基于BP神经网络算法的压缩机组运行优化模型
发布时间：2018-11-07 13:27
【摘要】：增压站运行方案制定的难点在于如何根据下游耗气量的变化,在不超出压缩机最大功率参数的情况下精准快速地调整进站压力,并根据具体需求提前制定多机组联合运行方案。以大牛地气田塔榆增压站6RDSA-1型压缩机组为研究对象,采用BP神经网络算法建立了压缩机组运行优化模型。选择已有的压缩机进气温度、排气压力及排气流量这3个基本参数作为模型输入值,计算得到了合适的进气压力和机组的轴功率。通过不同工况多组数据对比,模型对进气压力的预测结果与现场实测值的相对误差小于2.75%,验证了基于BP神经网络算法的压缩机组运行优化模型的可靠性,有助于增压站提前制定多机组联机运行方案,提升机组的运行效率,降低能耗和运维成本。
[Abstract]:The difficulty of working out the operation plan of supercharging station is how to adjust the inlet pressure accurately and quickly according to the change of downstream gas consumption and make the combined operation plan of multi-unit in advance according to the specific demand. Taking the 6RDSA-1 type compressor unit in Tayu supercharging station of Daniudi gas field as the research object, the operation optimization model of the compressor unit is established by using the BP neural network algorithm. Three basic parameters of compressor intake temperature exhaust pressure and exhaust flow are selected as the input values of the model and the proper intake pressure and the shaft power of the unit are calculated. Through the comparison of different working conditions and many groups of data, the relative error between the prediction result of the model and the field measured value is less than 2.75, which verifies the reliability of the optimized model of compressor unit operation based on BP neural network algorithm. It is helpful for the supercharging station to draw up the on-line operation plan of the multi-unit in advance, to improve the operation efficiency of the unit, and to reduce the energy consumption and the operation and maintenance cost.
【作者单位】： 长江大学机械工程学院;中石化石油机械股份有限公司压缩机分公司;
【基金】：国家科技重大专项“钻井协同自动控制系统研发”,2016ZX05022006-004 工业和信息化部科研项目“海洋大功率往复式压缩机研制”,工信部联装[2014]506号 湖北省技术创新专项“深部地热资源综合开发利用关键技术”,2016ACA181 长江大学地热资源开发研究所开放课题“往复式压缩机故障诊断与运行优化系统研究”,Geo TH2014-04;长江大学青年基金项目“天然气压缩机组节能优化运行软件开发”,2015CQN46
【分类号】：TE974
，
本文编号：2316510
[Abstract]:The difficulty of working out the operation plan of supercharging station is how to adjust the inlet pressure accurately and quickly according to the change of downstream gas consumption and make the combined operation plan of multi-unit in advance according to the specific demand. Taking the 6RDSA-1 type compressor unit in Tayu supercharging station of Daniudi gas field as the research object, the operation optimization model of the compressor unit is established by using the BP neural network algorithm. Three basic parameters of compressor intake temperature exhaust pressure and exhaust flow are selected as the input values of the model and the proper intake pressure and the shaft power of the unit are calculated. Through the comparison of different working conditions and many groups of data, the relative error between the prediction result of the model and the field measured value is less than 2.75, which verifies the reliability of the optimized model of compressor unit operation based on BP neural network algorithm. It is helpful for the supercharging station to draw up the on-line operation plan of the multi-unit in advance, to improve the operation efficiency of the unit, and to reduce the energy consumption and the operation and maintenance cost.
【作者单位】： 长江大学机械工程学院;中石化石油机械股份有限公司压缩机分公司;
【基金】：国家科技重大专项“钻井协同自动控制系统研发”,2016ZX05022006-004 工业和信息化部科研项目“海洋大功率往复式压缩机研制”,工信部联装[2014]506号 湖北省技术创新专项“深部地热资源综合开发利用关键技术”,2016ACA181 长江大学地热资源开发研究所开放课题“往复式压缩机故障诊断与运行优化系统研究”,Geo TH2014-04;长江大学青年基金项目“天然气压缩机组节能优化运行软件开发”,2015CQN46
【分类号】：TE974
，
本文编号：2316510
本文链接：https://www.wllwen.com/kejilunwen/shiyounenyuanlunwen/2316510.html
