Research On A Reliability Analysis Methodology Of Human-Computer Interaction In Advanced Main Control Room Of NPP

Ensuring nuclear safety is the basis and prerequisite for the operation of Nuclear Power Plants(NPPs).Human error is one of the main factors threatening the safe operation of nuclear power plants.The Main Control Room(MCR)is the central place for monitoring and controlling the operation of a NPP.The main control room operators are responsible for ensuring the safe operation of NPPs.Therefore,preventing the human errors of main control room operators is a major concern of any nuclear power plant.With the increasingly extensive application of digital technology in nuclear power plants,more and more main control rooms in nuclear power plants have adopted large screens,soft controls and digital operation procedure systems,which are called “advanced main control rooms”.Compared with the traditional main control rooms,the human-computer interaction means and human-computer interaction modes in advanced main control rooms have changed significantly.The design experience from the traditional main control rooms can not cope with the new features of the advanced main control rooms.Therefore,it is a hot issue in the nuclear industry to carry out human reliability analysis as early as possible in the early stage of the design of an advanced main control room,so as to timely find human factors engineering problems in the design of the advanced main control rooms,and avoid major design changes caused by the problems found in the late design of the advanced main control room.This thesis focuses on the human reliability analysis in the advanced main control room of nuclear power plant,which is a major issue that is generally concerned by nuclear industry.Starting from the data collection of human error in nuclear power plants,a comprehensive method for modeling,simulation and quantitative analysis of human error in advanced main control rooms of nuclear power plants was studied.Also through case analysis,how to use this comprehensive method for human error analysis to solve practical engineering problems was discussed.The following research works have been carried out in this thesis:(1)Study on a human error detection method and design of a human error detection system of NPP: the current analysis of operation logs in the main control room of NPP depends on manual operation with low efficiency and insufficient depth of data mining.A process and method for detecting operator human error using the operation log data of the main control room of NPP is studied.Based on the operation information and operator actions in the operation log,the operating conditions of the NPP and the standard operating procedures that should be implemented are identified.Based on this,the deviation between operators’ actual operation actions and standard operating procedures as well as the potential human errors that may exist in the operation log can be identified...(2)Study on a modeling and analysis method for human error of soft control in advanced main control room: Aiming at the current human error analysis method for NPPs,which is based on the development of simulated main control rooms and does not take into account the current situation of the technical characteristics of existing digital main control rooms(or advanced main control rooms),the thesis takes the advanced main control room as the object,and focuses on the basic concepts such as the characteristics of human error,the mode classification of human error,and task sequences of the advanced main control room operators in the context of implementing soft control,the modeling and quantification methods of human error probability for operators performing soft control are studied..(3)Research on human-computer interaction modeling and dynamic simulation technology: in view of the current situation that relatively independent and static analysis methods are used in the design of human-computer system in NPPs,and the lack of comprehensive and dynamic analysis methods for various human-computer elements and indicators,the thesis focuses on the characteristics of human-computer interaction in the advanced main control room of NPPs to perform operation tasks,and studies a human-computer interaction modeling and simulation method.By describing the personnel,equipment,task and other basic human-machine elements as well as the interaction characteristics between the elements are combined with the task complexity measurement and the failure probability of human-machine elements,a comprehensive and dynamic analysis of the workload of the operation team and the task reliability is presented.(4)Analysis of the impact of advanced computerized operation procedure system on human error: in view of the current situation that NPPs have doubts about the use of computerized intelligent diagnosis function and lack of demonstration means,the thesis focuses on the technical characteristics of the use of computerized operating procedure system in the advanced main control room,and studies the impact of automatic diagnosis function in the computerized operating procedure system on the decision-making of the main control room operation team and the workload of the operation team members.By integrating operators into the control loop and implementing collective decision-making,a scheme of collective decisionmaking between computer and operators is presented,and the effectiveness of collective decision-making scheme is compared and analyzed.The following research results have been achieved:(1)An intelligent human error detection process and method based on the operation log of the main control room is proposed,which can excavate the subtle human error hidden in the operation process of NPPs,and improve the digital level and work efficiency of the operation and maintenance work of NPPs.(2)A soft control human error prediction method is presented,which comprehensively considers the characteristics of digital manipulation technology,supports qualitative and quantitative analysis,can reveal weaknesses in the digital design and human-computer interaction of the main control room,and supports probabilistic safety analysis and humancomputer system design improvements in NPPs..(3)A dynamic human-computer interaction modeling and simulation method for the main control room is proposed,which comprehensively considers the basic human-computer elements of the advanced main control room.On the basis of realizing the dynamic simulation function of human-computer interaction scenarios,it can support the development of workload analysis and task reliability analysis for the operation team.(4)A human-machine collective decision-making scheme based on automatic diagnosis function is proposed,which balances the advantages and disadvantages of computer and human operators in decision-making,and has a more reliable decision-making ability than existing operating teams.The human error comprehensive analysis method proposed in this thesis can not only support the probabilistic safety analysis of NPPs,but also play a technical supporting role in the early stage of advanced MCR design to optimize the performance of operators.The method can also help reduce the probability of human error of operators by improving the design of human-computer interface,identify and improve the means to correct the probability of human error of operators,and evaluate the impact of artificial intelligence technology on human error of operators.Therefore,it has good engineering application value.