Nuclear power plant personnel reliability analysis method and system

Through the comprehensive personnel reliability analysis method of nuclear power plant, combined with the first model and the second model, comprehensively considering time and behavioral factors, the problem of insufficient accuracy of analysis results in the existing technology is solved, and the reliability analysis of nuclear power plant personnel is improved Accuracy and comprehensiveness.

CN118966759APending Publication Date: 2024-11-15SHANGHAI NUCLEAR ENGINEERING RESEARCH & DESIGN INSTITUTE CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411005163.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-25
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

The existing nuclear power plant personnel reliability analysis methods focus on time parameters or pressure factors, and fail to fully cover various factors affecting personnel reliability, resulting in insufficient accuracy of the analysis results.

Method used

A comprehensive nuclear power plant personnel reliability analysis method is provided, and the probability Pc1 of operators not responding in the cognitive response time window is calculated through the first model, and the probability Pc2 of diagnostic error is calculated by the second model, and the total is calculated by combining the personnel error analysis table. Probability of error HEP, taking into account time and behavioral factors in full.

Benefits of technology

It improves the accuracy and comprehensiveness of the reliability analysis of nuclear power plant personnel, can integrate the possibility of analysts' mistakes in different scenarios, and improves the level of safety evaluation of nuclear power plants.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118966759A_ABST
    Figure CN118966759A_ABST
Patent Text Reader

Abstract

The invention discloses a nuclear power plant personnel reliability analysis method. The method comprises the following steps: providing a first model for giving an unresponded probability Pc1 of an operator in a cognitive time window and a second model for giving a diagnosis error probability Pc2 of the operator; decomposing a given operation task into basic actions, providing a personnel error analysis table comprising a basic action error probability nominal value, and calculating a personnel error probability PE according to the personnel error analysis table; and calculating the total error probability HEP = Pc1 + Pc2 + PE of personnel errors. The method can effectively improve the efficiency and accuracy of nuclear power plant personnel reliability analysis, and provides reference for nuclear power plant probability safety analysis and calculation. The invention further provides a nuclear power plant personnel reliability analysis system.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the field of nuclear power, and in particular relates to a method and system for analyzing the reliability of personnel in a nuclear power plant. Background Art

[0002] According to the data of the World Association of Nuclear Operators (WANO), human error is the main cause of safety incidents and accidents in nuclear power plants. Therefore, for large and complex systems such as nuclear power plants, human reliability analysis (HRA) methods need to be used to qualitatively and quantitatively evaluate the reliability of personnel. HRA constitutes an important technical element of probabilistic safety assessment (PSA) of nuclear power plants and has an important impact on the quality of PSA assessment results. At present, some existing HRA methods for nuclear power plants have different emphases on different factors, such as focusing on the influence of relevant performance forming factors (PSF) such as time parameters or pressure factors, execution complexity, etc., and their analysis process needs to be further improved. Therefore, providing a more comprehensive method for analyzing the reliability of nuclear power plant personnel is of positive significance for improving the safety assessment level of nuclear power plants. Summary of the invention

[0003] The purpose of the present invention is to provide a method for analyzing the reliability of nuclear power plant personnel and improve the accuracy of HRA. The present invention also provides a system for analyzing the reliability of nuclear power plant personnel.

[0004] According to an embodiment of one aspect of the present invention, a method for analyzing the reliability of nuclear power plant personnel is provided, the method comprising the following steps: providing a first model and a second model, wherein the first model provides a probability P of the operator not responding within the cognitive response time window; c1 The second model provides the probability P of operator diagnosis error c2 ; Given an operation task, decompose the operation task into multiple basic actions; provide a personnel error analysis table for the operation task, the personnel error analysis table including the basic actions and the nominal values ​​of the error probabilities corresponding to the basic actions; calculate the personnel error probability P according to the personnel error analysis table E ; Calculate the total error probability HEP of the operator, HEP = P c1 +P c2 +P E .

[0005] The above method can comprehensively analyze the possibility of personnel errors in different scenarios of short-time margin and long-time margin during the execution of operation tasks, fully consider various factors affecting personnel reliability, and thus make a more reasonable evaluation of the reliability of nuclear power plant personnel.

[0006] Furthermore, in some embodiments, P c1 =1-Φ(ln(T w / T1 / 2 ) / σ), where Φ is the normal distribution function, T w is the cognitive response time window, T 1 / 2 is the median cognitive response time.

[0007] Furthermore, in some embodiments, σ, T w and T 1 / 2 Determined through accident condition statistics or simulation tests.

[0008] Furthermore, in some embodiments, P c2 Determined by: providing a distribution table of error mechanisms of personnel diagnosis errors, wherein the error mechanism distribution table includes the probability P of each error mechanism i i and the recovery factor P nr i ,

[0009] Furthermore, in some embodiments, P i The calculation method is as follows: establish a decision tree for the error mechanism i, and use the decision tree to calculate P i .

[0010] Furthermore, in some embodiments, the error mechanisms include: unavailability of information, failure to pay attention to information, misinterpretation or mistransmission of data, misleading information, omission of steps in procedures, misunderstanding of instructions, misunderstanding of decision logic, and intentional violation of procedures.

[0011] Furthermore, in some embodiments, the personnel error analysis table also includes factors affecting personnel behavior and recovery factors. E During calculation, the nominal value of the error probability is corrected according to the influencing factors of the personnel behavior and the recovery factor.

[0012] Furthermore, in some embodiments, the nominal value of the error probability adopts an average error probability obtained by statistics.

[0013] According to an embodiment of another aspect of the present invention, there is provided a nuclear power plant personnel reliability analysis system, the system comprising a memory and a processor, wherein the memory stores a calculation program, and when the calculation program is executed by the processor, the nuclear power plant personnel reliability analysis method provided in any of the aforementioned embodiments can be implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 It is a calculation flow chart of a method for analyzing the reliability of nuclear power plant personnel in one embodiment;

[0015] Figure 2 A schematic diagram of a decision tree where information is unavailable;

[0016] Figure 3 Schematic diagram of a decision tree for failure to pay attention to information;

[0017] Figure 4 Decision tree diagram for misinterpretation or misunderstanding of data;

[0018] Figure 5 A diagram of a decision tree with misleading information;

[0019] Figure 6 Schematic diagram of the decision tree for missing steps in the procedure;

[0020] Figure 7 Schematic diagram of the decision tree for misunderstood instructions;

[0021] Figure 8 A diagram of a decision tree that misunderstands the decision logic;

[0022] Fig. 9 Schematic diagram of the decision tree for intentional violation of regulations.

[0023] The purpose of the above drawings is to explain the present invention in detail so that those skilled in the art can understand the technical concept of the present invention, but it is not intended to limit the present invention. DETAILED DESCRIPTION

[0024] The present invention is further described in detail below through specific embodiments in conjunction with the accompanying drawings.

[0025] Mentioning "embodiment" herein means that the specific features, structures or characteristics described in conjunction with the embodiment may be included in at least one embodiment of the present invention. The phrase appearing in various locations in the specification does not necessarily refer to the same embodiment, nor is it limited to mutually exclusive independent or alternative embodiments. It should be understood by those skilled in the art that the embodiments herein can be combined with other embodiments without structural conflicts. In the description herein, terms such as "first" and "second" are only used to distinguish different objects, and cannot be understood as indicating relative importance or limiting the number, specific order or primary and secondary relationship of the described technical features. In the description herein, the meaning of "multiple" is at least two.

[0026] According to the operational event analysis reports collected by the World Association of Nuclear Operators (WANO) in recent years, human error is the main cause of safety incidents and accidents in nuclear power plants. Therefore, the reliability of researchers' behavior is crucial to ensuring nuclear power safety. Human reliability analysis (HRA) is an important technical element of probabilistic safety assessment (PSA) of nuclear power plants. The reliability of HRA results directly affects the quality of PSA assessment results. At present, there are many HRA methods available internationally, but these methods have different development backgrounds, scopes of application, and focuses. Some methods focus on the influence of time parameters on the probability of human behavior errors, while some methods focus on the influence of relevant performance forming factors (PSFs) such as pressure factors and execution complexity. However, most of the HRA methods currently used in China have yet to be improved in the analysis and consideration of factors affecting human behavior, which will affect the accuracy of HRA analysis results.

[0027] In order to solve the above problems, an embodiment of one aspect of the present invention provides a comprehensive nuclear power plant personnel reliability analysis method, which integrates the diagnosis process and execution process of nuclear power plant personnel error events to quantitatively calculate the probability of personnel error. In this method, personnel behavior is divided into two stages: diagnosis and execution, and personnel error events are also divided into two categories: diagnosis errors and execution errors. When an accident occurs in a nuclear power plant, if the operator fails to make a correct decision within the available time window (i.e., fails to start a correct response), it is a diagnostic error, and the probability of diagnostic failure is recorded as P. c When the operator makes a correct diagnosis but fails to complete the action in the execution phase, it is an execution error. The probability of execution error is recorded as P E When the operator's diagnosis and execution are both successful, the operator's operation task is determined to be valid.

[0028] The steps of the method include:

[0029] First, the first model and the second model are established, and the first model and the second model are used together to calculate the diagnosis failure probability P c , where the first model includes the probability P that the operator does not respond within the cognitive response time window c1 , the second model includes the probability of operator diagnosis error P c2 , probability of diagnosis failure P c =P c1 +P c2 .

[0030] The first diagnostic model is based on the operator's cognitive response time window T after the accident. w and median cognitive response time T 1 / 2 The ratio of the operator to the total event time window T swThe probability of not being able to respond within . The total event time window T sw It refers to the time from the occurrence of the accident and the transmission of the accident signal to the occurrence of undesirable accident consequences (such as core melting) without the intervention of the operator. The time taken by the operator from the transmission of the signal to the start of the correct response is T. This process is the operator's recognition and diagnosis process of the signal, which includes the time when the operator does not make any response, the time when the correct response is made, and if an incorrect response is made, it also includes the process from making an incorrect response to discovering and correcting the response. The operator is in the cognitive response time T w The probability of not responding within c1 =P(T>T w ). In a preferred embodiment, to calculate P c1 , assuming that the response time T is subject to the parameter T 1 / 2 and the standard deviation of the normal distribution of σ, using T w / T 1 / 2 After standardizing the time window, we can get P c1 =1-Φ(ln(T w / T 1 / 2 ) / σ). T w / T 1 / 2 The larger the value of , the more time the operator has available for diagnosis, and the higher the probability that the operator can complete the correct diagnosis. Conversely, the higher the probability of non-response. σ reflects the differences in operator diagnosis time, which can be determined by statistical or simulation tests on actual accident reports; T 1 / 2 、T w With T sw It can be determined based on the analysis and statistics of the accident conditions, simulation calculations or simulation tests. The simulation tests can be completed by operators simulating operations in the actual site or in a simulated environment; or T can be determined sw Subtract the signal transmission delay to get T w .

[0031] The second diagnostic model is obtained by calculating and adding the product of the probability and recovery factor of each error mechanism in the error mechanism distribution table of personnel diagnosis errors, and is used for scenario analysis with a large time margin as a supplement to the first diagnostic model. In a preferred embodiment, the personnel diagnosis error mechanism can be divided into two modes: power plant information-operator interface error and procedure-operator interface error, and further divided into eight error mechanisms: information cannot be obtained, failure to pay attention to information, data misreading or mistransmission, misleading information, omission of steps in the procedure, misunderstanding of instructions, misunderstanding of decision logic, and intentional violation of procedures. Each error mechanism is represented by a decision tree, and the branch path of the decision tree is fixed. Each decision tree covers the corresponding aspects of the cognitive behavior of the personnel. The title of the decision tree represents the influencing factor of the personnel error, and the bifurcation represents the affirmation or negation of the corresponding title. Each branch path corresponds to a probability value of personnel error. The path selection of the decision tree can be obtained based on the interview with the operator, and can be adjusted and corrected according to the specific accident conditions. For each error mechanism, the corresponding probability P is provided in the error mechanism distribution table. i and P nr i Recovery factor refers to the possibility that an operator can discover and correct an error by himself or others.

[0032] Specifically, the failure mechanisms include:

[0033] Information not availableP ca , its decision tree is as follows Figure 2 As shown in the figure, the decision tree only shows the part of the decision tree where information can be obtained inside the main control room. The information sources outside the main control room need to be analyzed separately. Information cannot be obtained when operators cannot obtain clear and accurate information. Information specifically refers to the information necessary for operators to accurately diagnose accidents. Its specific form varies according to different accidents, mainly signals, alarms or power plant parameters.

[0034] Failure to pay attention to information P cb , its decision tree is as follows Figure 3 Failure to pay attention to information means that the operator does not notice a certain signal, parameter or alarm, which is important for the operator to diagnose the accident. For example, the operator forgets to read the data of a certain instrument or display, or the operator is distracted by other signals before reading the data, or the operator does not monitor a certain instrument parameter at the required frequency and fails to discover the abnormal situation in time.

[0035] Misinterpretation or miscommunication of data cc , its decision tree is as follows Figure 4As shown. Data misreading or mistransmission means that the operator selects the wrong indicator, or makes an error when reading data from the indicator, or an error in communication between operators, resulting in the operator receiving incorrect data. Data misreading or mistransmission is related to the design of the control room panel, the layout of the indicators, and the communication method between operators (such as whether a standardized communication process is used to transmit data information).

[0036] Information is misleading cd , its decision tree is as follows Figure 5 Misleading information means that some information seems to be available, but will lead to wrong conclusions if interpreted in a normal way.

[0037] Missing step P in the procedure ce , its decision tree is as follows Figure 6 Missing steps in the procedure means that the operator fails to complete all steps according to the procedure. The influencing factors include heavy workload (such as the operator needs to perform multiple procedures at the same time) or the printing characteristics of the procedure.

[0038] Misunderstanding instruction P cf , its decision tree is as follows Figure 7 Factors that influence misunderstanding of instructions include incomplete or inaccurate description of procedures, unclear operating instructions, and operators not noticing the specific differences between similar operating steps.

[0039] Misunderstanding decision logic cg , its decision tree is as follows Figure 8 As shown in Figure 2, factors that influence misunderstanding of decision logic include the complexity of the operating instructions in the procedures. A feasible criterion for measuring the complexity of operating instructions is to determine the amount of information that the operator needs to consider (one signal or multiple signals) and the combination relationship between the information (AND, OR, NOT and their combination, etc.).

[0040] Intentional violation of regulations ch , its decision tree is as follows Fig. 9 An intentional violation of procedures is when an operator consciously decides not to follow the procedures, usually when the operator believes that their own approach is better than the procedures.

[0041] All decision trees are considered as a whole, and the same factor is only considered once when calculating the probability, and no repeated calculation is performed. For example, the operator load factor is only considered once in P ce It is calculated once in the decision tree and is not repeated in other decision trees.

[0042] The recovery factor of each failure mechanism includes five factors: self-review, additional team, STA review, shift review, and emergency response organization. The above eight failure mechanisms and recovery factors are included in the failure mechanism distribution table.

[0043] In one embodiment, the failure mechanism distribution table is shown in Table 1, where each decision tree and its branch probability are shown in Table 1. Figures 2 to 9 .

[0044]

[0045] Table 1 Failure mechanism distribution table

[0046] Among them, the recovery effect of STA on each team is only considered during the first review; None means that no recovery factor is considered; X means that its specific value needs to be determined based on the correlation between the reviewer and the operator. For the case of zero correlation, the recovery factor uses the error probability itself.

[0047] Next, the given operation task (preset scenario for personnel reliability analysis) is decomposed, and the operation task is divided into multiple subtasks and further decomposed into basic actions of the operator. Basic actions are independent and non-interfering behavior units that can be probability analyzed, such as starting a device or closing a valve. The decomposition of the operation task is based on the specific conditions of the power plant main control room design and related regulations. According to the regulations, the part that requires human intervention in the operation process is divided out, and it is screened and simplified according to the human-computer interaction information. The basic actions of the operation task are used as statistical items to formulate a personnel error analysis table, which includes each basic action and its corresponding error probability nominal value.

[0048] In a preferred embodiment, the personnel error analysis table also includes influencing factors and recovery factors on personnel behavior, wherein the influencing factors include stress, skill level, fatigue level, working environment and organizational influences, etc. The nominal value of the probability of personnel error usually considers personnel behavior as an isolated event. The nominal value is usually derived from HRA basic research data (experimental values, power plant statistical data, etc.), and usually does not consider the impact of other factors on personnel behavior. Therefore, in a preferred embodiment, it is necessary to correct the nominal value of the probability of personnel error based on the influencing factors and recovery factors. The specific correction calculation method is to multiply the nominal value of the probability of personnel error that needs to be corrected by all the influencing factors and recovery factors that need to be considered (that is, the performance formation factor PSF).

[0049] In one embodiment, the human error analysis table is shown in Table 2.

[0050]

[0051] Table 2 Personnel error analysis table

[0052] Finally, according to Figure 1 As shown, HEP=P c1 +P c2 +P EThe calculation results shall be recorded in documents or entered into a database for subsequent PSA analysis and calculation.

[0053] In a preferred embodiment, the above process can be written as a computing program with the aid of a general-purpose computer or a special computing device and stored in the memory of the computing device. When the computing program is executed by the processor of the computing device, the personnel reliability analysis method provided in any of the above embodiments can be automatically executed. The required input data can be pre-entered into the system or manually input at the corresponding computing node during the program operation.

[0054] In the nuclear power plant personnel reliability method provided by the embodiment of the present invention, the first model classifies the personnel error time, covering information such as the total task time window, diagnostic time margin, execution time and signal delay, which can more comprehensively and reasonably reflect the impact of time parameter changes on the probability of personnel errors; the second model fully considers the interaction between the operator and the power plant information and procedures, and complements the first model, which can improve the deficiency of a single model in not fully covering the factors affecting personnel behavior; the method also comprehensively considers various error mechanisms, and can comprehensively and comprehensively reflect the operating characteristics of different power plants. The implementation process of this method is concise and clear, and has good operability. It can effectively improve the efficiency, accuracy and reliability of nuclear power plant personnel reliability analysis, and provide a reference for the probabilistic safety evaluation of nuclear power plants.

[0055] The purpose of the above embodiments is to further explain the present invention in detail in conjunction with the accompanying drawings so that those skilled in the art can further understand the technical concept of the present invention. Within the scope of the present invention, the optimization or equivalent replacement of the method steps involved, and the combination of the implementation methods in different embodiments without conflict of structure and principle, all fall within the protection scope of the present invention.

Claims

1. A method for analyzing the reliability of personnel in a nuclear power plant, characterized in that: The following steps are involved: A first model and a second model are provided, wherein the first model provides a probability P of the operator not responding within the cognitive response time window. c1 The second model provides the probability P of operator diagnosis error c2 ; Given an operation task, decomposing the operation task into multiple basic actions; Providing a personnel error analysis table for the operation task, the personnel error analysis table including the basic actions and nominal values ​​of error probabilities corresponding to the basic actions; Calculate the probability of human error P of the operation task according to the human error analysis table E ; Calculate the total error probability HEP of the operator, HEP = P c1 +P c2 +P E .

2. The method for analyzing the reliability of nuclear power plant personnel according to claim 1, characterized in that: P c1 =1-Φ(ln(T w / T 1 / 2 ) / σ), where Φ is the normal distribution function, T w is the cognitive response time window, T 1 / 2 is the median cognitive response time, and σ is the standard deviation.

3. The method for analyzing the reliability of nuclear power plant personnel according to claim 2, characterized in that: σ、T w and T 1 / 2 Determined through accident condition statistics or simulation tests.

4. The method for analyzing the reliability of nuclear power plant personnel according to claim 1, characterized in that: P c2 Determined by: providing a distribution table of error mechanisms of personnel diagnosis errors, wherein the error mechanism distribution table includes the probability P of each error mechanism i i and the recovery factor P nr i , 5. The method for analyzing the reliability of nuclear power plant personnel according to claim 4, characterized in that: P i The calculation method is as follows: establish a decision tree for the error mechanism i, and use the decision tree to calculate P i .

6. The method for analyzing the reliability of nuclear power plant personnel according to claim 4 or 5, characterized in that: The failure mechanisms described include: unavailability of information, failure to pay attention to information, misinterpretation or miscommunication of data, misleading information, omission of steps in procedures, misunderstanding of instructions, misunderstanding of decision logic, and intentional violation of procedures.

7. The method for analyzing the reliability of nuclear power plant personnel according to claim 1, characterized in that: The personnel error analysis table also includes the factors affecting personnel behavior and the recovery factors. E During calculation, the nominal value of the error probability is corrected according to the influencing factors of the personnel behavior and the recovery factor.

8. The method for analyzing the reliability of nuclear power plant personnel according to claim 7, characterized in that: The nominal value of the error probability adopts the average error probability obtained by statistics.

9. A nuclear power plant personnel reliability analysis system, comprising a memory and a processor, characterized in that: The memory stores a calculation program, and when the calculation program is executed by the processor, the nuclear power plant personnel reliability analysis method according to any one of claims 1 to 8 can be implemented.