|
| 基于博弈论-模糊贝叶斯网络的储能预制舱火灾风险评估 |
| Fire Risk Assessment of Energy Storage Prefabricated Compartment Based on Game Theory and Fuzzy Bayesian Networks |
| 投稿时间：2025-07-29 修订日期：2025-12-22 |
| 查看/发表评论 下载PDF阅读器 |
| DOI： |
| 基金项目:国家消防救援局科技计划项目(2024XFZD04) |
|
| 中文关键词: 储能预制舱 贝叶斯网络 事故树分析 博弈论 组合赋权 |
| 英文关键词:Prefabricated cabins for energy storage fuzzy Bayesian network accident tree analysis game theory Combinatorial empowerment |
| 摘要点击次数: 124 |
| 全文下载次数: 0 |
| 中文摘要: |
| 储能预制舱在现代能源体系中具有重要地位，但由于事故数据不足、事故机制复杂，其火灾风险评估存在一定不确定性。通过引入专家打分方法，量化各类风险因素，提高评估的科学性和合理性。本文提出一种基于贝叶斯网络与博弈论的储能预制舱火灾风险评估方法，针对传统贝叶斯网络依赖精确概率数据而储能系统事故样本稀缺的矛盾，引入模糊集理论，利用三角模糊数将专家语言变量转化为定量概率，有效解决了评估过程中的模糊性和不确定性问题。依据专家经验构建了火灾风险的事故树结构模型，采用博弈论方法对主客观权重进行优化整合，运用贝叶斯推理技术推导火灾的发生概率，通过敏感性和关键致因链分析，识别关键风险因素及风险路径。实例表明，某储能预制舱发生火灾的概率高达33%，电池温度过高、电池内部短路和电气短路是主要诱因。电池排列密集和高温环境这两个因素对火灾风险的影响最为显著，该方法有效识别了系统薄弱环节，为储能电站的“防消结合”与运维决策提供了科学依据。 |
| 英文摘要: |
| Energy storage prefabricated cabins are vital in modern energy systems, but their fire risk assessment faces uncertainty due to limited accident data and complex mechanisms. An expert scoring method is introduced to quantify risk factors, improving the assessment's scientific rigor and rationality. This paper proposes a fire risk assessment method for energy storage prefabricated cabins based on Bayesian Network and Game Theory.To address the conflict between the traditional Bayesian Network's reliance on precise probability data and the scarcity of accident samples in energy storage systems, Fuzzy Set Theory is incorporated. Triangular Fuzzy Numbers are utilized to convert expert linguistic variables into quantitative probabilities, effectively resolving the issues of ambiguity and uncertainty in the assessment process. A Fault Tree structural model of fire risk is constructed based on expert experience. Game theory is employed to optimize and integrate subjective and objective weights, and Bayesian inference technology is applied to deduce the probability of fire occurrence. Furthermore, critical risk factors and risk paths are identified through sensitivity analysis and critical causal chain analysis. The case study indicates that the probability of fire in the target energy storage prefabricated cabin is 33%. High battery temperature, battery internal short circuits, and electrical short circuits are identified as the primary triggers. Specifically, dense battery arrangement and high-temperature environments exert the most significant influence on fire risk. The proposed method effectively identifies the weak links in the system, providing a scientific basis for the "combination of prevention and protection" strategy and operation and maintenance decision-making for energy storage stations. |
| 关闭 |
