Study on reliability of control system in deepwater CBHP MPD system

Deepwater managed pressure drilling technology, with unique advantages in dealing with narrow density window, has been widely used in deepwater drilling operations. As the key equipment to realize managed pressure drilling, the control system in constant bottomhole pressure managed pressure drilling system plays an important role in drilling operations. In this study, a new reliability evaluation method based on fault tree analysis and dynamic Bayesian network model has been proposed to analyze the reliability of the control system. First, the structure of the control system in the deepwater managed pressure drilling system was analyzed to set up a fault tree model of the control system. Second, considering the influence of time on the reliability, a dynamic Bayesian network model of the control system was established. Third, the proposed model was partly verified using the verification method based on three axioms. Fourth, in consideration of the repair factor, the reliability and availability of the control system were calculated by using the exponential distribution model of failure time. Finally, the sensitivity of the control system was analyzed, and the mutual information, variance and entropy were calculated to find out weak links in key components of the control system.