期刊:
Journal of Circuits, Systems and Computers,2024年34(05):2550120 ISSN:0218-1266
通讯作者:
Li, PC
作者机构:
[Li, Pengcheng; Wen, Jing; Li, PC] Univ South China, Sch Resources & Environm & Safety Engn, Hengyang 421001, Hunan, Peoples R China.;[Wen, Jing] Hunan Inst Technol, Sch Design & Art, Hengyang 421002, Hunan, Peoples R China.;[Wang, Ye; He, Jiayuan; Li, Pengcheng; Wen, Jing; Li, PC] Univ South China, Inst Human Factors, Hengyang 421001, Hunan, Peoples R China.;[Tan, Haibo] China Nucl Power Engn Co Ltd, State Key Lab Nucl Power Safety Technol & Equipmen, Shenzhen 518172, Guangdong, Peoples R China.
通讯机构:
[Li, PC ] U;Univ South China, Sch Resources & Environm & Safety Engn, Hengyang 421001, Hunan, Peoples R China.;Univ South China, Inst Human Factors, Hengyang 421001, Hunan, Peoples R China.
关键词:
Team error;digital control rooms;nuclear power plants;performance influencing factors;human factors research
摘要:
To identify the shortcomings in current research on team errors and to outline future research directions aimed at enhancing safety management in the nuclear field, a review on team errors of key employees at nuclear power plants (NPPs) is compiled. The introduction of digital control systems in NPPs has fostered new human issues. To date, resources have focused on individual errors; however, the event investigations reveal that team errors significantly contribute to poor performance and accidents. The importance of team errors cannot be ignored, and incorporating team error control into safety systems is necessary for efficiently functioning NPPs. This study reviews fundamental aspects of team errors in digital control rooms of NPPs, covering three main foci: concept and classification of team errors, identification and classification of team error performance influencing factors (PIFs), and mechanisms of team errors. The main contributions of this review include the findings that (1) team interaction is not adequately captured in the definition of team errors; (2) the identification of team error patterns lacks an in-depth exploration of cognitive mechanisms, and the classification of team errors lacks uniform standards and objective methods; (3) 21 key PIFs leading to team errors have been identified, but the correlations between error patterns and PIFs have not been explored in depth; (4) lacking reliable systematic methods or models for analyzing team error mechanisms, especially in dealing with team dynamics and psychological error mechanisms. This research contributes to a more comprehensive, systematic understanding of team errors in NPPs, providing theoretical references for subsequent team error prevention and control strategy development.
作者机构:
[Fu, Pengfei] Univ South China, Coll Comp Sci, Hengyang, Peoples R China.;[Dai, Licao; Dai, LC] Univ South China, Human Factor Inst, Hengyang, Peoples R China.
通讯机构:
[Dai, LC ] U;Univ South China, Human Factor Inst, Hengyang, Peoples R China.
关键词:
Human reliability analysis;operator knowledge base;causal knowledge extraction;deep learning;BERT model
摘要:
In human reliability analysis (HRA) for nuclear power plants, cognitive modeling-based approaches require the development of an operator knowledge base to simulate the cognitive processes of operators. However, existing automatic extraction methods fail to provide knowledge that meets the granularity requirements of cognitive modeling for the development of the operator knowledge base.To address this gap, this paper proposes a deep learning-based extraction method. Specifically, the method utilizes a bidirectional encoder representations from transformers (BERT)-bidirectional long short-term memory (Bi-LSTM)-1conditional random field (CRF) model to perform sequence labeling for extracting fine-grained knowledge, such as entities and their corresponding states, as well as the causal relationships between these pieces of knowledge. Additionally, we define mapping rules to structure the extracted causal knowledge to facilitate the integration of additional knowledge.To validate the extraction effectiveness of the BERT-Bi-LSTM-CRF model, experiments were conducted on a data set constructed from licensee event reports. The experimental results showed that the model achieved a macro-F1 score of 0.876 on the test set, indicating that the model is capable of effectively extracting the required knowledge and relationships from unstructured text. This method is expected to be applied in the development of operator knowledge bases, potentially reducing the workload involved.
作者机构:
[Zhang, Meihui; Li, Yu; Dai, Licao] Univ South China, Inst Human Factors, Hengyang 421001, Peoples R China.;[Li, Yu] Univ South China, Sch Comp Sci, Hengyang 421001, Peoples R China.;[Zhang, Meihui] Univ South China, Sch Nucl Sci & Technol, Hengyang 421001, Peoples R China.
通讯机构:
[Licao Dai] I;Institute of Human Factors, University of South China, Hengyang 421001, China<&wdkj&>Author to whom correspondence should be addressed.
关键词:
clustering;eyes’ blink rate;fatigue;mouse velocity;nuclear power plant main control room;PERCLOS;supervised learning
期刊:
Progress in Nuclear Energy,2022年144:104086 ISSN:0149-1970
通讯作者:
Pengcheng Li
作者机构:
[Luo, Zhuhua; Li, Pengcheng; Liu, Yahua; Dai, Licao] Univ South China, Sch Resources Environm & Safety Engn, Hengyang 421001, Hunan, Peoples R China.;[Jin, Xiao; Luo, Zhuhua; Li, Pengcheng; Liu, Yahua; Liu, Zhen; Dai, Licao] Univ South China, Human Factor Inst, Hengyang 421001, Hunan, Peoples R China.
通讯机构:
[Pengcheng Li] S;School of Resources, Environment and Safety Engineering, University of South China, Hengyang 421001, Hunan, People's Republic of China<&wdkj&>Human Factor Institute, University of South China, Hengyang 421001, Hunan, People's Republic of China
关键词:
Team situation awareness;Human reliability analysis;Dynamic Bayesian network;Digital nuclear power plants
摘要:
Team situation awareness (TSA) reliability is an important factor for team reliability. Moreover, situation awareness (SA) is a prominent problem in digital nuclear power plants (NPPs). Currently, since there is no suitable method to dynamically assess TSA reliability, we constructed a dynamical assessment method of TSA reliability based on a dynamic Bayesian network (DBN) to evaluate TSA reliability. First of all, a TSA causal concept model through qualitative analysis, expert group discussion and sample data analysis. On this basis, the quantitative assessment method of TSA reliability was constructed based on DBN and obtained probability distribution of variables. A standardized method was established to obtain the probability distribution of variables. Furthermore, we evaluated TSA dynamic reliability in a steam generator tube rupture (SGTR) accident. The results showed that the error probability of TSA decreased, and the level of TSA reliability continuously increased in SGTR. TSA reliability can be dynamically predicted by causal reasoning, the most important cause of TSA error could be identified by diagnostic reasoning, which provided theoretical support for the targeted prevention of human error. Finally, this established method was proved to be reasonable through sensitivity analysis.
作者机构:
[Li, Pengcheng; Zhang, Li; Li, Xiaofang; Dai, Licao] Univ South China, Human Factor Inst, Hengyang 421001, Hunan, Peoples R China.;[Li, Pengcheng] Univ N Carolina, Syst Engn & Engn Management, Charlotte, NC 28223 USA.;[Li, Pengcheng] Univ South China, Sch Nucl Sci & Technol, Hengyang 421001, Hunan, Peoples R China.
通讯机构:
[Dai, Licao] U;Univ South China, Human Factor Inst, Hengyang 421001, Hunan, Peoples R China.
关键词:
Model validation;Simulation experiment;SA reliability;Fuzzy logic-AHP;Digital nuclear power plants
摘要:
Situation awareness (SA) reliability of operators is an important component of human reliability analysis (HRA) in digital nuclear power plants (NPPs). Therefore, how to identify effectively and reliably the risk of SA errors is of great significance for SA error prevention and risk reduction. SA reliability assessment model or method based on fuzzy logic and analytic hierarchy process (AHP) is forwarded in the first paper. In order to prove its effectiveness and reliability, sensitivity analysis and simulator experiments are applied to testify our model in this paper. The results show that the proposed assessment model of SA reliability has a certain degree of sensitivity as well as an accuracy of prediction, and there are a strong correlation and consistency between the predicted value from fuzzy inference system and the real value (or observed value) from simulator experiments. Although there are some differences between the predicted value and the real value based on the perspective of HRA, the model can be used to predict SA error probability or SA reliability within the range of 10 error factors as well as provide data and theoretical support for SA reliability assessment.
作者机构:
[Li, Pengcheng; Zhang, Li; Li, Xiaofang; Dai, Licao] Univ South China, Human Factor Inst, Hengyang 421001, Hunan, Peoples R China.;[Li, Pengcheng] Univ N Carolina, Syst Engn & Engn Management, Charlotte, NC 28223 USA.;[Li, Pengcheng] Univ South China, Sch Nucl Sci & Technol, Hengyang 421001, Hunan, Peoples R China.;[Zou, Yanhua] Hunan Inst Technol, Inst Human Factor & Safety Management, Hengyang 421002, Hunan, Peoples R China.
通讯机构:
[Li, Pengcheng] U;Univ South China, Human Factor Inst, Hengyang 421001, Hunan, Peoples R China.
关键词:
Situation awareness reliability;Fuzzy logic;Analytic hierarchy process;Digital nuclear power plants
摘要:
In digital control rooms, situation awareness (SA) reliability has become an important element affecting operator's reliability. In order to establish a more reasonable assessment method of SA reliability under the condition of very lack of data, based on the established influential factor model of SA reliability considering the causality relationship of performance shaping factors (PSFs) in this paper, a fuzzy logic and analytic hierarchy process (AHP)-based method is established to more objectively assess SA reliability. The weight of PSFs is identified using AHP, and a fuzzy logic method is used to simulate the fuzzy assessment and reasoning process of operator's SA reliability, and a standardized method is built to determine the fuzzy rule base of fuzzy reasoning system for SA reliability assessment for reducing the subjectivity and uncertainty of expert judgment. Finally, an example is provided to illustrate the specific application of the proposed method. The results show that the established method takes account of the weight of PSFs and their causal influencing relationship, and the fuzzy logic method used to assess SA reliability can overcome the subjectivity and uncertainty of expert judgment, which makes the assessment results more objective and realistic. Furthermore, the method can be used to get more SA error data and have a wide range of application value.
摘要:
Shearing machines are the key pieces of equipment for spent–fuel reprocessing in commercial reactors. Once a failure happens and is not detected in time, serious consequences will arise. It is very important to monitor the shearing machine and to diagnose the faults immediately for spent–fuel reprocessing. In this study, an intelligent condition monitoring approach for spent nuclear fuel shearing machines based on noise signals was proposed. The approach consists of a feature extraction based on wavelet packet transform (WPT) and a hybrid fault diagnosis model, the latter combines the advantage on dynamic–modeling of hidden Markov model (HMM) and pattern recognition of artificial neural network (ANN). The verification results showed that the approach is more effective and accurate than that of the isolated HMM or ANN.