Digital Twin in Oil and Gas Equipment for Predictive Maintenance

Abstract Digital twins as a virtual model that is designed to reflect a physical system can improve the performance of electrical and mechanical components in industries. With digital twin technology, the industrial organization can be assumed in a way that they implement a sustainable solution with stability and predictability. By using MATLAB and Simulink, in this paper, a digital twin was demonstrated for Electro-Mechanical parts of the drilling for oil and gas industries. Hence, every single element of electrical and mechanical components was defined including the rotational parts separately, then by connecting those parts, the entire process in a closed-loop control system can be controlled. Having this digital twin, healthy data was generated for the situation that the system is working properly, as well as failure data that can be assigned for those components that are prone to wear and failure. Considering a fault in the digital twin in every specific part of the equipment, fault data was generated that can be used for data analysis and machine learning to find a Remaining Useful Life (RUL). To show how it is possible, in the next step, a digital twin for a triplex hydraulic pump that is crucial equipment in the oil and gas industries was investigated, then by having a simulation in healthy and faulty situations the required data were generated. Having data, an RUL was defined by implementing a data analysis and machine learning that can be implemented to have predictive maintenance for the system and it is shown how building a digital twin and having predictive maintenance for our system may boost productivity while decreasing unexpected downtime, which is costly and time-consuming.