STAMP框架下化学品船智能液货系统风险分析

Risk analysis of the intelligent liquid cargo system of the chemical tanker under the STAMP framework

为分析化学品船智能液货系统的安全性,采用系统理论事故模型与过程(System-Theoretic Accident Model and Processes, STAMP)方法,构建了化学品船智能液货系统控制反馈模型;基于系统理论过程分析(System-Theoretic Process Analysis, STPA),确定系统级事故,识别不安全控制行为并分析不安全控制行为关键致因;将不安全控制行为作为风险因素,利用三角模糊数对不安全控制行为风险进行量化。结果表明,26种不安全控制行为中,泵管阀动态调节、人员监测、数据采集、货品相容性判定等因素对风险影响程度较高。该分析结果可为化学品船智能液货系统的应用和安全管理提供参考。

To analyze the safety of the intelligent liquid cargo system of chemical tankers, the safety issue of equipment, crew and shore-based operator in the intelligent liquid cargo system is regarded as control and feedback problems. The control-feedback model of the intelligent liquid cargo system of the chemical tanker is established by adopting the System-Theoretic Accident Model and Processes(STAMP) method. The system-level accidents such as hull damage, loss of life, and equipment damage as well as the system-level hazards such as shore-based communication interruption, and safe space being occupied are determined through the System-Theoretic Process Analysis(STPA). According to the module composition of the intelligent liquid cargo system, the control-feedback structure of the intelligent liquid cargo system working process is analyzed, and 26 kinds of unsafe control acts that may lead to danger are identified. The key causes of unsafe control acts are analyzed from three aspects: control defects, feedback defects, and interactive coordination defects. The unsafe control act is regarded as the risk factor for intelligent liquid cargo systems, and the professionals are invited to make a subjective evaluation of the possible accidents caused by the risk factors. The weights of factors are obtained by Analytic Hierarchy Process(AHP), and the risks of unsafe control acts are quantified by triangular fuzzy numbers. The results show that dynamic adjustment of the valve, personnel monitoring, data collection, and compatibility judgment have a higher impact on the risk. Factors such as quality assessment, the concentration of attention, inspection of facilities, and matching of washing medium have a weak influence on the risk. Identification of information data, data conversion storage, verification of the validity of operation instructions, historical record calling, and cargo knowledge base updating have a slight impact on the risk. The analysis results can provide a reference for the application and safety management of the intelligent liquid cargo system of the chemical tanker.