Huang Lei

Bio: Huang Lei is an academic researcher from China University of Petroleum. The author has contributed to research in topics: Fault (power engineering) & Subsea. The author has an hindex of 3, co-authored 9 publications receiving 265 citations.

Bayesian Networks in Fault Diagnosis

Abstract: Fault diagnosis is useful in helping technicians detect, isolate, and identify faults, and troubleshoot. Bayesian network (BN) is a probabilistic graphical model that effectively deals with various uncertainty problems. This model is increasingly utilized in fault diagnosis. This paper presents bibliographical review on use of BNs in fault diagnosis in the last decades with focus on engineering systems. This work also presents general procedure of fault diagnosis modeling with BNs; processes include BN structure modeling, BN parameter modeling, BN inference, fault identification, validation, and verification. The paper provides series of classification schemes for BNs for fault diagnosis, BNs combined with other techniques, and domain of fault diagnosis with BN. This study finally explores current gaps and challenges and several directions for future research.

A reliability evaluation methodology of complex systems based on dynamic object oriented Bayesian networks

Abstract: A novel reliability evaluation methodology of complex systems is proposed by using Dynamic Object Oriented Bayesian Networks (DOOBNS). This modeling methodology consists of two main phases: one construction phase for Object Oriented Bayesian Networks (OOBNs) and another construction phase for DOOBNs. In the first phase, the network fragments that have the same structures and parameters are divided into classes, then the classes are encapsulated. OOBN construction is completed according to the relationship among the encapsulated classes. In the second phase, every fragment of DBNs which was constructed by the last phase is encapsulated as a class which is called DOOBN. DOOBN construction is completed according to the relationship between time fragments. The correctness of this methodology is validated by using an all series system, an all voting system, a voting after series system, a series after voting system, a parallel after series system and a series after parallel system. This methodology can model the system from global to local levels, effectively reducing the modeling difficulty, and can take more efficient arithmetic reasoning algorithms.

Intelligent monitoring and dynamic fault diagnosis system for subsea tree

Abstract: The invention belongs to the field of petroleum engineering and particularly relates to an intelligent monitoring and dynamic fault diagnosis system for a subsea tree. The intelligent monitoring and dynamic fault diagnosis system for the subsea tree comprises signal reception and fault diagnosis sub-systems and data acquisition and sonar signal emission sub-systems, wherein the signal reception and fault diagnosis sub-systems are located in a control room on a drilling platform, and the data acquisition and sonar signal emission sub-systems are located in undersea control modules; one data acquisition and sonar signal emission sub-system is arranged in each undersea control module; and data transmission between the signal reception and fault diagnosis sub-systems and data acquisition and the sonar signal emission sub-systems is carried out through sonar. According to the intelligent monitoring and dynamic fault diagnosis system for the subsea tree, the status information of multiple undersea tree systems can be monitored in real time and be respectively emitted to an offshore oil platform through sonar signals to carry out comprehensive diagnostic analysis, so that the safety for marine petroleum exploitation is improved.

Real-time reliability assessment system for deep water blowout preventer

Abstract: The invention belongs to the field of petroleum engineering and particularly relates to a real-time reliability assessment system for a deep water blowout preventer. The real-time reliability assessment system comprises a yellow box reliability assessment system positioned in an underwater yellow box, a blue box reliability assessment system positioned in an underwater blue box and a reliability parameter comprehensive processing module positioned in a central control unit, wherein the yellow box reliability assessment system and the blue box reliability assessment system have the same structure, and are both, through Ethernet buses, connected to the reliability parameter comprehensive processing module and, through signal cables, connected to pressure, temperature and flow sensors in a yellow box hydraulic control system and vibration and position sensors on a gate disc blowout preventer. When any one of data monitored by the vibration, position, pressure, temperature and flow sensors of the deep water blowout preventer system changes and the amplitude exceeds 3%, the real-time reliability assessment system automatically calculates system real-time reliability and remaining service life, and references are provided for the repair and maintenance of the blowout preventer system.

Bayesian Networks in Fault Diagnosis

Abstract: Fault diagnosis is useful in helping technicians detect, isolate, and identify faults, and troubleshoot. Bayesian network (BN) is a probabilistic graphical model that effectively deals with various uncertainty problems. This model is increasingly utilized in fault diagnosis. This paper presents bibliographical review on use of BNs in fault diagnosis in the last decades with focus on engineering systems. This work also presents general procedure of fault diagnosis modeling with BNs; processes include BN structure modeling, BN parameter modeling, BN inference, fault identification, validation, and verification. The paper provides series of classification schemes for BNs for fault diagnosis, BNs combined with other techniques, and domain of fault diagnosis with BN. This study finally explores current gaps and challenges and several directions for future research.

Improving the accuracy of medical diagnosis with causal machine learning.

Abstract: Machine learning promises to revolutionize clinical decision making and diagnosis. In medical diagnosis a doctor aims to explain a patient’s symptoms by determining the diseases causing them. However, existing machine learning approaches to diagnosis are purely associative, identifying diseases that are strongly correlated with a patients symptoms. We show that this inability to disentangle correlation from causation can result in sub-optimal or dangerous diagnoses. To overcome this, we reformulate diagnosis as a counterfactual inference task and derive counterfactual diagnostic algorithms. We compare our counterfactual algorithms to the standard associative algorithm and 44 doctors using a test set of clinical vignettes. While the associative algorithm achieves an accuracy placing in the top 48% of doctors in our cohort, our counterfactual algorithm places in the top 25% of doctors, achieving expert clinical accuracy. Our results show that causal reasoning is a vital missing ingredient for applying machine learning to medical diagnosis. In medical diagnosis a doctor aims to explain a patient’s symptoms by determining the diseases causing them, while existing diagnostic algorithms are purely associative. Here, the authors reformulate diagnosis as a counterfactual inference task and derive new counterfactual diagnostic algorithms.

Application of Bayesian Networks in Reliability Evaluation

Abstract: The Bayesian network (BN) is a powerful model for probabilistic knowledge representation and inference and is increasingly used in the field of reliability evaluation. This paper presents a bibliographic review of BNs that have been proposed for reliability evaluation in the last decades. Studies are classified from the perspective of the objects of reliability evaluation, i.e., hardware, structures, software, and humans. For each classification, the construction and validation of a BN-based reliability model are emphasized. The general procedural steps for BN-based reliability evaluation, including BN structure modeling, BN parameter modeling, BN inference, and model verification and validation, are investigated. Current gaps and challenges in reliability evaluation with BNs are explored, and a few upcoming research directions that are of interest to reliability researchers are identified.