SYSTEM IDENTIFICATION AND PARAMETRIC ESTIMATION OF INFERENTIAL CORIOLIS

Metering technology offers a number of possible options for the measurement of compressible natural gas (CNG). However, the accuracy of these measurements is dependent on various dynamic factors and fluid parameters. To avoid flow measurement from such dynamic errors, a new technique or novel concept of flowmeter is needed for measuring CNG. One of the options is to use natural force phenomenon that could be derived from fundamental physics, a force known as coriolis. It is used in mass flowmeter design that uses vibration tubes to guide and measure fluid or gas based on coriolis force. The motivation behind the research is to develop and apply coriolis in an embedded FieldPoint controller proclaimed as an inferential coriolis. The major challenge is to find a suitable algorithm for coriolis in the form of a mathematical model that could measure mass and mass flowrate of CNG with maximum permissible error. To define such system, an experimental approach known as System Identification (SYSID) theory is used. Performance of inferential coriolis is tested on experimental natural gas test rig which could be summarized into three areas: single pressure flow; continuous pressure flow; multi pressure flow with disturbances. When experiment was conducted, mass flowrate was measured using inferential coriolis and a commercial flowmeter from a manufacturer i.e., Micro Motion. To validate both methods, a load cell was used as the reference. Details evaluations of three pressure flow scenarios namely the single pressure flow, the continuous pressure flow, and the multi pressure flow for a CNG refueling system are presented. From percentage error analyses, it shows that in all measurements the inferential coriolis have less error compares to commercial coriolis manufactured by Micro Motion. The findings demonstrate the viability of the SYSID approach to provide a solution to the modeling of an inferential coriolis, and confirm the qualitative behavior of the inferential coriolis in response to the different flow measurements.