hours of electricity,2.68 million yuan of electricity cost and 2559 tons of carbon dioxide emission for the whole plant.

关 键 词：压缩空气流量预测 LSTM 鹈鹕优化算法 空压机节能 

分 类 号：TH138[机械工程—机械制造及自动化]