Analysis of urban hydrogen-blended natural gas pipeline leak failure and accident evolution based on the combination of causal inference and probabilistic machine learning

Key Points

• Pipeline leak failures were primarily influenced by factors related to hydrogen blending and natural gas infrastructure.
• The analysis reveals a significant correlation between hydrogen concentration and leak incidents.
• Assessment using causal inference and probabilistic machine learning enables detailed examination of failure patterns over time.
• Findings highlight the need for enhanced preventive measures in urban natural gas systems, particularly involving hydrogen blends.