Soft sensor for an orifice flowmeter in presence of disturbances

Abstract

Measurements using an orifice flowmeter are widely used in industry. In certain instances, the output of the flowmeter may be corrupted due to plate contamination, changes in fluid density, or incorrect insertion of the plate. This paper describes a method for estimating the correct output in the presence of such disturbances. First, a linear parameter-varying model of the orifice flowmeter is developed using data extracted from computational fluid dynamics simulations. The simulation and experimental output are found to have an average deviation of 6.5% and 3.49% in terms of the differential pressure and discharge coefficient, respectively. Observer-based estimators for the linear parameter-varying models are developed for different combinations of the settling time and maximum overshoot. These estimators enable the disturbance-induced output to be corrected close to the true value. The error in the disturbed output due to plate contamination is reduced from 45% to 0.82%. Similarly, the error due to an accidental change of plate decreases from 76% to 2.03%. Thus, the proposed estimator can be used to nullify the disturbances induced in the measurements from orifice flowmeters.