Using Artificial Intelligence to Aid Measurement Accuracy and Reliability in Coriolis Gas Flow meters

Abstract

Coriolis is one of the fastest growing technologies in the oil and gas flow measurement. Flow sensitivity, pressure drop, temperature changes and increased noise level affects the accuracy and reliability of these types of transmitters. However in most situations those parameters are not constant and not a mathematical model exist to include in the logic for the actual gas flow calculation. AGA Report Number 11 specifically concentrates on the measurement of natural gas and the impact of expanded compositional ranges on the flow calculation. Using Artificial intelligence enables performance characteristics that are much better than traditional metering technologies. Compressibility factors for natural gas and other Hydrocarbon gases are some of the factors that will be considered in artificial intelligence model for gas flow measurement. In this paper uncertain input parameters will be identified as fuzzy variables and will be integrated into fuzzy calculation of the gas flow measurement. It can be argued that Coriolis technology integrated with artificial intelligence greatly increases the accuracy and robustness of flow calculation. Uses of artificial intelligence in these types of Coriolis flow meters can minimize error and extend sensor life. This paper will discuss the appropriate implementation of such expert system with Coriolis flow measurement technology.