基于层析成像软测量的两相流流型识别  

Ect-based Soft-sensing Technique in Flow Regime Identification of Two-phase Flow

作　　者：张立峰[1] 

机构地区：[1]华北电力大学控制科学与工程学院,河北保定071003

出　　处：《电力科学与工程》2007年第3期6-8,16,共4页Electric Power Science and Engineering

基　　金：国家自然科学基金资助项目(60301008;50337020);天津自然科学基金资助项目(013614411.)

摘　　要：基于电容层析成像(ECT)和人工神经网络的软测量方法,实现了两相流流型识别。以油气两相流为例,建立了两相流流型识别的软测量模型。从ECT传感器的输出中提取特征参数作为软测量模型的辅助变量,两相流流型为主导变量,构建二级自组织竞争神经网络,进而实现对两相流流型的在线判别。仿真结果表明,该方法判别精度高、判别速度快。Taking the oil-gas two-phase flow as an example, this paper investigated the identification of flow regimes using soft-sensing technique, which is based on Electrical capacitance tomography （ECT） and artificial neural network. An online soft-sensing model was built. The character parameters extracted from ECT sensor outputs were the second variable. The primary variable was flow regime of oil-gas two-phase flow. Then, a two- layer self-organizing competitive neural network was built. Simulation results showed that the proposed method has good identification precision and fast identification speed, which means it is an effective tool in two-phase flow pattern online identification.

关 键 词：电容层析成像 软测量技术 流型识别 自组织竞争神经网络 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]