Digital twin-driven intelligent control of natural gas flowmeter calibration station

With the rapid growth of the natural gas industry, the handover volume of natural gas measurement is increasing every year, thus making the calibration of natural gas flowmeters more and more critical. The flow calibration process includes adjusting pressure and flow rate and performing calibration operations, which are manually carried out by testing personnel. The increasing demand for expertise and proficiency in the testing personnel poses significant challenges to on-site safety production. In this paper, we propose a digital twin-driven intelligent control of natural gas flowmeter calibration station to improve the efficiency of flowmeter calibration. First, a virtual simulation platform combining first-principle models and data-driven models is established, which can simulate the hydraulic state in the calibration station to provide verification for the control algorithm. Then, a fully automated flow control system is developed, which performs predictive control based on a hybrid driven model. This process first identifies the current process and automatically switches to the best process based on the diameter of the calibrated meter. Then, it performs initial control using a model predictive controller, and finally, improves control accuracy by using a BP neural network. The results show that the digital twin-driven intelligent control of natural gas flowmeter calibration station can accurately obtain hydraulic state and control strategy, and significantly improve the calibration efficiency.