基于回归分析法的离心式压缩机性能模型

A centrifugal compressor performance model based on regression analysis method

建立准确的压缩机性能曲线模型,是天然气长输管网优化运行的关键。受风机定律启发,分析了风机中流量、转速以及多变能量头的关系,将其应用于天然气长输管道压气站场的离心式压缩机,建立了压缩机特性曲线模型。假设一种线性化的压缩机特性曲线关系,采用偏最小二乘法对该特性曲线关系进行回归分析,得出压缩机特性曲线回归模型,并利用多年积累的现场数据将其应用于压气站工艺气流量虚拟测量。研究结果表明:该模型计算准确度较高,工艺气虚拟测量结果误差较小,能够较好地反映离心式压缩机组实际运行特性,为管网优化运行及实时掌握机组运行情况提供了技术保障。

The key to the operation optimization of natural gas pipeline network is to establish an accurate compressor performance curve model. Inspired by the law of fan, the relationships among flow rate, rotation speed and polytropic energy head of fans were analyzed and then applied to the centrifugal compressors in the compressor station of long-distance gas pipeline to establish the compressor performance curve model. A lincarized compressor performance curve relationship was assumed and then performed regression analysis by means of the partial least square method to obtain the regression model of compressor performance curve. And based on years＇ field data, this model was applied to the virtual measurement of process gas flow in the compressor station. It is shown that the compressor performance curve model based on regression analysis method is higher in the calculation accuracy and lower in the error of virtual measurement results of process gas, and can reflect better the actual operating characteristics of centrifugal compressor sets. It provides the technical guarantee for the operation optimization of pipeline network and the on-site understanding on the operation situations of compressor sets. （5 Figures, 4 Tables, 20 References）