Subsea gas pipeline leak may cause the catastrophic consequences , e.g., offshore fire and explosion, and the overturning of floating offshore structures . Efficient risk assessment is critical to prevent the unexpected offshore accident due to subsea gas pipeline leak. The modern technology advancement urges the need for developing new risk assessment approach in the case that a large volume of data becomes available. A new methodology comprising of BRANN implemented with BN is proposed for dynamic risk assessment of process operations, which can capture the nonlinear and the non-sequential features of accident escalation. BN is used to model the accident scenario from causations to consequences considering the conditional dependencies among accident contributory factors. Subsequently, BN model as an informative base is mapped into BRANN model. The data-driven nature makes it better to capture the uncertainty among the cause-consequence relationships. The practicability of the methodology is illustrated by a case study of subsea gas pipeline leak accident. It is observed that BRANN model is superior in the network performance and prediction accuracy. This methodology can help to perform more effectively real-time risk assessment of process operations. 

A machine learning methodology for probabilistic risk assessment of process operations: A case of subsea gas pipeline leak accidents

Residual strength prediction of corroded pipelines using multilayer perceptron and modified feedforward neural network

To quantify the progress of corrosion damage and develop pipeline integrity management strategies, it is necessary to establish a reliable corrosion growth model. Due to the complexity of the corrosion process, the availability of data, and the limitations of various models in their applicability, there is currently no generally accepted optimal corrosion growth prediction methodology. Corrosion data used for modeling, in-line inspection techniques for detecting defects, and sources of uncertainty in the modeling process are briefly described. This paper focuses on reviewing the concepts, the performance, and the application of existing pipeline corrosion growth models. The deterministic and probabilistic models are analyzed in detail according to the core methods involved, and the latest applications of machine learning and deep learning in corrosion growth modeling are also introduced. To leverage the strengths of various models, this paper presents hybrid approach models based on the combinations of the aforementioned models, which have greater performance and interpretability than single models and should be given more attention in the future development of corrosion growth prediction. Finally, some suggestions for future development are put forward in light of the challenges and deficiencies present in the current modeling process. 

Advances in corrosion growth modeling for oil and gas pipelines: a review

Evolution of corrosion prediction models for oil and gas pipelines: from empirical-driven to data-driven

Edge-cloud cooperation-driven smart and sustainable production for energy-intensive manufacturing industries

A KPCA-BRANN based data-driven approach to model corrosion degradation of subsea oil pipelines

Ultra-deep oil and gas wells have become a new development trend in onshore oil and gas exploitation. However, Ultra-deep oil and gas wellbore casing is with high failure risk due to the harsh environment. It is essential to evaluate the reliability of wellbore casing. This paper assesses the operational reliability of wellbore casing using data statistics and numerical simulation. Firstly, the theoretical model for reliability analysis of wellbore casing is established, and the variables in the model are determined, including rock mechanics, cement ring, and casing string strength factors. Subsequently, considering the random distribution of model variables, many statistics and analyses are performed to determine the distribution parameters of the model variables. Eventually, Monte Carlo based numerical simulations are carried out to obtain the residual strength distribution and the reliability of wellbore casing. The production casing in the ultra-deep well with a depth of 6.5 km in China as an industrial case is used to illustrate the present study. It is observed that this study can be useful to guide a more accurate assessment of the reliability of ultra-deep wellbore casing.

Shangyu Yang

Papers

A KPCA-BRANN based data-driven approach to model corrosion degradation of subsea oil pipelines

Reliability assessment of ultra-deep oil and gas wellbore casing using data statistics and numerical simulations