Abstract：To realize a real-time monitoring of ultrasonic flowmeter accuracy, a real-time in-use measurement system of ultrasonic flowmeter based on the high pressure closed loop facility was developed, which was combined with a prediction model of ultrasonic flowmeter with the relationship between signal parameters (including sound velocity deviation) and indicating value error based on random forest algorithm, and LabVIEW software platform. Within the measurement system, three characteristic signals of ultrasonic flowmeters, including signal quality, flow state index, sound speed index, can be real-time collected, and the simultaneous prediction of flowmeter indication error from the prediction model was obtained. The experimental tests were carried out in the range of 160~800m3/h, which showed that the deviation between the real error and the prediction error of the flowmeter was within the range of±0.42%.
RÍOS-MERCADO R Z, BORRAZ-SNCHEZ C. Optimization problems in natural gas transportation systems: A state-of-the-art review［J］. Applied Energy, 2015, 147: 536-555.
SONG C F, HAN W, GUO Z, et al. Analysis of typical cases of sound velocity verification method for ultrasonic flowmeter in use ［J］. Industrial Measurement, 2020, 30 (2): 83-86.
CAO H C. Realization of remote diagnosis of ultrasonic flowmeter in Yuji pipeline［J］. Petrochemical Technology, 2017, 24 (11): 97.
YAO K C, HUANG W T, WU C C, et al. Establishing an AI Model on Data Sensing and Prediction for Smart Home Environment Control Based on LabVIEW［J］. Mathematical Problems in Engineering, 2021, 2021: 1-18.
ZHANG H, LI C H, CUI L S, et al. The close loop standard device of high pressure gas flow ［J］. Acta Metrologica Sinica, 2017, 38 (3): 324-327.
［12］
STROBL C, MALLEY J, TUTZ G. An introduction to recursive partitioning: Rationale, application, and characteristics of classification and regression trees, bagging, and random forests［J］. Psychological Methods, 2009, 14(4): 323-348.
WANG P, GONG P, FENG D, et al. Soft Sensing Method for Water Cut of Crude Oil Based on Random Forest Algorithm［J］. Acta Metrologica Sinica, 2019, 40(5): 835-841.
HAO H, ZHANG H L. Design of flow measurement system based on LabVIEW［J］. IOP Conference Series: Materials Science and Engineering, 2020, 789(1): 012027.
［3］
Transmission Measurement Committee. Measurement of Gas by Multipath Ultrasonic Meter［R］. AGA Report No.10, 2014.
ZHAO W G, BU Q C, YAO H B, et al. A Data Fusion Method of Double-channel Ultrasonic Flowmeter Application in low pressure Gas［J］. Acta Metrologica Sinica, 2021, 42(7): 873-878.
［17］
FUKAMI K, FUKAGATA K, TAIRA K. Assessment of supervised machine learning methods for fluid flows［J］. Theoretical and Computational Fluid Dynamics, 2020, 34(4): 497-519.
LI M N, LYU C Z, WANG L, et al. In-use Measurement of Ultrasonic Flowmeter Based on Machine Learning Algorithms［J］. Acta Metrologica Sinica, 2022, 43(12): 1627-1633.
［15］
LIN C Y. Forward stepwise random forest analysis for experimental designs［J］. Journal of Quality Technology, 2021, 53(5): 488-504.
［8］
SHANAB A A, KHOSHGOFTAAR T M, WALD R, et al. Evaluation of the importance of data preprocessing order when combining feature selection and data sampling［J］. International Journal of Business Intelligence and Data Mining, 2012, 7(1/2): 116.
2024, Vol. 45 
