Development of importance measures reflecting the risk triplet in dynamic probabilistic risk assessment; The Concept and measures of risk importance

成川 隆文*; 高田 孝*; Zheng, X.; 玉置 等史; 柴本 泰照; 丸山 結; 高田 毅士

Journal of Nuclear Engineering (Internet), 6(4), p.49_1 - 49_14, 2025/12

https://doi.org/10.3390/jne6040049

Despite the advancements in dynamic probabilistic risk assessment methodologies that account for the dynamics of event progression, the development of risk importance measures for such methodologies remains a significant research challenge, particularly in terms of fully capturing the rich, multidimensional risk information provided by dynamic PRA. This study proposes novel risk importance measures from the perspective of the risk triplet: Timing-Based Worth (TBW), which captures the scenario occurrence timing (scenario diversity), Frequency-Based Worth (FBW), which reflects the likelihood of scenarios, and Consequence-Based Worth (CBW), which represents the consequences of scenarios. These three measures are formally defined, and a conceptual framework for integrated importance evaluation is presented to enable multidimensional assessment. As a preliminary demonstration, TBW and FBW are applied to a simplified reliability model using a dynamic PRA based on the continuous Markov chain Monte Carlo (CMMC) method to evaluate their interpretability and the coherence of the proposed conceptual framework. The results demonstrate that TBW and FBW enable a more comprehensive risk importance evaluation by capturing resilience effects and temporal diversity, alongside existing frequency-based evaluations. This advancement is expected to enhance the practical use of dynamic PRA outputs in risk-informed decision-making.

Dynamic probabilistic risk assessment of nuclear power plants using multi-fidelity simulations

Zheng, X.; 玉置 等史; 杉山 智之; 丸山 結

Reliability Engineering & System Safety, 223, p.108503_1 - 108503_12, 2022/07

AA2021-0588.pdf:1.29MB

https://doi.org/10.1016/j.ress.2022.108503

Dynamic probabilistic risk assessment (PRA) more explicitly treats timing issues and stochastic elements of risk models. It extensively resorts to iterative simulations of accident progressions for the quantification of risk triplets including accident scenarios, probabilities and consequences. Dynamic PRA leverages the level of detail for risk modeling while intricately increases computational complexities, which result in heavy computational cost. This paper proposes to apply multi-fidelity simulations for a cost- effective dynamic PRA. It applies and improves the multi-fidelity importance sampling (MFIS) algorithm to generate cost-effective samples of nuclear reactor accident sequences. Sampled accident sequences are paralleled simulated by using mechanistic codes, which is treated as a high-fidelity model. Adaptively trained by using the high-fidelity data, low-fidelity model is used to predicting simulation results. Interested predictions with reactor core damages are sorted out to build the density function of the biased distribution for importance sampling. After when collect enough number of high-fidelity data, risk triplets can be estimated. By solving a demonstration problem and a practical PRA problem by using MELCOR 2.2, the approach has been proven to be effective for risk assessment. Comparing with previous studies, the proposed multi-fidelity approach provides comparative estimation of risk triplets, while significantly reduces computational cost.