SVM based measurement method and implementation of gas-liquid two-phase flow for CMF. (October 2019)

• Record Type: Journal Article
• Authors: Yue, Jing; Xu, Ke-Jun; Liu, Wen; Zhang, Jian-Guo; Fang, Zheng-Yu; Zhang, Lun; Xu, Hao-Ran
• Is Part Of: Measurement. Volume 145(2019)
• Page Start: 160
• Page End: 171
• Publication Date: 2019-10
• Subjects: Coriolis mass flowmeters (CMF) -- Drive efficiency -- Zero-crossing detection algorithm -- Support vector machine (SVM) -- Mass flow measurement error (MFME) model -- Gas-liquid two-phase flow; Weights and measures -- Periodicals; Measurement -- Periodicals; Measurement; Weights and measures; Periodicals; 530.8
• DOI: 10.1016/j.measurement.2019.05.051
• Languages: English
• ISSNs: 0263-2241

Abstract

Highlights: A dual-core CMF transmitter is developed with a DSP chip and an FPGA chip. The update speed of the drive signal and drive efficiency are improved. The MFME model is established by the SVM method based on grid search. On-line real-time correction experiments are performed to verify the validity. The SVM method reduces the MFMEs and has good generalization performance. Abstract: The lack of timely flowtube drive and large measurement error are two major problems for Coriolis mass flowmeter (CMF) to measure gas-liquid two-phase flow. To solve above problems, a dual-core CMF transmitter is developed with a DSP chip and an FPGA chip. The FPGA is used to realize a digital drive method for driving the flowtube to vibrate normally. The DSP is utilized to realize a zero-crossing detection algorithm for calculating the mass flowrate. The Support Vector Machine (SVM) method is adopted to establish the mass flow measurement error (MFME) model for on-line real-time correction. The experimental results show that the developed transmitter improves the update speed of the drive signal and drive efficiency, and reduces the execution time of the algorithm. The SVM method has higher modeling accuracy and good generalization performance, and can be effectively applied to on-line real-time correction of the MFMEs under the gas-liquid two-phase flow.