Intelligent non-intrusive thermal flow rate meter designed for small diameter applications

Abstract

Current non-intrusive flow measurement techniques still need improvements as they have disadvantages in small-diameter applications. This work proposes to develop a non-intrusive thermal flow meter to obtain the slightest full-scale deflection possible in low liquid flows. The meter uses a copper duct with an internal diameter of $22mm$, six commercial K-type thermocouples, a microtubular heating resistance, and artificial intelligence to infer the flow rate from the thermal distribution on the duct surface. The sensors and the heater layout were calculated based on the theoretical temperature spread obtained from the physical model. To evaluate the meter, a test bench was built to control the heating resistor’s flow rate and temperature. The test bench is equipped with an electromagnetic flowmeter calibrated and certified in an external laboratory for reference and comparison, according to the ABNT (Brazilian Association of Norms Techniques) guidelines and the good practices used in the industries and calibration laboratories. In the meter evaluation, the resistance was activated so that the duct’s central region’s temperature remained at 70 degrees Celsius, and the thermal distribution data was collected with flow rates between 0.05 and $0.6m^3/h$ with intermediate increases of $0.01m^3/h$. The experiment’s collected data were used to train the following models: linear regression, K-Nearest Neighbor (K-NN), Decision Tree, Random Forests, and Gradient Boosting. Deep learning models were also trained. The best result was obtained with k-NN, demonstrating that the built prototype could infer the flow rate with a full-scale deflection equal to 0.14%. As a result, the evaluation indicated that artificial intelligence algorithms could improve non-intrusive flow measurement systems compared to the proposed analytical thermal flow model.