Errors In Nuclear Power Plants Research Articles

Using artificial intelligence to detect human errors in nuclear power plants: A case in operation and maintenance

Human error (HE) is an important concern in safety-critical systems such as nuclear power plants (NPPs). HE has played a role in many accidents and outage incidents in NPPs. Despite the increased automation in NPPs, HE remains unavoidable. Hence, the need for HE detection is as important as HE prevention efforts. In NPPs, HE is rather rare. Hence, anomaly detection, a widely used machine learning technique for detecting rare anomalous instances, can be repurposed to detect potential HE. In this study, we develop an unsupervised anomaly detection technique based on generative adversarial networks (GANs) to detect anomalies in manually collected surveillance data in NPPs. More specifically, our GAN is trained to detect mismatches between automatically recorded sensor data and manually collected surveillance data, and hence, identify anomalous instances that can be attributed to HE. We test our GAN on both a real-world dataset and an external dataset obtained from a testbed, and we benchmark our results against state-of-the-art unsupervised anomaly detection algorithms, including one-class support vector machine and isolation forest. Our results show that the proposed GAN provides improved anomaly detection performance. Our study is promising for the future development of artificial intelligence based HE detection systems.

Operation validation system to prevent human errors in nuclear power plants

The importance of human and organizational factors in the nuclear industry has been emphasized, and along these lines there have been various improvements to human–system interfaces (HSIs). Operator support systems are one of the key HSIs that can contribute to reduce human errors. The system proposed in this work, called the concealed intelligent assistant (CIA), is designed as an operator support system using artificial intelligence algorithms. The CIA system validates operator actions and notifies operators of human errors to prevent adverse effects on plant integrity. The operation validation system consists of two-step filtering, namely through a procedural compliance check module and a comparison of safety impact evaluation module. Both modules employ a prediction algorithm with deep neural networks, where the former predicts the results of operators’ decision-making, and the latter implements plant parameter prediction strategies. The CIA system minimizes any increase in the cognitive burden on operators because it provides additional information only when a mistake is actually expected to adversely affect the integrity of the power plant.

Operation Validation System to Prevent Human Errors in Nuclear Power Plants

Operation Validation System to Prevent Human Errors in Nuclear Power Plants

A data mining framework within the Chinese NPPs operating experience feedback system for identifying intrinsic correlations among human factors

With the continuous increase in the number of operating nuclear power plants (NPPs) in China, the amount of operating experience feedback (OEF) increases significantly. On the other hand, the safe operation of NPPs has become an urgent problem that the National Nuclear Safety Administration (NNSA) must solve. To this end, NNSA established a nationalwide OEF system to improve the safety level of NPPs and strengthen the exchange of operating experience. Analyzing the human factors events (HFEs) is an important part of OEF and it is significant to improve human performance and prevent human error. Data mining has been recognized as an effective way to analyze data. With the continuous increase in operating event reports, data mining related to nuclear safety becomes a new domain of study. In this paper, we propose a data mining framework in support of the OEF system. The framework combines three statistical approaches (i.e., correlation analysis, cluster analysis and association rule mining) for identifying intrinsic correlations among human factors: correlation analysis measures the strength of linear relationship between human factors; cluster analysis classifies human factors into relevant groups; association rule mining identifies associations and causalities among human factors. For illustration, we apply the proposed framework to 162 human factors events (screened out from 313 events collected from the OEF system), and the results reflect the feasibility and effectiveness of the framework in identifying the intrinsic correlations among human factors. Besides, further suggestions for improving human performance and preventing human errors in NPPs are also discussed.

Systematic development of scenarios caused by cyber-attack-induced human errors in nuclear power plants

The digitalization of the instrumentation and control (I&C) systems in nuclear power plants has increased the threat from cyber-attack. This paper addresses cyber-attack vulnerability from potential attacks on non-safety I&C systems that result in a hazardous plant status. In safety-critical applications such as nuclear power plants, safety systems are separated and isolated from non-safety systems by design. Cyber-attacks on the non-safety systems can however escalate into plant safety threats by inducing wrong operator actions. We focus here on the operator actions that lead to the unavailability of the safety system and cause the failure of accident mitigation. In this study, the effect of safety system unavailability on plant safety is carefully modeled by using n conventional fault tree (FT) analyses, and human actions are analyzed based on emergency operating procedures. Based on the results of the FT analyses in combination with human action analyses, we suggest a novel method to systematically develop cyber-attack propagation scenarios, where a cyber-attack is linked to its consequences. As a case study, the feed-and-bleed operation is analyzed with the developed FT, producing multiple scenarios that lead to core damage. The results of this study are expected to be useful in establishing preventive measures

The Effect of Personality Type on Human Performance Tool Compliance and General Recommendations for Enhancement of the its Practical Utilization

Objective: The purpose of this study is to investigate the effect of personality type on human performance tool compliance in nuclear power plants (NPPs) and to propose general recommendations for an enhancement of its practical utilization. Background: Various guidelines, regulating criteria, and recommendations have been developed to prevent human errors in NPPs. Despite these efforts, the accidents sometimes caused by human errors have steadily occurred, and therefore, various human performance tools have been adopted as countermeasures against human errors. The major and inevitable contributing factors among the many hazards to human errors might be the trait and personality, which are considered to be the inner world of humans. Thus, we try to investigate the utilization of human performance tools by considering the different types of operating crew personalities, and we suggested more practical recommendations to prevent human errors according to the personality. Method: We developed the Questionnaire using the Big 6 (HEXACO) models, which are human performance tools for in NPPs, and individual (condition) variables to investigate the effect of personality types on human performance tools. We slightly modified them to help the survey respondents understand them better. A survey was conducted for ordinary people over the age of 20. SPSS 22.0 was used to perform a correlation analysis and a hierarchical regression analysis to find the relationship between personality types and human performance tools. Results: The utilization of human performance tools shows significant differences statistically by personality. The correlation result reveals that the types of Honesty (H), Extraversion (X), Conscientiousness (C), and Openness to experience (O) show a higher utilization of human performance tools. In hierarchical regression results, human performance tools of task preview, questioning attitude, stopping when unsure, self-checking, effective communication, and place-keeping show a higher utilization with personality types. However, the Agreeableness (A) type did not show significant differences statistically with human performance tools. Conclusion: We tried to investigate the utilization of human performance tools by considering the different types of human personality and provide more practical recommendations to prevent human errors according to the personality. These results will be able to prevent human errors owing to the characteristics (advantages and disadvantages) of personality types. Application: This information can be utilized as guidelines for proactive recommendations according to the workers personalities for more practical human performance tools to prevent human errors in an NPP.

Development of Management Guidelines and Procedure for Anthropometric Suitability Assessment: Control Room Design Factors in Nuclear Power Plants

Objective: The aim of this study is to develop management guidelines and a procedure for an anthropometric suitability assessment of the main control room (MCR) in nuclear power plants (NPPs). Background: The condition of the MCR should be suitable for the work crews in NPPs. The suitability of the MCR depends closely on the anthropometric dimensions and ergonomic factors of the users. In particular, the MCR workspace design in NPPs is important due to the close relationship with operating crews and their work failures. Many documents and criteria have recommended that anthropometry dimensions and their studies are one of the foremost processes of the MCR design in NPPs. If these factors are not properly considered, users can feel burdened about their work and the human errors that might occur. Method: The procedure for the anthropometric suitability assessment consists of 5 phases: 1) selection of the anthropometric suitability evaluation dimensions, 2) establishment of a measurement method according to the evaluation dimensions, 3) establishment of criteria for suitability evaluation dimensions, 4) establishment of rating scale and improvement methods according to the evaluation dimensions, and 5) assessment of the final grade for evaluation dimensions. The management guidelines for an anthropometric suitability assessment were completed using 10 factors: 1) director, 2) subject, 3) evaluation period, 4) measurement method and criteria, 5) selection of equipment, 6) measurement and evaluation, 7) suitability evaluation, 8) data sharing, 9) data storage, and 10) management according to the suitability grade. Results: We propose a set of 17 anthropometric dimensions for the size, cognition/perception action/behavior, and their relationships with human errors regarding the MCR design variables through a case study. The 17 selected dimensions are height, sitting height, eye height from floor, eye height above seat, arm length, functional reach, extended functional reach, radius reach, visual field, peripheral perception, hyperopia/myopia/astigmatism, color blindness, auditory acuity, finger dexterity, hand function, body angle, and manual muscle test. We proposed criteria on these 17 anthropometric dimensions for a suitability evaluation and suggested an improvement method according to the evaluation dimensions. Conclusion: The results of this study can improve the human performance of the crew in an MCR. These management guidelines and a procedure for an anthropometric suitability assessment will be able to prevent human errors due to inadequate anthropometric dimensions. Application: The proposed set of anthropometric dimensions can be integrated into a managerial index for the anthropometric suitability of the operating crews for more careful countermeasures to human errors in NPPs.

Quantification of unavailability caused by random failures and maintenance human errors in nuclear power plants

Risk analysis of any equipment or system estimates the unavailability of redundant components due to hardware failure, periodic test and repair work, and human errors in maintenance tasks. A model has been developed in this study to estimate the unavailability of a periodically repairable component of a system that considers hardware failure and maintenance errors due to human failure, and which is applicable to estimate system unavailability in nuclear power plants as well as other industries. A sensitivity analysis has been performed based on the newly developed model considering multiple levels of human errors in maintenance tasks to observe the effects of maintenance errors on overall unavailability of a component. The effect of human error on the optimal frequency of periodic maintenance of a device has been analyzed based on the newly developed unavailability quantification model.

A framework for evaluating the effects of maintenance-related human errors in nuclear power plants

This paper proposes a framework for estimating the qualitative and quantitative consequences of human errors that occur during maintenance tasks involving the balance of plant in nuclear power plants. One of the remedies for unexpected reactor shutdowns may be a methodical tool designed to warn potential hazards arising from given maintenance tasks, taking into account human error modes in a proactive manner, in terms of risk and/or loss of electrical power. The entire framework that we established is composed of four components: (1) the human-error analyzer to connect possible failure modes resulting from human errors with other estimators, (2) the frequency estimator to quantify the occurrence of maintenance-related failure modes, (3) the risk estimator to determine minimal cutsets and to compute the variation of the core damage frequency using the fault tree analysis and turbine cycle simulation, and (4) the derate estimator to determine the electrical power loss under abnormal plant configurations caused by human error. The final result is characterized by a cost metric that can be used for decision-making possibly resulting in revisions of procedures, or task modifications. This paper also discusses case studies to illustrate the feasibility of the proposed framework.

Measure Tables for Preventing Worker Errors in Nuclear Power Plants and Its Application

Measure Tables for Preventing Worker Errors in Nuclear Power Plants and Its Application

A model-based and computer-aided approach to analysis of human errors in nuclear power plants

Since the operator's mission in NPPs is increasingly defined by cognitive tasks such as monitoring, diagnosis and planning, the focus of human error analysis should also move from external actions to internal decision-making processes. While more elaborate analysis of cognitive aspects of human errors will help understand their causes and derive effective countermeasures, a lack of framework and an arbitrary resolution of description may hamper the effectiveness of such analysis. This paper presents new model-based schemes of event description and error classification as well as an interactive computerized support system. The schemes and the support system were produced in an effort to develop an improved version of HPES. The use of a decision-making model enables the analyst to document cognitive aspects of human performance explicitly and in a proper resolution. The stage-specific terms used in the proposed schemes make the task of characterizing human errors easier and confident for field analysts. The support system was designed to help the analyst achieve a contextually well-integrated analysis throughout the different parts of HPES.

078 Detection of measurement errors in nuclear power plants

078 Detection of measurement errors in nuclear power plants

A method for quantification of nuclear reactor seismic risk due to design and construction errors based on deficiencies identified in the past

This paper proposes a plausible methodology for quantification of risk due to seismic related design and construction errors in nuclear power plants. This is based on information available on errors discovered in the past, as reported in construction deficiency reports pursuant to US NRC regulations. Deficiencies found during construction whose recurrence is considered possible are included. The possibility of deficiencies not being discovered is evaluated by comparison of data between operating plants and those under construction. Error consequences are also evaluated from construction deficiency reports; in particular, the impact of errors on the seismic capacity of the plants is quantified by an extension of seismic probabilistic risk assessment (PRA) methods. The application of the method is illustrated based on a limited review of data, showing its capabilities and limitations. Some tentative results are presented, but these by no means represent a formal assessment.

Maintenance and Testing Errors in Nuclear Power Plants: A Preliminary Assessment

Safety and reliability in the operation of nuclear power plants have been topics for intense analysis and debate in recent years. Once crucial factor which has received little attention until very recently, is human reliability in the operation, maintenance and testing of nuclear plants. Maintenance and testing errors are of particular importance in that, despite the fact that they are relatively few in number compared with hardware related events, their occurrence programs the system for future operator errors bv introducing erroneous data in operator-hardware system communication or by eliminating a portion of that communication. Erroneous data or a lack of anticipated feedback following control actions by the operator can tend to cause further inappropriate responses and to escalate the frequency of operator errors. Such effects are most undesirable during emergencies. A number of improvements based upon improved human factors engineering will be necessary to eliminate maintenance errors of the type encountered in this study. These include: standardization of design for given types of subsystems, standardization of maintenance operations, work simplification involving both physical and psychological aspects of maintenance work, the use of training simulators to increase the amount of practice in performing maintenance operations, automanual control to show predicted outcome of results, assignment of individual responsibility and accountability to specific hardware subsystems, measures of job performance against some criterion, programmed instruction, improved divisions of responsibility among maintenance, operating and technical personnel, and improved management practices.