Quantitative risk assessment plays an important role in facilitating deepwater drilling safety. This paper proposes an accident causation network method for risk assessment during deepwater drilling. The method involves four basic steps: risk identification, failure scenarios definition, uncertainty evaluation, and accident path calculation. Risk factors regarding different sources such as the operator, process, equipment, and environment are firstly identified. Then a network topology based on graph theory models the underlying accident scenarios. Risk entropy is adopted to model the uncertain influence of multi-type risk factors. Dijkstra algorithm is finally used to calculate the shortest path from an initial event to a blowout accident. A deepwater drilling system is chosen as a case study to demonstrate the applicability of the proposed method. The result shows that changes of failure probabilities of risk factors lead to the variation of the shortest path both in probabilities and event sequences. With the assessment result, decision makers can take targeted measures to prevent accidents or reduce system risks.

定量风险评估在促进深水钻井安全中起着重要作用。提出了一种事故原因网络法，用于深水钻井过程中的风险评估。该方法包括四个基本步骤：风险识别，故障场景定义，不确定性评估和事故路径计算。首先确定与不同来源（如操作员，过程，设备和环境）有关的风险因素。然后，基于图论的网络拓扑对潜在的事故场景进行建模。采用风险熵对多种风险因素的不确定性影响进行建模。Dijkstra算法最终用于计算从初始事件到井喷事故的最短路径。选择了一个深水钻井系统作为案例研究，以证明该方法的适用性。结果表明，风险因素失效概率的变化导致概率和事件序列最短路径的变化。有了评估结果，决策者可以采取针对性的措施来预防事故或降低系统风险。