Analysis Of Soft Control Error At NPPs Main Control Room Based On Eye Movement Experiment

With the rapid development of digitalized technologies,computer-based soft control in the main control rooms of modern industries such as nuclear power plant has become an inevatable trend. The operators will obtain state information on system operation and dispatch orders of control. The quality of man-machine interface and skills of operators in the main control room will have direct influence on the safe operation of systems. However, so far the designs of computer interface in most main control rooms have greatly depended on the experience and judgement of designers, less of experimental foundations and quantitative methodologies for the validation and verification of the computer interface,and difficult to instruct the improvement of man-machine interface design.This thesis took the Chemistry and Volume Control System (CVCS) as an example, a design of soft control interface was presented. The structure, working principle and operating procedures of CVCS were analyzed. Based on the relevant national and international standards, the design principles and design process for the interface of CVCS were sumarized.According to the cognitive principles, the design rules such as interface layout and semiotic meanings of graphcial symbols were given. The division of functional areas as well as the design of menus, tool bars, system flow charts, shortcut symbols and dialog boxes were implimented. Two types of soft control interfaces including a system structure-oriented interface and a task-oriented interface were developed. The human-machine interactions for soft controls were realized.On the basis of above mentioned works, an eye movement experiment was conducted for evaluating the performance of the developed soft control interfaces of CVCS. In this thesis,a block of operating procedures of starting CVCS under the reactor normal cold-shutdown condition was selected and the experement schedule was designed. 10 testee were devided into two groups including a professional group and a normal group according to their academic background. The experiemnt of testee implimenting operating procedures through the soft control interfaces as well as their eye movements during the expeirement were recorded.Finally, the human errors of testee during eye movement experiment were identified. The number and the failure probability of each human error mode were counted.The experiment results were analzyed and discussed. Some suggestions for human-manchine interface design and optimization were also given.The research results of this thesis show that eye movement experiment can be used to identify the human error modes of soft control in the main control room, and is helpful for the statistical data processing and analyzing. The proposed technology can provide not only an objective basis for the performance evaluation of the soft control interface,the skills of operators implimenting the soft controls,but also instructive suggestions for the improvement and optimization of soft control interface.