WO2014173270A1

WO2014173270A1 - Human-machine interface detection method and system

Info

Publication number: WO2014173270A1
Application number: PCT/CN2014/075823
Authority: WO
Inventors: 张力; 李鹏程; 戴立操; 胡鸿; 黄卫刚; 戴忠华; 邹衍华; 王春辉; 苏德颂; 李晓蔚
Original assignee: 湖南工学院; 南华大学; 中广核核电运营有限公司; 大亚湾核电运营管理有限责任公司
Priority date: 2013-04-23
Filing date: 2014-04-21
Publication date: 2014-10-30
Also published as: CN103198230A; CN103198230B

Abstract

The present invention discloses a human-machine interface detection method and system, said human-machine interface detection method including the following steps: identifying, on the basis of an HRC method, risk scenarios for a probability of no response to actions exceeding a predetermined threshold; using a CREAM method in regard to the identified risk scenarios in order to determine the probability of failure, within a risk scenario, of each task or of each action; checking, in order of the probability of failure, a human-machine interface and preset human factor engineering checklist affected by each task or each action, so as to determine whether any human-machine interface units are flawed. The present invention uses a combination of HRC, CREAM and HEC methods to evaluate digitized human-machine interfaces of nuclear power plants, in order to quickly and efficiently detect flawed human-machine interface units within digitized human-machine interfaces, and to find deep-seated design flaws leading to human error, thus improving as a whole the safety of operations of nuclear power plant digital control systems.

Description

Human-machine interface detection; ^ and system

TECHNICAL FIELD The present invention relates to the field of digital control of nuclear power plants, and in particular, to a human-machine interface detection method and system for digital control of a nuclear power plant. BACKGROUND OF THE INVENTION Nuclear power plants (NPPs) are complex high-risk systems. Due to the Sancha Island and Chernobyl accidents, people recognize the importance of human-machine interaction, Man-machine interface (Man-machine). Interface, MMI) The quality of the design is critical to human performance and nuclear safety. For example, Smith and Mosier pointed out in 1986 (Guidelines for designing user interface software) that data processing from poor human-machine interfaces can lead to frequent and serious human error; Bellamy and Geyer were in 1988. Addressing human factors issues in the safe design and operation of computer controlled process systems)) also investigates that nearly 60% of events occurring in computer-controlled process systems are related to human error occurring during operation. Despite spending a lot of manpower and material resources to carry out research to improve the design quality of human-machine interface, the human-machine interface design and human-machine interaction technology of nuclear power plants in China have been in the early stage of development, one learning and absorption stage, SP" The design is ready to use, and the detailed review of the human-machine interface has been in the development stage. Especially after digitizing the nuclear power plant's main control room, the digital human-machine interface changes the situation environment in which the operator is located. If the designer does not understand the adverse effects of the digital human-machine interface characteristics on the operator, then the designer is in the person. - The design process of the machine interface follows the general design principles and references some human factor engineering (HFE) design guidelines, but they may overlook some deeper performance problems or ignore the manipulation in the design. The role of the personnel in complex human-computer interactions is not sufficient to ensure that personnel performance and operating system security can be improved by reference to general human factors engineering design guidelines. Similarly, although there are some human-machine interface evaluation methods, the principles, methods, styles, cultures, etc. of digital human-machine interfaces designed by countries are different. Therefore, combined with the human-machine interface design features and Chinese cultural characteristics, we need The development of digital human-machine interface evaluation methods and tools to review the human-machine interface to improve the performance of operators and the safety level of nuclear power plants. Man-machine interface evaluation methods can generally be divided into experimental methods and theoretical methods. The experimental method generally evaluates the availability of the human-machine interface in the control room through simulators and simulation experiments, such as Paulo VR Carvalho, Isaac L. dos Santos, Jose Orlando Gomes, Marcos RS Borges and Stephanie Guerlain. Human Factors approach for evaluation and redesign of ¹ J ( DISPLAY) Human-system interfaces of a nuclear power plant simulator ^■ The use of small simulator experiments in the text to review the advanced man-machine interface design through on-site observation to identify specific defects in the design. Chen Xiaoming, Gao Zuyu, Zhou Zhiwei, Zhao Bingquan of Tsinghua University's Nuclear Energy Technology Design and Research Institute and Nakagawa Takashi of Japan's Mitsubishi Electric Advanced Technology Research Institute and Converse Co., Ltd. published in the journal "Atomic Energy Science and Technology" in 2004. The human-machine interface evaluation system of technology introduces a DIAS (Dynamic Interaction Analysis Support) software evaluation system for nuclear power plant control room based on computer simulation technology, which is used as the main control room of the nuclear power plant. The design and improvement of the machine interface provides good technical support. Although the experimental and simulation methods are one of the most accurate human-machine interface evaluation methods, it is difficult to be consistent with the real risk scenario. Only a few behavioral influence factors can be considered, and the experimental methods are time-consuming, and the results often depend on The experience of the participants requires a lot of money. Theoretical methods generally include expert judgment, human-machine interface design guidelines, human-machine interface evaluation models and methods, etc. The theoretical method allows system designers to estimate the impact of human-machine interface on operator performance without experimentation. This method is effective, especially in the early stages of design. Methods based on expert judgment, design guidelines, and evaluation models have their own advantages and disadvantages. Among them: The shortcomings of the theoretical methods based on expert judgment are mainly subjective judgments of experts, and the evaluation results are difficult to quantify. Such as

Sheue-Ling Hwang, Sheau-FarnMax Liang, Tzu-Yi Yeh Liu, Yi-Jhen Yang, Po-Yi Chen and Chang-Fu Chuang published in "Nuclearation Engineering and Design" in 2009 (Evaluation of human factors in Interface design in main control rooms" Zhongmi interviewed the human-machine interface, Yung-Tsan Jou, Chiuhsiang Joe Lin, Tzu-Chung Yenn, Chi-Wei Yang, Li-Chen Yang and Ruei-Chi Tsai in 2009 The "The implementation of a human factors engineering checklist for human-system interfaces upgrade in nuclear power plants", published in "Safety Science"), establishes a mixed human factors engineering checklist to review the human-machine interface design. However, the above methods lack systematic and comprehensive considerations from operator information processing and operation, so that people's performance is improved in some aspects (such as monitoring and data acquisition), and in other aspects (such as operational control behavior). Declining, did not achieve the best overall performance. In the human-machine interface design guide, in 1995, the US Nuclear Regulatory Commission (USNRC) established in the Human-System Interface Design Review Guidelines from information display, user-interface interaction and management, and control. A specific digital human-machine interface review project. In 2002, the US Nuclear Regulatory Commission updated the Human-System Interface Design Review Guidelines. Similarly, in 2004, the Electric Power Research Institute (EPRI) in the "Human Factors Guidance for Control Room and Digital Human-System Interface Design and Modification") from information display, user-interface interaction and management, soft control, Alarms, computerized procedures systems, computerized operator support systems, etc., have presented specific review items. Based on the design guidelines The subjectivity of the management has been reduced, but due to the emergence of new technologies or new states, it takes a lot of time to update the design guide, and due to various trade-offs, there may be conflicting guiding rules, resulting in design guidelines. Misinterpretation and error application. In terms of human-machine interface evaluation models and methods, in order to overcome the limitations of expert judgment methods, some methods and models based on mathematical statistics techniques have been developed, such as regression analysis models, fuzzy comprehensive evaluation methods, grey relational analysis, neural networks. Methods such as quantitative prediction and evaluation. For example, in 2006, Jiang Tao used the gray theory to quantitatively and comprehensively evaluate the human-machine interface of DCS (Digital Control System) system in his doctoral thesis "Comprehensive Evaluation of Human-machine Interface of DCS System in Thermal Power Plants". However, the above method overemphasizes the quantitative evaluation results, but fails to examine the digital human-machine interface features in detail from the perspective of human reliability. It is difficult to find the deep-rooted causes of human error in the human-machine interface. In short, with the digitization of the main control room of nuclear power plants in China, the traditional human-machine interface evaluation method is difficult to meet the requirements of analysis and evaluation, and the human factor engineering checklist for review is not established from the characteristics of the digital human-machine interface. The human factors engineering checklist has not been comprehensively established from the perspective of human reliability and personnel performance, and the existing human-machine interface evaluation method is inefficient and reliable. SUMMARY OF THE INVENTION It is an object of the present invention to provide a human-machine interface detection method for solving the technical problem of quickly and reliably identifying an interface item having a defect in a human-machine interface. Another object of the present invention is to provide a human-machine interface detecting system for quickly and reliably detecting a technical problem of a defective interface item in a human-machine interface. In order to achieve the above object, the technical solution adopted by the present invention is as follows: A human-machine interface detection method, applicable to a nuclear power plant digital control system, comprising the following steps: Identifying a risk that the non-response probability of operator behavior exceeds a predetermined threshold based on the HCR method a scenario; determining, by using a CREAM method, the probability of failure of each task in the risk scenario and each behavior in the task for the identified risk scenario; and sequentially, for each task or each person involved in each behavior according to the size of the error probability The machine interface is checked against a pre-set human factor checklist to determine the human-machine interface unit with defects. Further, the formula used by the HCR method to calculate the non-response probability is as follows:

Where t is the available time to complete the task, T _{1 / 2} is the median response time for the operator to complete the task, α is the scale parameter, β is the shape parameter, ^ is the position parameter, and P(t) is the non-response probability. Further, the CREAM method specifically includes the following steps: analyzing, by using a Hierarchical Task Analysis (HTA) method, each task and each behavior of the identified risk scene to construct an event sequence; and analyzing the context according to the context (Context) Analysis) or common performance conditions (CPC), Cognitive activity analysis to determine the human error model corresponding to the sequence of events; according to the human error model in the CREAM method The corresponding basic error probability determines the first human cause error probability of each task or each behavior, and corrects the first human factor error probability by the context environment analysis, and obtains the corrected second corresponding to each task or each behavior. Human error probability; Sort each task or behavior according to the probability of the second person due to error. Further, the human factor engineering checklist is established according to an information processing process of the digital control system of the nuclear power plant, and the specific considerations include: information monitoring, state evaluation, response planning, and response execution. Further, the human-machine interface unit specifically includes: an information display module, a soft control module, a digitization procedure module, a user interface management module, and an alarm module. According to another aspect of the present invention, a human-machine interface detecting system is further provided, which is applicable to a nuclear power plant digital control system, and the human-machine interface detecting system comprises: a risk scene screening unit, which identifies an operator behavior based on the HCR method. a risk scenario in which the response probability exceeds a predetermined threshold; a human factor reliability analysis unit uses a CREAM method to determine a probability of error in each task and task in the risk scenario for the identified risk scenario; a human factor engineering review unit, The human-machine interface involved in each task or each behavior is sequentially checked against the preset human factors checklist according to the magnitude of the error probability to determine the human-machine interface unit with defects. Further, the human factor reliability analysis unit specifically includes: an event sequence construction module, configured to analyze each task and each behavior of the identified risk scene by using a hierarchical task analysis method to construct an event sequence; a human error mode a determining module, configured to determine a human error mode corresponding to the event sequence according to the situation environment analysis and the cognitive behavior analysis; the error probability determination module determines the human error mode classification and the corresponding basic error probability according to the CREAM method The first person of each task or each behavior is corrected for the probability of error, and the first human error probability is corrected by the contextual environment analysis, and the corrected second human error probability corresponding to each task or each behavior is obtained; , sorting the tasks or behaviors according to the second person's probability of error. Further, the human factor engineering checklist is established according to an information processing process of the digital control system of the nuclear power plant, and the specific considerations include: information monitoring, state evaluation, response planning, and response execution. Further, the human-machine interface unit specifically includes: an information display module, a soft control module, a digitization procedure module, a user interface management module, and an alarm module. The invention has the following beneficial effects: The inventor-machine interface detecting method adopts a combination of HCR+CREAM+HEC to evaluate the digital human-machine interface of the nuclear power plant, so as to quickly and efficiently detect the existence of the digital human-machine interface. Defective human-machine interface unit. Among them, the HCR method is used to identify important and risky risk scenarios; the CREAM method fully considers the cognitive and contextual environmental impacts to identify human error with high probability of error; further, combined with HEC, from human reliability And the perspective of personnel performance to establish a human factors engineering review form, to quickly and reliably verify the human error caused by the high probability of error or the human-machine interface unit involved in the task, in order to find deep design defects that induce human error, thereby Improve the safety of the operation of the digital control system of the nuclear power plant as a whole. The human-machine interface detection system of the present invention comprises a risk scene screening unit, a human factor reliability analysis unit and a human factors engineering review unit, wherein the risk scene screening unit is used to identify an important and risky risk scene; The sexual analysis unit fully considers the cognitive and contextual environmental impacts to identify human error with high probability of error; further, the human factors engineering review unit establishes the human factors engineering review form from the perspective of human reliability and personnel performance, and makes mistakes. High probability probabilities are quickly and reliably verified by mistakes or human-machine interface units involved in the mission to find deep design flaws that induce human error, thereby improving the overall safety of the nuclear power plant digital control system. In addition to the objects, features and advantages described above, the present invention has other objects, features and advantages. The invention will now be described in further detail with reference to the drawings. BRIEF DESCRIPTION OF THE DRAWINGS The accompanying drawings, which are incorporated in FIG. BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a schematic diagram showing the steps of a human-machine interface detecting method according to a preferred embodiment of the present invention; FIG. 2 is a schematic diagram showing a preferred step of step S20 of FIG. 1. FIG. 3 is a human-machine interface of a preferred embodiment of the present invention. A schematic block diagram of the detection system. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The embodiments of the present invention are described in detail below with reference to the accompanying drawings. Explanation of terms:

HCR: Human Cognitive Reliability;

CREAM: Cognitive Reliability and Error Analysis Method;

HEC: Human Engineering Checklist. The invention relates to a human-machine interface detecting method, which is suitable for a nuclear power plant digital control system. Since the nuclear power plant adopts digital control, there are a large number of human-machine interfaces. The inventor-machine interface detection method detects the human-machine interface through a combination of HCR, CREAM and HEC methods, thereby efficiently and reliably finding the induced Defects in human-machine interface due to human error. Referring to FIG. 1 , the method for detecting a human-machine interface of the present invention specifically includes the following steps: Step S10: Identify, according to the HCR method, a risk scenario in which the probability of non-response of the operator behavior exceeds a predetermined threshold, so that a risk scenario with high risk can be selected. Targeted human-machine interface detection. Step S20: The CREAM method is used to determine the error probability of each behavior in each task and task in the risk scenario, and the risk scenario exceeding the predetermined threshold is used as a human factor reliability analysis. In order to identify the task or behavior of a high probability of human error and a high contribution rate to the accident risk, in order to further conduct risk assessment. In step S30, the human-machine interface involved in each task or each behavior is sequentially checked against the preset human factors checklist according to the magnitude of the error probability to determine the human-machine interface unit with the defect. Preferably, the HCR method calculates the non-response probability in step S10 by using the following formula:

In formula (1), t is the available time to complete the task, T _1/2 is the median response time of the operator to complete the task, α is the scale parameter, β is the shape parameter, ^ is the position parameter, P(t) is Does not respond to probability.

T _1/2 = Τ _1/2 , _η (1 + Κ ₁ )(1 + Κ ₂ )(1 + Κ ₃ ) (2) In the formula (2), Τ _{1/2 η} is a general condition (such as simulation The execution time of the machine training, Κ ₂ , Κ ₃ respectively represent the "training" correction factor, the "psychological pressure" correction factor and the "human-machine interface" correction factor. In practice, different parameters are obtained by identifying the available time t for the operator to complete the task, and the behavior type of the operator in different scenarios, and evaluating the state level of the behavior forming factor in the HCR model, and obtaining different correction factors, and By substituting each parameter value into the HCR model, the error probability of the operation team can be obtained, thereby identifying a risk scenario in which the operator does not respond with high probability, so as to further analyze the key risk scenarios. Preferably, in step S20, referring to FIG. 2, the CREAM method specifically includes the following steps: Step S21, using a Hierarchical Task Analysis (HTA) method to analyze each task and each behavior of the identified risk scene to construct an event. Sequence S22, determining a corresponding human error pattern of the event sequence according to Context analysis or Common Performance Condition Assessment (CPC) and Cognitive Activity Analysis; the situational environment analysis is a common performance condition (common performance condition, CPC) As the analysis object, a quantization level is proposed for each CPC. Cognitive behavior analysis is to analyze the specific cognitive behaviors required by each task. For example, the specific cognitive behaviors in practice include: coordinate, communication, comparison, diagnosis, evaluation ( evaluate ) , execute , identify , maintain , monitor , observe , plan , record , adjust , scan , confirm (verify) and so on. After confirming the recognition of the task After knowing the behavior, according to the correspondence between cognitive behavior and cognitive function and the situational environment analysis, the possible human failure mode can be predicted. Step S23, determining a first human cause error probability of each task or each behavior according to a human error mode and a corresponding basic human error mode probability, and correcting the first human cause error probability according to the situation environment analysis, and obtaining each The corrected second human factor error probability corresponding to the task or each behavior; the correction of the first human factor error probability is to consider the weight of CPC influence on cognitive behavior, and the weight factor of CPC that has no influence on cognitive behavior is 1 Considering the weight of each CPC's influence on cognitive behavior, the correctness of the first person's error probability is corrected by the method of multiplication, and the corrected second human factor error probability is obtained. In step S24, each task or each behavior is sorted according to the second person's probability of error to identify a key task or behavior. Preferably, the human factor engineering check table preset in step S30 is established according to the information processing process of the nuclear power plant digital control system, and the specific information processing behavior includes: information monitoring, status evaluation, response planning, and response execution. The human-machine interface unit specifically includes: an information display module, a soft control module, a digital procedure module, a user interface management module, and an alarm module. The following is an example of a false alarm scene to analyze the probability of non-response. According to formula (1) and formula (2), collect relevant data. See Table 1 for details: Table 1 Data sheet for scene collection of false alarms

Man-machine interface: The K ₃ value corresponding to the human-machine interface design is equal to 0. 44 Considering the human factors engineering problem, but the test

Insufficient and need an operator

Integrate some of the information, so the evaluation is

Generally, the available time for completing the task. In the case of the accidental safety note, as long as the operator who has calculated the task by the thermal hydraulics stops the pump, the available time is 20 minutes, and the event is considered to be relieved.

Open the safety valve. The median time to actually complete the task. Interview with the operator and get good performance. Therefore, it is assumed that the operator in the completed task will complete the value in 10 minutes for 10 minutes.

Mission, poor performance will be at 13

Complete the task in minutes or so, averaging

About 10 or so

Substituting the value of Κ ₃ into the formula (2) gives:

Τ _1/2 =10x1.8432=18.432 minutes; Substituting α=0.601, β=0.9, _r =0.6, t=20, T _1/2 =18.432 into formula (1) can obtain the non-response probability of the scene : P(t) = 0.4384. Due to the high probability of non-response to the scene, it is necessary to optimize the relevant human-machine interface in the scene of the error. Further, the pre-execution behavior (PRE-ACT) in the scenario of false alarm is used to determine the probability of error in the behavior of the task. After analyzing, the key human error tasks are: "13 interpretation delay" and " 02 error identification "etc. The key task order is "REA 503KA alarm appears?", "REA404KA alarm appears?", "Is there any chemical injection in this process?" Therefore, it is possible to focus on these key or important human error or task analysis, and focus on the human error caused by the human-machine interface, save resources, and achieve targeted. Further, the human-machine engineering checklist is used to detect the human-machine interface involved in the behavior with high probability of error, thereby identifying specific defects and making corresponding suggestions to find out the deep-level person who induces human error. - Defects in machine interface design. Referring to FIG. 3, the human-machine interface detecting system of the present invention is applicable to a digital control system of a nuclear power plant, and the human-machine interface detecting system includes: The risk scene screening unit identifies a risk scenario in which the non-response probability of the operator behavior exceeds a predetermined threshold based on the HCR method; the human factor reliability analysis unit uses the CREAM method to determine each task in the risk scenario for the identified risk scenario And the probability of error in each activity in the task; the human factors engineering review unit checks the human-machine interface involved in each task or each behavior in sequence with the pre-set human factors checklist according to the order of the probability of error. A human-machine interface unit with defects. The risk scene screening unit, the human factor reliability analysis unit, and the human factors engineering review unit may be operated by the terminal processor. Further, the human factor reliability analysis unit specifically includes: an event sequence construction module, configured to analyze each task and each behavior of the identified risk scene by using a hierarchical task analysis method to construct an event sequence; a human error mode a determining module, configured to determine a human error mode corresponding to the event sequence according to the situation environment analysis and the cognitive behavior analysis; the error probability determination module determines the human error mode classification and the corresponding basic error probability according to the CREAM method The first person of each task or each behavior is corrected for the probability of error, and the first human error probability is corrected by the contextual environment analysis, and the corrected second human error probability corresponding to each task or each behavior is obtained; , sorting each task or behavior according to the second person's probability of error. The event sequence building module, the human error mode determining module, the error probability determining module, and the sorting module may be run by the terminal processor. Further, the human factor engineering checklist is established according to the operator information processing process of the digital control system of the nuclear power plant, and the specific information processing behaviors include: information monitoring, state evaluation, response planning, and response execution. The person was established from the perspective of human reliability and personnel performance due to the engineering checklist. It comprehensively considered the characteristics of the digital human-machine interface and the general human factors engineering principle, and served as a human factor review service for the digital human-machine interface. The human-machine interface unit specifically includes: an information display module, a soft control module, a digital procedure module, a user interface management module, and an alarm module, and each of the above modules can be operated by a terminal processor. The human-machine interface detecting method and system of the invention adopts a combination of HCR+CREAM+HEC to evaluate the digital human-machine interface of the nuclear power plant, and quickly and efficiently detect the human-machine with defects in the digital human-machine interface. Interface unit. Among them, the HCR method is used to identify important and risky risk scenarios; CREAM side The law fully considers the cognitive and contextual environmental impacts to identify tasks or behaviors with high probability of error; further, combined with HEC, establishes a human factors engineering review table from the perspective of human reliability and personnel performance, and performs man-machine interface units. Quick and reliable verification to find deep design flaws that induce human error, thereby improving the safety of the operation of the digital control system of the nuclear power plant as a whole. The above is only the preferred embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes can be made to the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and scope of the present invention are intended to be included within the scope of the present invention.

Claims

Claim

A human-machine interface detection method suitable for a nuclear power plant digital control system, characterized in that it comprises the following steps:

Human-based cognitive reliability The HCR method identifies a risk scenario in which the probability of non-response of operator behavior exceeds a predetermined threshold;

For the identified risk scenarios, the cognitive reliability and error analysis CREAM method is used to determine the probability of failure of each task in the risk scenario and each behavior in the task;

The human-machine interface involved in each task or the respective behaviors is sequentially checked against a preset human factors checklist according to the magnitude of the error probability to determine a human-machine interface unit with defects.

2. The human-machine interface detecting method according to claim 1, wherein:

The formula used by the HCR method to calculate the non-response probability is as follows:

Where t is the available time to complete the task, T _{1 / 2} is the median response time for the operator to complete the task, α is the scale parameter, β is the shape parameter, is the position parameter, and P(t) is the non-response probability.

3. The human-machine interface detecting method according to claim 2, wherein:

The T _{1 / 2} is calculated by the following formula:

_{1/2 1/2} = Τ _1/2 , _η (1 + Κ ₁ )(1 + Κ ₂ )(1 + Κ ₃ ) ^ where ^Tl/2 , ⁿ is the execution time, ^Kl and ^Κ represent the training correction factor, Psychological stress correction factor and human-machine interface correction factor.

4. The human-machine interface detecting method according to claim 1, wherein:

Using Cognitive Reliability and Error Analysis The CREAM method determines the probability of failure of each task in the risk scenario and each behavior in the task, including the following steps:

The hierarchical task analysis method is used to analyze each task and each behavior of the identified risk scene to construct an event sequence; Determining the human error mode corresponding to the sequence of events according to the situational environment analysis and the cognitive behavior analysis;

Determining, according to the human error mode classification and the corresponding basic error probability in the CREAM method, a first human cause error probability of the tasks or the behaviors, and analyzing the situation by the context environment The first person is corrected by the probability of error, and the corrected second human factor error probability corresponding to each task or the respective behaviors is obtained;

Sorting the respective tasks or the behaviors according to the second human error probability.

5. The human-machine interface detecting method according to claim 1, wherein:

The human factor engineering checklist is established according to the information processing process of the digital control system of the nuclear power plant, wherein the information processing behavior includes: information monitoring, state evaluation, response planning, and response execution.

6. The human-machine interface detecting method according to claim 1, wherein:

The human-machine interface unit includes: an information display module, a soft control module, a digital procedure module, a user interface management module, and an alarm module.

7. A human-machine interface detection system, suitable for a nuclear power plant digital control system, characterized in that the human-machine interface detection system comprises:

Risk scene screening unit, based on human cognitive reliability HCR method identifies a risk scenario in which the probability of non-response of operator behavior exceeds a predetermined threshold;

The human factor reliability analysis unit uses the cognitive reliability and error analysis CREAM method to determine the probability of error in each task and task in the risk scenario for the identified risk scenario; the human factor engineering review unit The magnitude of the error probability sequentially checks the human-machine interface involved in each task or the respective behaviors with a preset human factor engineering checklist to determine a human-machine interface unit with defects.

8. The human-machine interface detection system according to claim 7, wherein: the risk scene screening unit comprises:

The processing module is configured to calculate the non-response probability by using the HCR method as follows:

9. The human-machine interface detection system according to claim 8, wherein: the processing module comprises: a calculation module, configured to calculate the T _1/2 by the following formula:

_{1/2 1/2} = Τ _1/2>η (1 + Κ ₁ )(1 + Κ ₂ )(1 + Κ ₃ ) ^ where, for the execution time, ^Kl and ^{Κ Κ3} represent the training correction factor and the psychological stress correction factor, respectively. And human-machine interface correction factor.

The human-machine interface detecting system according to claim 7, wherein the human factor reliability analyzing unit comprises:

An event sequence building module is configured to analyze each task and each behavior of the identified risk scene by using a hierarchical task analysis method to construct an event sequence;

a human error mode determination module, configured to determine a human error mode corresponding to the sequence of events according to the situation environment analysis and the cognitive behavior analysis;

a failure probability determination module, according to the human error mode classification and the corresponding basic error probability in the CREAM method, determining a first human cause error probability of the tasks or the behaviors, and passing the situation The environmental analysis corrects the first human factor error probability, and obtains the corrected second human factor error probability corresponding to each task or the behavior;

The sorting module sorts the tasks or the behaviors according to the second human factor error probability.

11. The human-machine interface detecting system according to claim 7, wherein:

12. The human-machine interface detection and system according to claim 7, wherein: the human-machine interface unit comprises: an information display module, a soft control module, a digitization procedure module, a user interface management module, and an alarm module.