Santhosh K. Venkata and Binoy K. Roy
Artificial neural network, Ultrasonic flow meter, sensor modelling, temperature compensation, Non contact type
This paper aims to design an intelligent non contact type flow measurement scheme using Ultrasonic Flow Meter UFM). The objectives of the present work are (a) to extend the linearity range of the proposed flow measurement scheme using UFM, (b) to make the proposed flow measurement scheme adaptive w.r.t variation in (i) pipe diameter, (ii) liquid density and (iii) temperature of liquid, all within a range. The output of ultrasonic flow meter is frequency. It is converted to voltage by using a Frequency to Voltage (F-V) converter. An ANN block is added in cascade to the F-V converter. This arrangement helps to linearise the overall measurement system and makes the output adaptive of variations in pipe diameters, densities and temperatures of liquid. Since, the proposed intelligent flow measurement scheme produce output considering variation in physical dimensions of pipe, liquid densities and temperatures. Thus the proposed work avoids the requirement of repeated calibration every time for any change in liquid, pipe and temperature of liquid, all within the specified range. ANN is trained, tested and validated with simulated data considering variations in pipe diameters, liquid densities and temperatures of the liquid. These variations are considered within a specified range. When an unknown flow is tested with an arbitrary pipe diameter, liquid density and temperature of liquid, all within the specified range, the proposed scheme has measured the flow correctly. Results show that the proposed scheme has fulfilled the objectives.
Important Links:
Go Back
