WO2014173259A1 - 数字化主控室工作人员人因可靠性的确定方法及装置 - Google Patents

数字化主控室工作人员人因可靠性的确定方法及装置 Download PDF

Info

Publication number
WO2014173259A1
WO2014173259A1 PCT/CN2014/075738 CN2014075738W WO2014173259A1 WO 2014173259 A1 WO2014173259 A1 WO 2014173259A1 CN 2014075738 W CN2014075738 W CN 2014075738W WO 2014173259 A1 WO2014173259 A1 WO 2014173259A1
Authority
WO
WIPO (PCT)
Prior art keywords
failure probability
probability
staff
type
failure
Prior art date
Application number
PCT/CN2014/075738
Other languages
English (en)
French (fr)
Inventor
张力
黄俊歆
戴立操
李鹏程
胡鸿
陈青青
方小勇
邹衍华
蒋建军
黄卫刚
戴忠华
王春辉
苏德颂
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 (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 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 filed Critical 湖南工学院
Publication of WO2014173259A1 publication Critical patent/WO2014173259A1/zh

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00Administration; Management
    • G06Q10/06Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00Administration; Management
    • G06Q10/06Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063Operations research, analysis or management
    • G06Q10/0639Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00Administration; Management
    • G06Q10/06Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063Operations research, analysis or management
    • G06Q10/0639Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06398Performance of employee with respect to a job function

Definitions

  • the present invention relates to the field of personnel reliability analysis, and in particular to a method and apparatus for determining human reliability of a digital control room staff.
  • MCR main control room
  • MMI man-machine interface
  • the information display has changed from a light plate and an alarm prompt to Large-screen display (Plant Display System, PDS for short) and computer display (Video Display Unit (VDU) for short); operator control and manipulation mode is converted from the control button of the traditional control panel to the mouse operation using the computer terminal
  • the procedures used by the operator are transferred from traditional paper procedures to electronic procedures that are displayed on a computer screen and are State-Oriented Procedure (SOP).
  • SOP State-Oriented Procedure
  • Embodiments of the present invention are directed to providing a method and apparatus for determining human factor reliability of a digital control room staff, so as to solve the problem that the human factor reliability analysis of the digital main control room staff cannot be realized in the prior art.
  • a method for determining a human factor reliability of a digital control room staff including: determining a failure probability of a worker's response to each stage of the task, or determining a difference The failure probability of the type worker; the total failure probability is determined according to the failure probability.
  • the worker includes: a first type of staff, a second type of staff, and a third type of staff, wherein the first type of staff performs an accident handling, and the second type of staff monitors the unit status Reference The number changes, monitors the execution of the first type of staff, and independently verifies the execution.
  • the third type of staff independently checks the unit status, determines the nature of the accident, and evaluates the safety status of the unit.
  • the respective phases include: a monitoring phase, a state evaluation phase, a response planning phase, and a response execution phase, wherein the monitoring phase includes the staff monitoring system state transition, the state assessment phase including the work The personnel evaluates the monitored state, the response planning phase including the response policy employed by the worker to determine the monitored state, the response execution phase including the staff performing the response policy.
  • the total failure probability is a total failure probability of each stage of each of the first type of workers; determining a total failure probability of each stage of the first type of workers according to the following manner: The failure probability is integrated by two event trees.
  • determining a total failure probability of each stage of the first type of staff according to the failure probability further comprises: adjusting a total failure probability F t of each stage of the first type of staff according to the following formula.
  • Tal the final total failure probability F T of the first class of workers is obtained :
  • F T F t . tal / (l_T), where, T is the task of managing two types of human reliability impact factor, T than 0 and less than 1.
  • the total failure probability is a total failure probability of a team consisting of the first type of staff, the second type of staff, and the third type of staff; determining the failure probability of different types of workers includes: Obtaining a first failure probability of the first type of staff; acquiring a second failure probability of the second type of staff; determining, by the medium correlation MD, the third type of the third type of staff according to the second failure probability The probability of failure; determining the total failure probability of the contingency according to the failure probability comprises: determining a total failure probability of the contingency according to the first failure probability, the second failure probability, and a third failure probability.
  • a device for determining the human factor reliability of a digital control room staff including: a first determining module, configured to acquire a failure probability of each stage of a worker's response to the task, Or determining a failure probability of different types of workers; and a second determining module, configured to determine a total failure probability according to the failure probability.
  • the monitoring phase includes the staff monitoring system state transition, the state assessment phase including the staff assessing the monitored state, the response planning phase including the staff determining a response to the monitored state
  • the policy execution phase includes the staff performing the response policy.
  • the first determining module includes: a first acquiring unit, configured to acquire a first failure probability of the first type of staff; a second acquiring unit, configured to acquire a second failure probability of the second type of staff; a unit, configured to determine, by using a medium correlation MD, a third failure probability of the third type of worker according to the second failure probability; the second determining module, configured to use, according to the first failure probability, the second failure probability And a third failure probability determining a total failure probability of the team consisting of the first type of staff, the second type of staff, and the third type of staff; wherein the first type of staff performs an accident handling
  • the second type of staff monitors changes in the state parameters of the unit, monitors the execution of the first type of staff, and independently verifies the execution.
  • the third type of staff independently checks the status of the unit and determines the nature of the accident. , evaluation of the crew and safety status.
  • FIG. 1 is a flow chart showing a method for determining human factor reliability of a digitized main control room worker according to an embodiment of the present invention
  • FIG. 2 is a flow chart showing a method for determining a failure probability of a monitoring phase according to an embodiment of the present invention.
  • FIG. 3 is a schematic diagram of a state evaluation phase model according to an embodiment of the present invention
  • FIG. 4 is a schematic diagram of a response planning phase model according to an embodiment of the present invention
  • FIG. 5 is a schematic diagram of a human error path according to an embodiment of the present invention
  • 6 is a schematic diagram of a total failure probability of an operator according to an embodiment of the present invention
  • FIG. 7 is a schematic diagram of a total failure probability of a team according to an embodiment of the present invention
  • FIG. 8 is a digital master room staff member according to an embodiment of the present invention.
  • FIG. 9 is a schematic diagram of a system structure frame according to an embodiment of the present invention.
  • a method for determining human factor reliability of a digital control room staff is provided.
  • 1 is a flow chart of a method for determining human reliability of a digitized main control room worker according to an embodiment of the present invention. As shown in FIG. 1, the method includes steps S102 to S104. In step S102, the failure probability of each stage of the task response of the worker is determined, or the failure probability of different types of workers is determined.
  • Step S104 determining a total failure probability according to the failure probability.
  • the foregoing method may determine a total failure probability of a type of staff, and may also determine a total failure probability of a team composed of different types of workers. The above two aspects are described separately below. I.
  • the total failure probability of a type of staff In the embodiment of the present invention, the first type of staff is taken as an example for description, and the reliability of the staff of the first type of staff performing the accident handling is in the nuclear power plant.
  • the staff performing the above processing may be the first and second circuit operators.
  • the response of the staff to the task can be divided into: monitoring phase, state assessment phase, response planning phase, and response execution phase.
  • the monitoring phase includes a staff monitoring system state transition
  • the state assessment phase includes a staff assessment of the monitored state
  • the response planning phase includes the staff determining a response strategy for the monitored state, the response execution phase including the work
  • the person performs the above response strategy.
  • the total failure probability of a class of workers can be determined as follows: The failure probability of each phase is integrated by using two event trees, and the total failure probability is obtained according to the following formula:
  • determining the total failure probability of a class of workers according to the failure probability of each stage may further include: adjusting a total failure probability F t of the class of workers according to the following formula. ta, to get the final total failure probability F T:
  • T F t . tal / (l_T), where, T is the task of managing two types of human reliability impact factor, T than 0 and less than 1. In practical applications, T can be 10%.
  • T is the task of managing two types of human reliability impact factor, T than 0 and less than 1. In practical applications, T can be 10%.
  • the failure probability of each type of staff response to each stage of the task response is not considered.
  • the focus of the embodiment of the present invention is to determine the total failure probability according to the failure probability of each stage.
  • an example of determining the failure probability of each stage is described by taking the operator of the nuclear power plant main control room as an example.
  • the operator's important response can be broken down into monitoring, status assessment, response planning, response execution of four main tasks, and assessment of the failure (success) probability of these responses.
  • the Markov model of the corresponding monitoring behavior, the Bayesian network model of the state evaluation, the Bayesian network model of the response plan, and the event tree quantitative evaluation model of the response execution are respectively constructed for each of the above-mentioned tasks, and the main tasks are calculated. Failure probability.
  • the basic principles of the four models and the method for determining the probability of failure are described below.
  • (1) Markov model monitoring behavior The monitoring behavior of nuclear power plant operators is the act of obtaining information from a complex and dynamic working environment. From the point of view of the monitoring activity itself, in general, the shift of the operator's monitoring of the system state is usually based on the current state of the system, regardless of the previous state of the system.
  • the monitoring target has certain expectations (especially in the event of an accident/event)
  • the monitoring path and the transfer process have no expectation and obvious regularity, and have obvious randomness. Therefore, the monitoring process can be approximated as a random process.
  • the Markov model is typically used to describe dynamic, continuous stochastic process functions. According to the foregoing analysis, it can be assumed that the direction of the monitoring transition is only related to the state and factors of the current monitoring point, so it can be assumed that the transfer process is ineffective.
  • the continuous transfer process of sex in time series, with Markov property can be simulated by Markov model.
  • Step S1 Judging The operating state of the power plant, that is, before starting to determine the monitoring task, the current operating state of the power plant needs to be determined before monitoring and analysis, and only "normal” and "abnormal” are taken.
  • Step S2 The monitoring process is decomposed, that is, the operator monitoring process is decomposed according to certain rules and logics based on the monitoring task, and the (transfer) node (N) of the monitoring process is drawn.
  • Step S3 Determine the monitoring behavior time window, that is, divide the starting point TO and the ending point TE of the monitoring activity, and determine the time period (window) of the monitoring activity.
  • the monitoring starting point can be set to enter the SOP time (time 0); in other analysis, the monitoring starting time should be reasonably determined based on the actual situation.
  • Step S4 Determine the monitoring transfer node, that is, based on the above steps, draw a schematic diagram of the monitoring node (transfer node) of the monitoring process, and determine the monitoring transfer node.
  • Step S5 calculating the success probability of the node i monitoring, that is, respectively calculating "the perceived success probability P of the operator node i" and the operator shifting from the node i-1 to the node i operator monitoring the transition success probability P T f , and taking both The probability product gives the probability of the node i operator monitoring success.
  • Step S6 Calculate the success probability of node i monitoring! ⁇ Two? ⁇ ? ⁇ .
  • Bayesian network model state assessment of state assessment involves two related models, namely state model and mental model.
  • the state model is the operator's understanding of a particular state, and when new information is collected, the state model is updated frequently.
  • the mental model is built through formal education, specific training, and operator experience, and is stored in the brain.
  • the state assessment process primarily develops a state model to assess the current state of the plant.
  • the assessment of the state is mainly influenced by the parameters, and different combinations of different parameters and their states may result in different states.
  • the above state evaluation Bayesian network model will be further described below. First, introduce the state assessment behavior of the operator of the digital nuclear power plant main control room.
  • the state model is the operator's understanding of the state presented by a particular plant system, and the state model is updated frequently as new information is collected.
  • the mental model is built through formal education, specific training, and operator experience, and is stored in the brain.
  • the state assessment process primarily develops a state model to describe the current plant state. If an event (such as an alarm) is very simple, the operator does not need any reasoning for the identification of the state of the plant, and is considered a skill-based state assessment.
  • an abnormal event belongs to a so-called "problem"
  • the operator is required to produce the original problem.
  • the process is constructed by describing the impact and the state model is constructed, and the constructed state model is matched with the operator's mental model (ie, similarity matching).
  • This process is called a regular state assessment.
  • operators are required to evaluate and predict possible plant states, then analyze the more abstract logical relationships between the structure and function of the problem space, perform in-depth analysis, and gradually form a state model and verify it. Finally, determine the state of the plant, which is considered a knowledge-based state assessment.
  • the assessment involves two related models, the state model and the mental model.
  • the state model is the operator's understanding of the specific state of the system or component, and the state model is updated frequently as new information is collected.
  • the mental model is built through formal education, specific training, and operator experience, and is stored in the brain.
  • the state assessment process primarily develops a state model to assess the current state of the plant. If the operator is to evaluate the current state of the real power plant well, the operator needs to use his own mental model to identify the current state of the plant.
  • the mental level/mind model is derived from the knowledge and experience of the operator.
  • the knowledge and experience are mainly influenced by the training of the organization and the communication and cooperation of the team. If the training is not enough, the knowledge and experience of the operator will be affected. Cooperation with can complement the knowledge and experience of individual operators.
  • the recognizability of the state presented by the power plant is mainly affected by the automation level of the digital human-machine interface and system. If the digital human-machine interface is designed well, the information is awake, and it is easy to collect information and identify.
  • the stress level has a great influence on the operator's matching between the state model and the mental model.
  • the stress level is mainly affected by the severity of the event, the complexity of the task and the available time.
  • the complexity of the same task is mainly digitized.
  • the quality of the design of the procedure and the influence of the design of the digital human-machine interface The complexity of the tasks in the procedure is complicated by the tasks that the operator needs to complete. The procedures or procedures are good for guiding the operator to respond to the plan.
  • the human-machine interface is not good.
  • the status assessment is influenced by factors such as the level of communication and cooperation of the team, the level of training, digital procedures, digital human-machine interface, the severity of the incident, the available time of the incident, and the level of automation of the system.
  • the relationship diagram of the state assessment (or the Bayesian network model for state evaluation) is shown in Figure 3.
  • the Bayesian network model is a generalized state model (the graph can also add corresponding nodes), where The lowest level of state assessment is reliable. A state assessment reliability node.
  • Bayesian network model of response plan In general, the reliability of the response plan is mainly affected by the mental state of the frontline operator, the information in the memory, and the inherent attributes of the personality. The knowledge and experience of the operator will recognize that the specific plant state corresponds to What kind of response strategy or plan to take. Knowledge and experience are mainly influenced by the training of the organization and the communication and cooperation of the team. If the training is not enough, the knowledge and experience of the operator will be affected. The exchange and cooperation of the team can supplement the lack of knowledge and experience of the individual operators. In addition, the stress level has a great influence on the formulation of the response plan. The stress level is mainly affected by the severity of the event, the complexity of the task and the available time.
  • the complexity of the same task is mainly influenced by the design of the procedure and the human-machine.
  • the influence of the interface design is complicated.
  • the tasks in the procedure are complicated.
  • the tasks that the operator needs to complete are complicated.
  • the procedures or procedures are good for guiding the operator to make a response plan. If the human-machine interface is not good, the operator is difficult to obtain a favorable response. Useful information for planning.
  • the response plan is also affected by the attitude of the operator.
  • the attitude and responsibility of the operator is good. It is difficult to violate the rules and focus.
  • the attitude of the operator is mainly affected by the safety culture and management of the organization. If the safety culture is not good, the operator's risk awareness and safety attitude are not good.
  • Figure 4 shows a schematic diagram of the response planning phase model, the operator responding to the planned Bayesian network model.
  • the response plan can be affected by the team's communication and cooperation level, training level, digital procedures, digital human-machine interface, the severity of the incident, the available time of accident handling, safety culture and organizational management level, etc. The impact of factors.
  • post-accident response execution behavior means that the operator configures the VDU screen with the mouse and clicks on the SOP procedure.
  • the success path of the human factor event is shown in Figure 5.
  • the success path of the human factor event can include two processes, the operator cognitive process and the action process, monitoring the plant information in the cognitive process operator, and monitoring the plant status.
  • the assessment based on the evaluation results, develops a response plan in which the operator performs the formulated response plan.
  • the operator reliability model uses the above four models to calculate the probability of success when the operator performs monitoring, status evaluation, response planning, and response to the execution of four main tasks.
  • a two-branch event tree is used for integrated integration.
  • the above method may be used to determine the reliability of the staff performing the accident processing, and the staff performing the above processing in the nuclear power plant are the first and second circuit operators.
  • F total P( A) + a ⁇ P(B) + ab ⁇ P(C) + abc ⁇ P(D)
  • the second type of management task results in a 10% reduction in overall mission performance.
  • the staff may include: a first type of staff, a second type of staff and a third type of staff, wherein the first type of staff performs an accident Processing, in a nuclear power plant, the first type of staff can be the first and second circuit operators; the second type of staff monitors the changes in the unit status parameters, monitors the performance of the first type of staff, and independently verifies the
  • the second type of staff in a nuclear power plant may be the captain/coordinator; the third type of staff independently checks the status of the unit, determines the nature of the accident, evaluates the unit's and safety status, and performs the third type of work in the nuclear power plant.
  • Determining the failure probability of different types of workers includes: obtaining the first failure probability of the first type of staff; obtaining the second failure probability of the second type of staff; using the medium correlation to determine the third type of staff according to the second failure probability Three failure probability.
  • Determining the total failure probability according to the failure probability includes: determining a total failure probability according to the first failure probability, the second failure probability, and the third failure probability.
  • the first failure probability is determined by the above method provided by the embodiment of the present invention, and details are not described herein. The following describes an example of determining the total failure probability of different types of workers.
  • the accident state of the DCS of the East Nuclear Power Plant is composed of the Angong, the captain/coordinator and the first and second circuit operators.
  • the first and second circuit operators execute the DOS program and related accident handling procedures.
  • the captain/coordinator monitors the main unit.
  • the status parameter changes, monitors the execution of the first or second circuit operator SOP program or the corresponding accident handling program, and independently verifies the key criteria and key operations.
  • the independent inspection of the power plant unit status and the nature of the accident to evaluate the nuclear safety status of the unit.
  • the coordinator performs the procedures separately, monitors the main state parameter changes of the crew, and the coordinator monitors the first and second circuit operators and independently verifies their key criteria and key controls.
  • the coordinator is related to the first and second circuit operators. Due to the role of "independent verification" in the organizational structure design, the method adopts a moderately conservative strategy and uses medium correlation (MD: Moderate Dependence) for calculation.
  • MD Moderate Dependence
  • the calculation method is carried out by using the recovery factor method.
  • the coordinator uses the paper procedure, and the security worker judges the state of the power plant without performing the operation.
  • the performance degradation of the above two is not considered due to the second type of management task. According to the above analysis, as shown in Figure 7
  • the total failure probability of the operator team is:
  • FIG. 8 is a structural block diagram of a device for determining human reliability of a digitized main control room staff according to an embodiment of the present invention.
  • the apparatus includes: a first determining module 10 and a second determining module 20.
  • the first determining module 10 is configured to obtain a failure probability of each stage of the worker's response to the task, or determine a failure probability of the different types of workers.
  • the second determining module 20 is connected to the first determining module 10, and is configured to The total failure probability is determined based on the probability of failure.
  • the foregoing apparatus may determine a total failure probability of one type of staff, and may also determine a total failure probability of a team composed of different types of workers.
  • the above two aspects are described separately below.
  • the second determining module 20 uses the two event trees for integration in the failure probability of each stage, and obtains the following formula according to the following formula.
  • P (A) is the probability of failure in the monitoring phase
  • a is the probability of success in the monitoring phase
  • P (B ) is the probability of failure in the state assessment phase
  • b is the probability of success in the state assessment phase
  • P (C) is the phase of the response planning
  • c is the probability of success in response to the planning phase
  • P (D) is the probability of failure in response to the execution phase.
  • the monitoring phase includes a staff monitoring system state transition
  • the state assessment phase includes a staff assessment of the monitored state
  • the response planning phase includes a staff member determining a response strategy for the monitored state
  • the response execution phase includes staff execution. Response strategy.
  • the first determining module 10 may include: a first acquiring unit, configured to acquire a first failure probability of the first type of staff; a second acquiring unit, configured to obtain a second failure probability of the second type of staff; and a determining unit, configured to determine, by using the medium correlation MD, a third failure probability of the third type of worker according to the second failure probability.
  • the second determining module 20 is configured to determine a total failure probability according to the first failure probability, the second failure probability, and the third failure probability.
  • the first type of staff performs accident handling.
  • the first type of staff can be the first and second circuit operators; the second type of staff monitors the change of the unit status parameters and monitors the first type of work.
  • the second type of staff in the nuclear power plant can be the captain / coordinator; the third type of staff independently check the status of the unit, determine the nature of the accident, evaluate the unit and safety status
  • the third type of staff can be an employee.
  • the third failure probability, ⁇ ( ⁇ ) is the second probability of failure.
  • F crew P A xP (B/A) x P B , wherein , F Dining pw is the total failure probability, and P A is the first failure probability.
  • the Markov model for monitoring behavior, the Bayesian network model for state evaluation, the Bayesian network model for response planning, and the four known models for the event tree quantitative evaluation model for response execution can be integrated.
  • 9 is a schematic structural diagram of a system according to an embodiment of the present invention, and the main model, data and the like of the embodiment of the invention are connected.

Landscapes

  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Engineering & Computer Science (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Development Economics (AREA)
  • Educational Administration (AREA)
  • Operations Research (AREA)
  • Marketing (AREA)
  • Game Theory and Decision Science (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Testing And Monitoring For Control Systems (AREA)

Abstract

本发明公开了一种数字化主控室工作人员人因可靠性的确定方法及装置,其中,该方法包括:确定工作人员对任务响应的各个阶段的失效概率,或者确定不同类型工作人员的失效概率;根据上述失效概率确定总失效概率。通过本发明,实现了对数字化主控室工作人员人因可靠性的分析。

Description

数字化主控室工作人员人因可靠性的确定方法及装置
技术领域 本发明涉及人员可靠性分析领域, 具体而言, 涉及一种数字化主控室工作人员人 因可靠性的确定方法及装置。 背景技术 由于核电厂主控室 (Main Control Room, 简称为 MCR) 数字化以后, 人机界面 (Man-machine Iinterface, 简称为 MMI) 发生了巨大变化, 信息显示从光字牌、 报警 器提示转变成大屏幕显示(Plant Display System,简称为 PDS )和计算机终端显示(Video Display Unit, 简称为 VDU); 操纵员控制和操纵方式从传统的控制盘台的控制键操纵 转换成使用计算机终端的鼠标操纵; 操纵员使用的规程由传统的纸质规程转为在计算 机屏幕上显示的以电厂状态为导向 (State-Oriented Procedure, 简称为 SOP) 的电子规 程。 电厂系统的数字化势必会引起人的因素的诸多方面的变化。 为了探究这种变化的 内部机制和影响模式以及这种变化带来的后果, 需要建立与之相适应的新的人的行为 模型和人员可靠性分析 (Human Reliability Analysis, 简称为 HRA) 方法。 由于人的生理和心理因素复杂, 加之与系统和周围环境的交互性和相关性, 导致 在某种程度上人的行为不像机械、 电子设备那样具有确定性,并难以进行定量化描述。 相关技术中的人员可靠性定量分析方法尚不多见, 其中具有代表性的人员可靠性定量 分析方法为 CREAM定量分析法和 THERP+HCR定量分析法。但这些方法仅适用于传 统主控室中的人员可靠性分析, 未考虑主控室数字化之后的特征。 发明内容 本发明实施例旨在提供一种数字化主控室工作人员人因可靠性的确定方法及装 置, 以解决现有技术中无法实现对数字化主控室工作人员人因可靠性的分析的问题。 为了实现上述目的, 根据本发明实施例的一个方面, 提供了一种数字化主控室工 作人员人因可靠性的确定方法, 包括: 确定工作人员对任务响应的各个阶段的失效概 率, 或者确定不同类型工作人员的失效概率; 根据所述失效概率确定总失效概率。 优选地, 所述工作人员包括: 第一类工作人员、 第二类工作人员和第三类工作人 员, 其中, 所述第一类工作人员执行事故处理, 所述第二类工作人员监控机组状态参 数的变化、 监控所述第一类工作人员的执行情况, 并独立验证所述执行情况, 所述第 三类工作人员独立检查机组状态、 判断事故性质、 评价机组的安全状态。 优选地, 所述各个阶段包括: 监视阶段、 状态评估阶段、 响应计划阶段、 响应执 行阶段, 其中, 所述监视阶段包括所述工作人员监视系统状态的转移, 所述状态评估 阶段包括所述工作人员评估监视到的状态, 所述响应计划阶段包括所述工作人员确定 对监视到的状态所采用的响应策略, 所述响应执行阶段包括所述工作人员执行所述响 应策略。 优选地, 所述总失效概率为各个所述第一类工作人员的各个阶段的总失效概率; 按照以下方式确定所述第一类工作人员的各个阶段的总失效概率: 对所述各个阶段的 失效概率采用两支事件树进行集成, 按照以下公式得到所述第一类工作人员的各个阶 段的总失效概率为: Ftotal = P( A) + a · P(B) + ab · P(C) + abc · P(D); 其中, P (A) 为监 视阶段的失效概率, a为监视阶段的成功概率, P (B ) 为状态评估阶段的失效概率, b 为状态评估阶段的成功概率, P (C) 为响应计划阶段的失效概率, c为响应计划阶段 的成功概率, P (D) 为响应执行阶段的失效概率。 优选地, 根据所述失效概率确定所述第一类工作人员的各个阶段的总失效概率, 还包括: 根据如下公式调整所述第一类工作人员的各个阶段的总失效概率 Fttal, 得到 所述第一类工作人员的最终的总失效概率 FT : FT = Fttal/(l_T), 其中, T是二类管理 任务对人因可靠性的影响因子, T大于等于 0且小于 1。 优选地, 所述总失效概率为所述第一类工作人员、 所述第二类工作人员和所述第 三类工作人员构成的班组的总失效概率; 确定不同类型工作人员的失效概率包括: 获 取所述第一类工作人员的第一失效概率; 获取所述第二类工作人员的第二失效概率; 采用中等相关 MD根据所述第二失效概率确定所述第三类工作人员的第三失效概率; 根据所述失效概率确定所述班组的总失效概率包括: 根据所述第一失效概率、 所述第 二失效概率和第三失效概率确定所述班组的总失效概率。 优选地, 按照以下方式确定所述第三失效概率: MD,P(B/ A) = 1 + 6^(B\ 其中,
P(B/ A)为所述第三失效概率, P(B)为所述第二失效概率。 优选地, 按照以下方式确定所述班组的总失效概率: FCTew = PAxP(B/A) x PB,其中, F„pw为所述班组的总失效概率, ?八为所述第一失效概率。 根据本发明实施例的另一个方面, 提供了一种数字化主控室工作人员人因可靠性 的确定装置, 包括: 第一确定模块, 用于获取工作人员对任务响应的各个阶段的失效 概率, 或者确定不同类型工作人员的失效概率; 第二确定模块, 用于根据所述失效概 率确定总失效概率。 优选地, 所述第二确定模块, 用于对所述各个阶段的失效概率采用两支事件树进 行集成, 按照以下公式得到所述工作人员的各个阶段的总失效概率为: Ftotai = P( + - P(B) + ab - P(C) + abc - P(D) , 其中, Ρ (Α) 为监视阶段的失效概率, a为监视阶段的成功概率, P (B ) 为状态评估阶段的失效概率, b 为状态评估阶段的 成功概率, P (C)为响应计划阶段的失效概率, c为响应计划阶段的成功概率, P (D) 为响应执行阶段的失效概率; 其中, 所述监视阶段包括所述工作人员监视系统状态的 转移, 所述状态评估阶段包括所述工作人员评估监视到的状态, 所述响应计划阶段包 括所述工作人员确定对监视到的状态所采用的响应策略, 所述响应执行阶段包括所述 工作人员执行所述响应策略。 优选地, 所述第二确定模块, 还用于根据如下公式调整所述工作人员的各个阶段 的总失效概率 Fttal, 得到所述工作人员的最终的总失效概率 FT : FT = Fttal/a-T), 其中, T是二类管理任务对人因可靠性的影响因子, T大于等于 0且小于 1。 优选地, 所述第一确定模块包括: 第一获取单元, 用于获取第一类工作人员的第 一失效概率; 第二获取单元, 用于获取第二类工作人员的第二失效概率; 确定单元, 用于采用中等相关 MD根据所述第二失效概率确定第三类工作人员的第三失效概率; 所述第二确定模块, 用于根据所述第一失效概率、 所述第二失效概率和第三失效概率 确定由所述第一类工作人员、 所述第二类工作人员和所述第三类工作人员构成的班组 的总失效概率; 其中, 所述第一类工作人员执行事故处理, 所述第二类工作人员监控 机组状态参数的变化、 监控所述第一类工作人员的执行情况, 并独立验证所述执行情 况, 所述第三类工作人员独立检查机组状态、判断事故性质、评价机组的和安全状态。 优选地, 所述确定单元, 用于确定所述第三失效概率为 MD, P(B/ A) = 1 + 6 B), 其中, P(B/ A)为所述第三失效概率, P(B)为所述第二失效概率。 优选地, 所述第二确定模块, 用于确定所述班组的总失效概率为: FCTew = PAx P(B/A) x PB, 其中, FCTew为所述总失效概率, PA 所述第一失效概率。 应用本发明实施例的技术方案,确定工作人员对任务响应的各个阶段的失效概率, 或者确定不同类型工作人员的失效概率, 并根据上述失效概率确定总失效概率, 实现 了对数字化主控室工作人员人因可靠性的分析。 附图说明 构成本申请的一部分的说明书附图用来提供对本发明的进一步理解, 本发明的示 意性实施例及其说明用于解释本发明, 并不构成对本发明的不当限定。 在附图中: 图 1是根据本发明实施例的数字化主控室工作人员人因可靠性的确定方法的流程 图; 图 2是根据本发明实施例监视阶段的失效概率的确定方法的流程图; 图 3是根据本发明实施例的状态评估阶段模型的示意图; 图 4是根据本发明实施例的响应计划阶段模型的示意图; 图 5是根据本发明实施例的人因失误路径的示意图; 图 6是根据本发明实施例的操纵员的总失效概率的示意图; 图 7是根据本发明实施例的班组的总失效概率的示意图; 图 8是根据本发明实施例的数字化主控室工作人员人因可靠性的确定装置的结构 框图; 以及 图 9是根据本发明实施例的系统结构框架的示意图。 具体实施方式 需要说明的是, 在不冲突的情况下, 本申请中的实施例及实施例中的特征可以相 互组合。 下面将参考附图并结合实施例来详细说明本发明。 根据本发明实施例, 提供了一种数字化主控室工作人员人因可靠性的确定方法。 图 1是根据本发明实施例的数字化主控室工作人员人因可靠性的确定方法的流程 图, 如图 1所示, 该方法包括步骤 S102至步骤 S104。 步骤 S102, 确定工作人员对任务响应的各个阶段的失效概率, 或者确定不同类型 工作人员的失效概率。 步骤 S104, 根据上述失效概率确定总失效概率。 应用本发明实施例的技术方案,确定工作人员对任务响应的各个阶段的失效概率, 或者确定不同类型工作人员的失效概率, 并根据上述失效概率确定总失效概率, 实现 了对数字化主控室工作人员人因可靠性的分析。 在本发明实施例中, 上述方法可以确定一种类型的工作人员的总失效概率, 也可 以确定不同类型的工作人员组成的班组的总失效概率。 下面分别对上述两个方面进行 描述。 一、 一种类型的工作人员的总失效概率 在本发明实施例中, 以第一类工作人员为例进行说明, 第一类工作人员执行事故 处理的工作人员的可靠性, 在核电厂中的执行上述处理的工作人员可以是第一、 二回 路操纵员。 该类工作人员对任务的响应可以分为: 监视阶段、 状态评估阶段、 响应计 划阶段、 响应执行阶段。 其中, 监视阶段包括工作人员监视系统状态的转移, 状态评估阶段包括工作人员 评估监视到的状态, 响应计划阶段包括所述工作人员确定对监视到的状态所采用的响 应策略, 响应执行阶段包括工作人员执行上述响应策略。 在本发明实施例的一个优选实施方式中, 可以按照以下方式确定一类工作人员的 总失效概率: 对各个阶段的失效概率采用两支事件树进行集成, 按照以下公式得到总 失效概率为:
Ftotal = P( A) + a · P(B) + ab · P(C) + abc · P(D) 其中, P (A) 为监视阶段的失效概率, a为监视阶段的成功概率, P (B ) 为状态 评估阶段的失效概率, b为状态评估阶段的成功概率, P (C) 为响应计划阶段的失效 概率, c为响应计划阶段的成功概率, P (D ) 为响应执行阶段的失效概率。 优选地, 根据各个阶段的失效概率确定一类工作人员的总失效概率还可以包括: 根据如下公式调整该类工作人员的总失效概率 Ftta, 得到最终的总失效概率 FT :
FT = Fttal/(l_T), 其中, T是二类管理任务对人因可靠性的影响因子, T大于等于 0 且小于 1。 在实际应用中 T可以为 10%。 在本发明实施例中, 并不关注一类工作人员对任务响应的各个阶段的失效概率如 何确定, 本发明实施例的重点在于根据各个阶段的失效概率确定总的失效概率。 下面 以核电厂主控室操纵员为例, 对确定各个阶段的失效概率的一种实施方式进行描述。 优选地, 操纵员的重要响应可分解为监视、 状态评估、 响应计划、 响应执行 4项 主要任务, 并评估这些响应的失效 (成功) 概率。 对上述各个阶段任务分别构建相应的监视行为的马尔科夫模型、 状态评估的贝叶 斯网络模型、 响应计划的贝叶斯网络模型以及响应执行的事件树定量评价模型, 计算 出各项主要任务的失效概率。 下面介绍 4个模型的基本原理, 以及失效概率的确定方 法。 ( 1 ) 监视行为的马尔科夫模型 核电厂操纵员的监视行为就是从复杂动态的工作环境中获取信息的行为。 从监视 行为活动本身来看, 一般来说, 操纵员对系统状态的监视的转移, 通常是根据系统的 当前状态, 而与系统以前的状态无关。 虽然监视目标具有一定的预期性 (特别是事故 / 事件状态下),但是监视路径与转移过程没有预期性与明显的规律, 具有较明显的随机 性, 因此, 监视过程可近似看作为随机过程。 这类过程无确定的变化形式 (无必然的 变化规律), 从而不可能用精确的数学关系式表示, 但可以用随机函数来描述。 马尔科 夫模型是典型的用来描述动态、 连续随机过程函数, 根据前述的分析, 可以假设监视 转移的去向仅与本次监视点的状态和因素相关, 因此可以假设该转移过程是无后效性 的在时间序列上的连续转移过程, 具有马尔科夫性, 可以用马尔科夫模型来模拟。 图
2给出了计算监视成功及失败概率的算法流程图, 如图 2所示, 基于操纵员监视行为 分析要求, 以及监视行为各节点的逻辑关系, 制定监视行为定量分析流程如下: 步骤 S1 : 判断电厂运行状态, 即在开始确定监视任务, 进行监视分析前需要确定 电厂的当前运行状态, 只取"正常"与"异常"两种。 步骤 S2: 监视过程分解, 即基于监视任务将操纵员监视过程按照一定规则与逻辑 进行分解, 划出监视过程的 (转移) 节点 (N)。 当电厂状态为 "正常", 基于监视任务、 相关监视参数、 操作规程与操纵员监视该 类任务经验, 采取专家判断方法合理地确定监视节点; 当电厂状态为 "异常", 对于 PSA框架下的 HRA监视任务, 依据分析事件 /事故的 SOP规程来划分监视节点。 对于 节点划分还可基于对操作流程与事件处理规程的知识表征方法来实现。 步骤 S3: 确定监视行为时间窗口, 即划分监视活动的起点 TO与终点 TE, 确定监 视活动的时间段(窗口)。在 PSA-HRA分析中, 监视起点可设置为进入 SOP时刻(记 0时刻); 在其他分析中要基于实际情况合理确定监视起点时刻。 步骤 S4: 确定监视转移节点, 即基于上面步骤, 划出监视过程的监视节点 (转移 节点) 逻辑示意图, 确定监视转移节点。 步骤 S5:计算节点 i监视成功概率,即分别计算"操纵员节点 i的察觉成功概率 P " 与操纵员自节点 i-1转移到节点 i操纵员监视转移成功概率 PTf , 并取两者概率乘积得 到节点 i操纵员监视成功概率。 步骤 S6: 计算节点 i监视成功概率!^二?^ ?^。 步骤 S7: 计算监视成功概率 = ΠΡ 。
i=l 步骤 S8: 计算监视失败概率 P^ =1_P^。
(2) 状态评估的贝叶斯网络模型 状态评估涉及两个相关的模型, 即状态模型和心智模型。 状态模型就是操纵员对 特定状态的理解, 并且当收集到新信息的时候, 状态模型会被经常更新。 心智模型是 通过正式的教育、 具体的培训和操纵员经验来构建的, 并且存储在大脑中。 状态评估 过程主要就是发展一个状态模型来评估当前的电厂状态。 状态的评估主要受参数的影 响, 不同的参数及其状态的不同组合可能得到不同的状态。 下面进一步对上述状态评 估贝叶斯网络模型其进行说明。 首先, 介绍数字化核电厂主控室操纵员的状态评估行为。 当核电厂发生异常状态时, 操纵员将根据核电厂的状态参数情况构建一个合理的和合乎 逻辑的解释, 来评估电厂所处的状态, 作为后续的响应计划和响应执行决策的依据。 这一系 列过程称为状态评估, 并涉及两个相关的模型, 即状态模型和心智模型。 状态模型就是操纵 员对特定电厂系统所呈现出的状态的理解, 并且当收集到新信息的时候, 状态模型会被经常 更新。 心智模型是通过正式的教育、 具体的培训和操纵员经验来构建的, 并且存储在大脑中。 状态评估过程主要就是发展一个状态模型来描述当前的电厂状态。 如果一个事件(如报警) 非常简单, 操纵员对电厂状态的辨识不需要任何推理, 则认为 是技能型的状态评估。 如果一个异常事件属于所谓的 "问题", 要求操纵员对该问题产生的原 因和影响进行说明来构建状态模型,并且构建好的状态模型与操纵员的心智模型进行匹配(即 相似性匹配), 则这个过程称为规则型的状态评估。 同样, 对于不熟悉的状态模式, 要求操纵 员评估和预测可能的电厂状态, 然后分析问题空间的结构和功能之间更加抽象的逻辑关系, 进行深层次的分析, 逐渐形成一个状态模型并进行验证, 最后确定电厂状态, 这个过程被认 为是知识型的状态评估。 其次, 介绍操纵员的状态评估的贝叶斯网络模型。 通过组建专家组 (包括核电厂操纵员班组以及人因专家) 识别出影响操纵员状态评估可 靠性的影响因素以及它们的因果关系, 一般来说, 当核电厂发生异常事件后, 操纵员的状态 评估涉及两个相关的模型, 即状态模型和心智模型。 状态模型就是操纵员对系统或组件的特 定状态的理解, 并且当收集到新信息的时候, 状态模型会被经常更新。 心智模型是通过正式 的教育、 具体的培训和操纵员经验来构建的, 并且存储在大脑中。 状态评估过程主要就是发 展一个状态模型来评估当前的电厂状态。 如果操纵员要很好地评估出真实的电厂当前状态, 则操纵员需要利用其自身的心智模型去辨识出电厂当前的状态, 这个过程受电厂状态呈现的 易识别性、 操纵员的心智水平 /心智模型以及心理压力的影响。 心智水平 /心智模型来源于操纵 员的知识和经验, 知识和经验主要受组织培训的影响和班组的交流与合作的影响, 如果培训 不够, 则操纵员的知识和经验会受影响, 班组的交流与合作可以补充操纵员个体的知识和经 验的不足。 电厂所呈现的状态的易识别性 (状态模型的另一种解释) 主要受数字化人机界面和系统 的自动化水平的影响, 如果数字化人机界面设计好, 则信息醒匿, 容易搜集信息和识别出系 统所处的状态, 如果系统自动化水平高, 则操纵员没有参与到具体的任务中, 则容易丧失与 任务相关的系统状态的理解。 另外, 压力水平对操纵员在状态模型和心智模型之间的匹配有 很大的影响, 压力水平主要受事件的严重度、 任务的复杂性及可用时间的影响, 同样任务的 复杂性主要受数字化规程设计的好坏与数字化人机界面设计的好坏的影响, 规程中的任务复 杂则操纵员需要完成的任务复杂, 规程或程序好有利于指导操纵员做出响应计划, 人机界面 不好 (如诸多的界面管理任务) 则操纵员难以获取有利于任务完成的有用信息。 再者, 事件 越严重, 操纵员的心理压力越大, 完成任务的可用时间越短, 则操纵员的心理压力越大。 通 过上述分析, 状态评估受班组的交流与合作水平、 培训水平、 数字化规程、 数字化人机界面、 事件的严重度、 事故处置的可用时间以及与系统的自动化水平等因素的影响, 这些 PSF因子 与状态评估的影响关系图 (或称状态评估的贝叶斯网络模型) 见图 3 所示, 为一般化的状态 模型的贝叶斯网络模型 (该图同样也可以增加相应的节点), 其中, 最下层的状态评估可靠新 就是一种状态评估可靠性节点。
( 3 ) 响应计划的贝叶斯网络模型 一般来说, 响应计划的可靠性主要受一线操纵员的心理状态、 记忆中的信息以及 个性固有属性的影响。 操纵员的知识和经验丰富, 则会认识到特定的电厂状态对应该 采取何种响应策略或计划。 知识和经验主要受组织培训的影响和班组的交流与合作的 影响, 如果培训不够, 则操纵员的知识和经验会受影响, 班组的交流与合作可以补充 操纵员个体的知识和经验的不足。 另外, 压力水平对响应计划的制定有很大的影响, 压力水平主要受事件的严重度、 任务的复杂性及可用时间的影响, 同样任务的复杂性 主要受规程设计的好坏与人-机界面设计的好坏的影响,规程中的任务复杂则操纵员需 要完成的任务复杂, 规程或程序好有利于指导操纵员做出响应计划, 人-机界面不好则 操纵员难以获取有利于响应计划制定的有用信息。 再者, 响应计划还受操纵员的态度 的影响, 操纵员的态度和责任心好, 则难以违规, 注意力集中, 操纵员的态度主要受 组织的安全文化和管理的好坏的影响, 如安全文化不好, 则操纵员的风险意识和安全 态度则不好。 图 4示出了响应计划阶段模型的示意图, 操纵员响应计划的贝叶斯网络模型。 通 过上述分析及图 4可以看出, 响应计划可以受到班组的交流与合作水平、 培训水平、 数字化规程、 数字化人机界面、 事件的严重度、 事故处置的可用时间、 安全文化与组 织管理水平等因素的影响。 (4) 响应执行的事件树定量评价模型 电厂事故后, 操纵员对于电厂信息进行监视, 对电厂状态进行评估并做出响应计 划, 以上认知过程完成之后, 操纵员需针对响应计划进行响应执行行为。 DCS中, 事 故后响应执行行为是指操纵员利用鼠标配置 VDU画面, 并点击 SOP规程执行。 人因事件的成功路径如图 5所示, 人因事件的成功路径可以包括两个过程, 操纵 员认知过程和动作过程, 在认知过程操纵员对电厂信息进行监视, 并对电厂状态进行 评估, 根据评估结果制定响应计划, 在动作过程, 操纵员执行制定的响应计划。 操纵员可靠性模型 利用上述 4个模型, 分别计算出操纵员执行监视、 状态评估、 响应计划、 响应执 行 4项主要任务时的成功概率。 对于电厂操纵员的可靠性模型, 采用两分支事件树进 行综合集成。 事故发生后, 操纵员需准确监视, 有效评估电厂状态并做出响应计划, 执行响应动作。 在本发明实施例中, 上述方法可以用于确定执行事故处理的工作人员的可靠性, 在核电厂中的执行上述处理的工作人员为第一、 二回路操纵员。 图 6是根据本发明实施例的操纵员的总失效概率的示意图, 如图 6所示, 操纵员 共有 4条失效分支, 即 Fl、 F2、 F3、 F4。操纵员行为失效概率使用如下公式进行计算: Ftotal = P( A) + a · P(B) + ab · P(C) + abc · P(D) 二类管理任务导致任务整体绩效下降 10%, 即最终一、二回路操纵员单个 PSA始 发事件人因失效概率: FT = Fttal/90%。 二、 不同类型的工作人员的总失效概率 在本发明实施例中, 工作人员可以包括: 第一类工作人员、 第二类工作人员和第 三类工作人员, 其中, 第一类工作人员执行事故处理, 在核电厂中, 第一类工作人员 可以是第一、 二回路操纵员; 第二类工作人员监控机组状态参数的变化、 监控所述第 一类工作人员的执行情况, 并独立验证所述执行情况, 在核电厂中第二类工作人员可 以是机组长 /协调员; 第三类工作人员独立检查机组状态、 判断事故性质、 评价机组的 和安全状态, 在核电厂中第三类工作人员可以是安工。 确定不同类型工作人员的失效概率包括: 获取第一类工作人员的第一失效概率; 获取第二类工作人员的第二失效概率; 采用中等相关根据第二失效概率确定第三类工 作人员的第三失效概率。 根据失效概率确定总失效概率包括: 根据第一失效概率、 第 二失效概率和第三失效概率确定总失效概率。 在本发明实施例的一个优选实施方式中, 可以按照以下方式确定第三失效概率: MD,P(B/ A) = 1 + 6P(B\ 其中, P(B/ A)为第三失效概率, P(B)为第二失效概率。 进一步的,可以按照以下方式确定总失效概率: FCTew = PAx P(B/A) x PB,其中, FCTew 为总失效概率, PA为第一失效概率。 其中, 第一失效概率可以通过本发明实施例提供 的上述方法确定, 在此不再赘述。 下面以一个实例对不同类型工作人员的总失效概率的确定方法进行描述。 岭东核电厂 DCS 中事故状态下班组由安工、 机组长 /协调员和一、 二回路操纵员 构成。 一、 二回路操纵员执行 DOS程序和相关事故处理程序。 机组长 /协调员监控机 组主要状态参数改变, 监控一、 二回路操纵员 SOP程序或相应事故处理程序的执行, 独立验证其关键判据和关键操纵。 安工独立检查电厂机组状态, 判断事故性质, 评价 机组的核安全状态。 协调员单独执行规程, 监控机组主要状态参数改变, 协调员监视一、 二回路操纵 员, 且独立验证其关键判据和关键操纵。 协调员与一、 二回路操纵员具有相关性, 由 于组织结构设计中考虑其"独立验证"的作用, 本方法采用适当保守的策略, 采用中等 相关 (MD: Moderate Dependence) 进行计算。 计算公式为: ΜΡ, Ρ(Β/ Α) = 1 + 6 7 Ρ(Β\ 安工独立检查电厂机组状态, 判断事故性质, 评价机组的核安全状态, 为组织机 构中最后一道防线, 本方法采用恢复因子的方法进行计算。 协调员使用纸质规程, 安工对于电厂状态进行判断, 不需执行操纵。 不考虑二类 管理任务导致的上述二者的绩效下降。 根据以上分析, 如图 7 所示的班组模型。 操纵员班组总的失效概率为:
Fcrew = PA xP(B/A) xP(B)。 根据本发明实施例, 对应于上述方法, 还提供了一种数字化主控室工作人员人因 可靠性的确定装置。 图 8是根据本发明实施例的数字化主控室工作人员人因可靠性的确定装置的结构 框图, 如图 8所示, 该装置包括: 第一确定模块 10和第二确定模块 20。 其中, 第一 确定模块 10, 用于获取工作人员对任务响应的各个阶段的失效概率, 或者确定不同类 型工作人员的失效概率; 第二确定模块 20, 与第一确定模块 10相连接, 用于根据失 效概率确定总失效概率。 在本发明实施例中, 上述装置可以确定一种类型的工作人员的总失效概率, 也可 以确定不同类型的工作人员组成的班组的总失效概率。 下面分别对上述两个方面进行 描述。 一、 一种类型的工作人员的总失效概率 在本发明实施例的一个实施方式中, 第二确定模块 20, 用于各个阶段的失效概率 采用两支事件树进行集成, 按照以下公式得到所述总失效概率为: Ftotal = P( A) + a · P(B) + ab · P(C) + abc · P(D) 其中, P (A) 为监视阶段的失效概率, a为监视阶段的成功概率, P (B ) 为状态 评估阶段的失效概率, b为状态评估阶段的成功概率, P (C) 为响应计划阶段的失效 概率, c为响应计划阶段的成功概率, P (D ) 为响应执行阶段的失效概率。 其中, 监视阶段包括工作人员监视系统状态的转移, 状态评估阶段包括工作人员 评估监视到的状态, 响应计划阶段包括工作人员确定对监视到的状态所采用的响应策 略, 响应执行阶段包括工作人员执行响应策略。 进一步的, 第二确定模块 20, 还用于根据如下公式调整总失效概率 Fttal, 得到最 终的总失效概率 FT : FT = Fttal/(l -T), 其中, T 是二类管理任务对人因可靠性的影 响因子, T大于等于 0且小于 1。 二、 不同类型的工作人员的总失效概率 在本发明实施例的一个实施方式中, 第一确定模块 10可以包括: 第一获取单元, 用于获取第一类工作人员的第一失效概率; 第二获取单元, 用于获取第二类工作人员 的第二失效概率; 确定单元, 用于采用中等相关 MD根据第二失效概率确定第三类工 作人员的第三失效概率。 第二确定模块 20, 用于根据第一失效概率、 第二失效概率和第三失效概率确定总 失效概率。 其中, 第一类工作人员执行事故处理, 在核电厂中, 第一类工作人员可以是第一、 二回路操纵员; 第二类工作人员监控机组状态参数的变化、 监控所述第一类工作人员 的执行情况, 并独立验证所述执行情况, 在核电厂中第二类工作人员可以是机组长 / 协调员; 第三类工作人员独立检查机组状态、判断事故性质、评价机组的和安全状态, 在核电厂中第三类工作人员可以是安工。 在本发明实施例的一个实施方式中, 确定单元, 用于确定第三失效概率为: ΜΡ, Ρ(Β/ Α) = 1 + 6 7 Ρ(Β\ 其中, Ρ(Β/ Α)为第三失效概率, Ρ(Β)为第二失效概率。 进一步的,所述第二确定模块,用于确定所述总失效概率为: Fcrew = PAxP(B/A) x PB , 其中, F„pw为所述总失效概率, PA 所述第一失效概率。 在实际应用中, 可以将监视行为的马尔科夫模型、 状态评估的贝叶斯网络模型、 响应计划的贝叶斯网络模型以及响应执行的事件树定量评价模型 4个已知模型予以集 成, 图 9给出了本发明实施例系统的结构框架示意图, 将该发明实施例的主要模型、 数据等内容进行了衔接。 从以上的描述中, 可以看出, 本发明上述的实施例实现了如下技术效果: 将计算 技术有效地引入到数字化主控室背景下的人员可靠性定量分析中, 为数字化后的核电 厂主控室人员及班级可靠性的定量分析提供了一种新的有效系统。 以上所述仅为本发明的优选实施例而已, 并不用于限制本发明, 对于本领域的技 术人员来说, 本发明可以有各种更改和变化。 凡在本发明的精神和原则之内, 所作的 任何修改、 等同替换、 改进等, 均应包含在本发明的保护范围之内。

Claims

权 利 要 求 书
1. 一种数字化主控室工作人员人因可靠性的确定方法, 包括:
确定工作人员对任务响应的各个阶段的失效概率, 或者确定不同类型工作 人员的失效概率;
根据所述失效概率确定总失效概率。
2. 根据权利要求 1所述的方法, 其中, 所述工作人员包括: 第一类工作人员、 第 二类工作人员和第三类工作人员, 其中, 所述第一类工作人员执行事故处理, 所述第二类工作人员监控机组状态参数的变化、 监控所述第一类工作人员的执 行情况, 并独立验证所述执行情况, 所述第三类工作人员独立检查机组状态、 判断事故性质、 评价机组的安全状态。
3. 根据权利要求 2所述的方法, 其中, 所述各个阶段包括: 监视阶段、 状态评估 阶段、 响应计划阶段、 响应执行阶段, 其中, 所述监视阶段包括所述工作人员 监视系统状态的转移,所述状态评估阶段包括所述工作人员评估监视到的状态, 所述响应计划阶段包括所述工作人员确定对监视到的状态所采用的响应策略, 所述响应执行阶段包括所述工作人员执行所述响应策略。
4. 根据权利要求 3所述的方法, 其中, 所述总失效概率为各个所述第一类工作人 员的所述各个阶段的总失效概率; 按照以下方式确定所述第一类工作人员的所 述各个阶段的总失效概率:
对所述各个阶段的失效概率采用两支事件树进行集成, 按照以下公式得到 所述第 一类 工 作人员 的 所述各个 阶段 的 总 失效概率 为 : Ftotai = P(A) + a - P(B) + ab- P(C) + a b-P(D) ; 其中, P (A) 为监视阶段的失效概率, a为监视阶段的成功概率, P (B ) 为状态评估阶段的失效概率, b为状态评估阶段的成功概率, P (C) 为响应计 划阶段的失效概率, c为响应计划阶段的成功概率, P (D) 为响应执行阶段的 失效概率。 根据权利要求 4所述的方法, 其中, 根据所述失效概率确定所述第一类工作人 员的所述各个阶段的总失效概率, 还包括:
根据如下公式调整所述第一类工作人员的所述各个阶段的总失效概率
Fttal, 得到所述第一类工作人员的最终的总失效概率 FT : FT = Fttal/(l_T), 其中, T是二类管理任务对人因可靠性的影响因子, T大于等于 0且小于 1。 根据权利要求 2至 5中任一项所述的方法, 其中, 所述总失效概率为所述第一 类工作人员、 所述第二类工作人员和所述第三类工作人员构成的班组的总失效 概率;
确定不同类型工作人员的失效概率包括: 获取所述第一类工作人员的第一 失效概率; 获取所述第二类工作人员的第二失效概率; 采用中等相关 MD根据 所述第二失效概率确定所述第三类工作人员的第三失效概率;
根据所述失效概率确定所述班组的总失效概率包括: 根据所述第一失效概 率、 所述第二失效概率和第三失效概率确定所述班组的总失效概率。 根据权利要求 6所述的方法, 其中, 按照以下方式确定所述第三失效概率:
MD,P(B/ A) = 1 + 6P(B\ 其中, P(B/ A)为所述第三失效概率, P(B)为所 述第二失效概率。 根据权利要求 7所述的方法,其中,按照以下方式确定所述班组的总失效概率: FCTew = PA x P(B/ A) x PB, 其中, FCTew为所述班组的总失效概率, ?八为所述第 一失效概率。 一种数字化主控室工作人员人因可靠性的确定装置, 包括:
第一确定模块, 用于获取工作人员对任务响应的各个阶段的失效概率, 或 者确定不同类型工作人员的失效概率;
第二确定模块, 用于根据所述失效概率确定总失效概率。 根据权利要求 9所述的装置, 其中, 所述第二确定模块, 用于对所述各个阶段 的失效概率采用两支事件树进行集成, 按照以下公式得到所述工作人员的所述 各个阶段的总失效概率为: Fttal = P(A) + a ' P(B) + al P(C) + ab P(D), 其中,
P (A) 为监视阶段的失效概率, a为监视阶段的成功概率, P (B ) 为状态 评估阶段的失效概率, b为状态评估阶段的成功概率, P (C) 为响应计划阶段 的失效概率, C为响应计划阶段的成功概率, P (D) 为响应执行阶段的失效概 率; 其中, 所述监视阶段包括所述工作人员监视系统状态的转移, 所述状态评 估阶段包括所述工作人员评估监视到的状态, 所述响应计划阶段包括所述工作 人员确定对监视到的状态所采用的响应策略, 所述响应执行阶段包括所述工作 人员执行所述响应策略。
11. 根据权利要求 10所述的装置, 其中, 所述第二确定模块, 还用于根据如下公式 调整所述工作人员的所述各个阶段的总失效概率 Fttal, 得到所述工作人员的最 终的总失效概率 FT : FT = Fttal/(l_T), 其中, T是二类管理任务对人因可靠性 的影响因子, T大于等于 0且小于 1。
12. 根据权利要求 9至 11中任一项所述的装置, 其中,
所述第一确定模块包括: 第一获取单元, 用于获取第一类工作人员的第一 失效概率; 第二获取单元, 用于获取第二类工作人员的第二失效概率; 确定单 元, 用于采用中等相关 MD根据所述第二失效概率确定第三类工作人员的第三 失效概率;
所述第二确定模块, 用于根据所述第一失效概率、 所述第二失效概率和第 三失效概率确定由所述第一类工作人员、 所述第二类工作人员和所述第三类工 作人员构成的班组的总失效概率;
其中, 所述第一类工作人员执行事故处理, 所述第二类工作人员监控机组 状态参数的变化、 监控所述第一类工作人员的执行情况, 并独立验证所述执行 情况, 所述第三类工作人员独立检查机组状态、 判断事故性质、 评价机组的和 安全状态。
13. 根据权利要求 12所述的装置, 其中, 所述确定单元, 用于确定所述第三失效概 率为 MD,P(B/ A) = 1 + 6(B), 其中, P(B/ A)为所述第三失效概率, P(B)为所 述第二失效概率。
14. 根据权利要求 13所述的装置, 其中, 所述第二确定模块, 用于确定所述班组的 总失效概率为: F£rew = PA x P(B/ A) >< PB, 其中, F£rew为所述总失效概率, PA为所 述第一失效概率。
PCT/CN2014/075738 2013-04-23 2014-04-18 数字化主控室工作人员人因可靠性的确定方法及装置 WO2014173259A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201310142589.4A CN103198232B (zh) 2013-04-23 2013-04-23 数字化主控室工作人员人因可靠性的确定方法及装置
CN201310142589.4 2013-04-23

Publications (1)

Publication Number Publication Date
WO2014173259A1 true WO2014173259A1 (zh) 2014-10-30

Family

ID=48720786

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2014/075738 WO2014173259A1 (zh) 2013-04-23 2014-04-18 数字化主控室工作人员人因可靠性的确定方法及装置

Country Status (2)

Country Link
CN (1) CN103198232B (zh)
WO (1) WO2014173259A1 (zh)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103198232B (zh) * 2013-04-23 2016-08-03 湖南工学院 数字化主控室工作人员人因可靠性的确定方法及装置
CN104965652A (zh) * 2015-06-15 2015-10-07 湖南工学院 一种人误概率计算方法和装置
CN104965978A (zh) * 2015-06-15 2015-10-07 湖南工学院 一种诊断失误概率计算方法和装置
CN104915116A (zh) * 2015-06-15 2015-09-16 湖南工学院 一种人误概率计算方法和装置
CN106950850A (zh) * 2017-02-20 2017-07-14 上海核工程研究设计院 一种数字化仪控系统动态可靠性集成分析方法
CN109284925A (zh) * 2018-09-21 2019-01-29 南华大学 一种班组情景意识的测量方法、装置、设备及存储介质

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103198232A (zh) * 2013-04-23 2013-07-10 湖南工学院 数字化主控室工作人员人因可靠性的确定方法及装置

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103198232A (zh) * 2013-04-23 2013-07-10 湖南工学院 数字化主控室工作人员人因可靠性的确定方法及装置

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
JIANG, JIANJUN ET AL.: "Research on Cognitive Reliability Model for Main Control Room Considering Human Factors in Nuclear Power Plants", NUCLEAR POWER ENGINEERING, vol. 33, no. 1, 29 February 2012 (2012-02-29), pages 66 - 72 *
LI, CHENGPENG ET AL.: "Methodology for Analysis of Dependency between Operators in Digital Control System in a Nuclear Power Plant", NUCLEAR POWER ENGINEERING, vol. 32, no. 6, 31 December 2011 (2011-12-31), pages 17 - 21 *
ZHOU, YONG ET AL.: "Effects of Digitalized Main Control Room on Operators' Implementing Situation Assessment Tasks in Nuclear Power Plant", CHINA SAFETY SCIENCE JOURNAL, vol. 23, no. 1, 31 January 2013 (2013-01-31), pages 41 - 46 *

Also Published As

Publication number Publication date
CN103198232B (zh) 2016-08-03
CN103198232A (zh) 2013-07-10

Similar Documents

Publication Publication Date Title
WO2014173259A1 (zh) 数字化主控室工作人员人因可靠性的确定方法及装置
Park et al. An experimental investigation on relationship between PSFs and operator performances in the digital main control room
Lee et al. Development of a qualitative evaluation framework for performance shaping factors (PSFs) in advanced MCR HRA
Sundaramurthi et al. Human reliability modeling for the next generation system code
CN103218689B (zh) 操作员状态评估的可靠性分析方法及装置
Zou et al. Human reliability analysis for digitized nuclear power plants: case study on the LingAo II nuclear power plant
Kim et al. An analytic model for situation assessment of nuclear power plant operators based on Bayesian inference
Rashid et al. Eradicating root causes of aviation maintenance errors: introducing the AMMP
WO2014173276A1 (zh) 通过hra判定dcs人机界面的可靠性的方法、系统
WO2014173258A1 (zh) 响应计划的可靠性分析方法及装置
CN107016457A (zh) 一种实现社区危险行为预警的系统和方法
Li et al. An assessment method of operator’s situation awareness reliability based on fuzzy logic-AHP
Li et al. Study on operator’s SA reliability in digital NPPs. Part 3: A quantitative assessment method
WO2014173275A1 (zh) 核电厂数字化主控室操作员监视行为可靠性判定方法
Wu et al. A review of alarm system design for advanced control rooms of nuclear power plants
Ekanem A model-based human reliability analysis methodology (Phoenix method)
Li et al. Study on operator's SA reliability in digital NPPs. Part 2: Data-driven causality model of SA
Lee et al. A computational model for evaluating the effects of attention, memory, and mental models on situation assessment of nuclear power plant operators
Zhang et al. Developing a taxonomy and a dependency assessment model of performance influencing factors for intelligent coal mines
Yan et al. Using artificial neural network for predicting and evaluating situation awareness of operator
Zhang et al. The simulator experimental study on the operator reliability of Qinshan nuclear power plant
CN115828607A (zh) 一种基于多智能体的人机环协同建模方法
Wu et al. Safety archetypes identification and behavior simulation for nuclear power plant operation human reliability improvement
Park et al. Investigating the effect of task complexities on the response time of human operators to perform the emergency tasks of nuclear power plants
Petkov Symptom-based approach for dynamic HRA and accident management

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 14787820

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 14787820

Country of ref document: EP

Kind code of ref document: A1