Reliability and Safety Assessment of Passive Safety Systems Through Coupling of Fault Tree Analysis and Artificial Neural Network
Bibliographic record
Abstract
Abstract Small Modular Reactors (SMRs) have attracted much attention in recent years, and they could play a significant role in the future of energy supply and the nuclear industry. Many factors have contributed to the advancement of SMRs, including their affordability and zero greenhouse gas emissions. However, the most significant advantage associated with SMRs is their increased safety level, which has been achieved by introducing a wide range of new design features. Despite the diversity of design techniques, a similar set of design principles, such as Passive Safety Systems (PSSs), has been adopted to improve plant safety and robustness, eliminate design vulnerabilities, minimize accident likelihood, and mitigate accident effects. Reliability and safety evaluation of PSSs are crucial from the design phase to achieve these objectives. Probabilistic Safety Assessment (PSA) is a well-known methodology for analyzing risk levels associated with safety-critical systems in many industries, such as the aerospace, oil and gas, and nuclear industries. Probabilistic safety assessment utilizes the combination of Event Tree (ET) and Fault Tree (FT) techniques to estimate risks associated with certain undesired top events, such as core meltdown in the nuclear industry. Although PSA offers a range of advantages for safety assessment compared with traditional deterministic risk analysis technology, it also has some limitations. There are still many challenges associated with dynamic PSA analysis due to the demand for computational power for oversized FTs and ETs. Moreover, the final assessment result is prone to a significant uncertainty level due to human-related errors. Some of the challenges associated with PSA might be alleviated by Artificial Neural Networks (ANNs), as ANNs address the limitations of PSA, such as adaptive capacity, learning ability, and real-time calculation, which are challenging for dynamic process systems. Apart from ANNs, Bayesian Networks (BNs) are used to establish the collection of stochastic processes and their conditional dependencies through graphical connections. Bayesian Network is a graph layout that models accident scenarios and various real-world problems. This paper investigates the application of artificial intelligence (Deep Learning (DL)) to enhance FT analysis through the conversion of FT and ANN models. The potentiality of extending this technique to analyze the reliability and safety of PSSs in SMRs is examined. In SMRs, natural circulation has a low driving force, and PSSs are easily manipulated by system variables such as heat loss, flow friction, and oxidation, leading to system instability and jeopardizing the system’s safety. As a result, FT analysis is inadequate to capture these effects in real-time to analyze the reliability and safety of PSSs. This paper demonstrates that the introduction of ANN could help address some of these limitations.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".