Markov reliability model research of monitoring process in digital main control room of nuclear power plant☆
Research highlights
► This paper proposes a reliability Markov model that can analyze next monitoring object probability in terms of current information and state. ► The authors in the paper divide digital human–machine interface into two parts that are referred as logical homogeneous Markov and logical heterogeneous Markov. ► For homogeneous Markov model, the authors propose many dynamic functions. ► For heterogeneous Markov model, the authors consider the association degree of monitoring information.
Introduction
The monitoring process is an important part in digital human–machine interface, it is referred to as the action (Kim and Seong, 2006) of extracting information from surrounding correctly. The authors mainly research the transition probability of monitoring process in digital human–machine interface. Digital is considered as the flexible ways provided by computer. Digital system has converted the way by common detecting into the method by computer station in control room. From the capability offered and handled information, computer work platforms have tremendous superiorities. On the contrary, computer working consoles bring new challenges to operators. The challenge particularly focuses on the changing of monitoring process. So, in digital system control room, the operators not only monitor the traditional information sources (Vicente et al., 2001) that mainly include shift turnover, testing, control room panels, field operators, other units, log, alarm screens, CRT displays and Field tour, but also monitor that thousands of complex and dynamic information derived from computer. Thus, previous monitoring models and methods have not been adapted for digital system monitoring reliability evaluation in control room. To facilitate monitoring and decrease events, the authors build a monitoring transition probability model and present some functions in this paper. The paper has two prominent contributions that are listed as follows: (1) The model exploited by Markov chain can effectively assess the transition probability of the next monitoring object; (2) The model is formed under certain conditions that include human factors, other relating factors and the specific field of digital human–machine interface in monitoring process; (3) The authors propose many novel methods and dynamic functions.
Section snippets
Review of monitoring process researches
In fact, some researchers have proposed many monitoring methods and models toward monitoring tasks in human–machine interface. Such as, in 1980, Sheridan (1980) developed a mathematical model of monitoring behavior for automated systems. The model depicted how monitor behavior is allocated in time to multiple task. In 1983, Rasmussen (1983) proposed two basic models in monitoring: the situation model and the mental model. An accurate mental model is viewed as a limited feature of expert
The description of model
In reliability analysis, we may well say that the study considering human factors and other relating factors in monitoring digital human–machine interface is a blank yet. In this paper, the authors fill a vacancy in the field.
In this paper, the authors propose the logical partition theory that refers as that if the components belong to the same category, we will divide them into a block, otherwise, we will divide them into an other block. For digital human–machine interface, there are thousands
The proposed method
To deeply understand homogeneous Markov transition probability method, the authors firstly consider a structure with three conditions. The model structure is shown in Fig. 2.
The model structure shown in Fig. 2 is called Hidden Markov Model (HMM) with conditions. According to the model structure, homogeneous Markov transition probability, namely, P{objectn+1, Hi, Sj, Ak|Objectn}, is defined as:
For the convenience of
The proposed method
For heterogeneous Markov transition probability, we write it as:
For the convenience, we divide Eq. (14) into and . ζ1 can be determined by Eq. (15):
In the three influence factors, human factors is a constant. f(Cs) and f(Ca) can be obtained as follows respectively:where Rs = Cp/CS, Ra = Cp/Ca.
According to Eq. (8), human factors influence, namely p(Hi),can be obtained as
Conclusions
With the current increasing human factors accidents in monitoring process, the authors propose a transition probability calculation model and corresponding methods in term of human factors and other relating factors. Proposed methods can successfully assess the next monitoring object probability so that human factors accidents can be greatly decreased by the methods. The calculation methods proposed by the authors are also an excellent means to estimate transition probability, and consider many
Acknowledgments
The research presented in this paper is supported in part by National nature Science Foundation (Nos. 70873040, 71071051) of China, and research projects of Ling Dong Nuclear Power Company Ltd. (KR70543) of China.
References (32)
- et al.
Analysis of sequential failures for assessment of reliability and safety of manufacturing systems
Reliability Engineering & System Safety
(2002) Assessment of human reliability factors: a fuzzy cognitive maps approach
International of Industrial Ergonomics
(2007)- et al.
Construction of event-tree/fault-tree models from a Markov approach to dynamic system
Reliability Engineering & System Safety
(2008) - et al.
Cognitive modeling and dynamic probabilistic simulation of operating crew response to complex system accidents, part 4: IDAC causal model of operator problem-solving response
Reliability Engineering & System Safety
(2007) - et al.
Assessing new product development project risk by Bayesian network with a systematic probability generation methodology
Expert System with Application
(2009) - et al.
A generic method for estimating system reliability using Bayesian networks
Reliability Engineering & System Safety
(2009) - et al.
Automatic creation of Markov models for reliability assessment of safety instrumented systems
Reliability Engineering & System Safety
(2008) - et al.
A computational model for knowledge-driven monitoring of nuclear power plant operators based on information theory
Reliability Engineering & System Safety
(2006) - et al.
A computational model for knowledge-driven monitoring of nuclear power plant operators based on information theory
Reliability Engineering & System Safety
(2006) - et al.
Risk-based reconfiguration of safety monitoring system using dynamic Bayesian network
Reliability Engineering & System Safety
(2007)