Research On Petri Nets Modelling Theory And Method Of Complex Dynamic Monitoring Task Workload

Complex dynamic monitoring task is a kind of complex behavior activity set with high information,time and space operation uncertainty,which widely exists in the safety-critical systems such as train/aircraft driving,traffic dispatching,air control,and nuclear power monitoring.In recent years,with the continuous application of intelligent equipment,the connotation of the task is constantly changing,the traditional system security risk concentration point is changing from machine to human.How to ensure the performance of complex dynamic monitoring task,building a safety evaluation system combined with man and machine has become a complicated man-machine system design problem to be solved.As one of the important ergonomic indexes,workload is generally regarded as an effective measure of human-computer interaction task design level at present,and Petri Net is widely used in modeling and analysis of complex dynamic systems.Therefore,on the background national natural fund project"Task-based measurement and optimization of complex human-computer interaction system manipulation adaptability"and the Beijing municipal science and technology plan projects"Key technology research and demonstration application in Full Automatic Operation in urban rail transit system.",from the perspective of complex dynamic monitoring task workload,this paper introduces the formal method and uses Colored Petri Net to explore and demonstrate the feasibility and effectiveness of continuous monitoring task workload measurement.On this basis,through in-depth analysis of the connotation and expression of the workload and the mapping relationship with Petri Net structure,a concurrency short-term multi-task workload prediction model based on Time Petri Net was established and verified by experiments in the laboratory environment.Finally,taking the manual driving task in departure phase under large area behind time in emergency scenario of metro fully-automatic operation as example,this paper discusses the optimization of time series for concurrency short-time multi-task,proves the feasibility of engineering application of prediction model,provides theoretical method and technical support for the optimization design of human-computer interaction task of complex system,and has very important theoretical and practical significance.The thesis has completed the following main work:(1)The definition,evaluation method,prediction model and application status of workload were deep analyzed,and the problems in workload related research were summarized.Taking railway industry and ergonomic application as the medium,this paper analyzes the relevant background research of Petri Net and workload,points out its potential connection,and determines the research purpose,method and technical route.(2)The complex dynamic monitoring task and its applicable environment,classification and characteristics are defined.This paper analyzes the components and workload characteristics of complex dynamic monitoring tasks,discusses the internal relations between Petri Nets and workload,and discusses the feasibility of using Petri Nets to describe workload.(3)On the basis of VACP and Multiply Resource Theory,a behavior-based workload measurement model for task process under Colored Petri Net was established.By introducing the influence of repeated behaviors on the workload during the task process,the measurement accuracy was effectively improved,and the model was verified in the MATB-? multi-task environment.(4)Aiming at the concurrency short-term multi-tasks in human-computer interaction in complex systems and on the basis of behavior based workload measurement model of task process,a further workload prediction model for resource allocation under Time Petri Net is constructed.The feasibility of applying the model to human-computer interaction task optimization is demonstrated by designing and implementing relevant experiments.(5)A test platform of metro driving tasks is developed,and the time sequence optimization of the emergency manual off-station task of metro is realized by applying the short-term parallel multi-task workload prediction model.The difference before and after the task optimization is analyzed through the platform simulation experiment,and the engineering application of the model is discussed.