Underground gas storage reservoirs are vital to China's energy infrastructure, facilitating natural gas peak shaving and supply security. However, cyclic alternating loads from significant pressure fluctuations during injection and production operations lead to severe erosion and wear of tubing strings, complicating safe operation. This study developed a custom jet flow gas–solid–liquid erosion experimental apparatus to analyze the effects of impact angle, impact velocity, and particle size on tubing erosion rates. An erosion model was established from the experimental data. Using computational fluid dynamics flow field analysis, the study evaluated how various factors—such as production rates, particle properties, reservoir pressure, and gas–liquid ratio—affect erosion patterns under production conditions. It also explored the impact of these factors under gas injection conditions. This research offers valuable insights for enhancing the functionality of supply demand regulation and strategic reserves in underground gas storage facilities.

The high hazard mechanical system (HHMS) has three characteristics in the petroleum and petrochemical industry (PPI): high risk, high cost, and high technology requirements. For a HHMS, part, component, and subsystem failures will result in varying degrees and various types of risk consequences, including unexpected downtime, production losses, economic costs, safety accidents, and environmental pollution. Thus, obtaining the quantitative risk level and distribution in a HHMS to control major risk accidents and ensure safe production is of vital importance. However, the structure of the HHMS is more complex than some other systems, making the quantitative risk analysis process more difficult. Additionally, a variety of uncertain risk data hinder the realization of quantitative risk analysis. A few quantitative risk analysis techniques and studies for HHMS exist, especially in the PPI. Therefore, a study on the quantitative risk analysis method for HHMS was completed to obtain the risk level and distribution of high-risk objects. Firstly, Fuzzy Set Theory (FST) was applied to address the uncertain risk data for the occurrence probability (OP) and consequence severity (CS) in the risk analysis process. Secondly, a fuzzy fault tree analysis (FFTA) and a fuzzy event tree analysis (FETA) were used to achieve quantitative risk analysis and calculation. Thirdly, a fuzzy bow-tie model (FBTM) was established to obtain a quantitative risk assessment result according to the analysis results of the FFTA and FETA. Finally, the feasibility and practicability of the method were verified with a case study on the quantitative risk analysis of one reciprocating pump system (RPS). The quantitative risk analysis method for HHMS can provide more accurate and scientific data support for the development of Asset Integrity Management (AIM) systems in the PPI.