Research On Cognitive Load Recognition Of High-speed Rail Dispatcher In Monitoring Work

The perception and risk identification of high-speed rail dispatchers on the driving order and driving environment is the primary link of a complete operation unit.The lack of monitoring and the missed observation of signals and information or the wrong identification of signals and information is the primary root cause of human error.The improvement of the degree of automation has profoundly changed the interaction mode between the HSR dispatcher and the centralized dispatching control system.The interaction mode of HSR dispatching work,especially the dispatching monitoring work,is prone to the decline of situational awareness and the phenomenon of "loss of the loop".Detecting and identifying the cognitive load in the monitoring work of high-speed rail dispatchers is a key technical issue to improve the interaction of the dispatching control system and ensure the effective supply of cognitive resources related to dispatcher monitoring.Firstly,based on task analysis and cognitive model analysis,this study designs a highspeed rail dispatcher monitoring work information perception processing density calculation model,calibrates the cognitive load of tasks from the perspective of cognitive time requirements of different cognitive functions,and conducts standard task tests to measure Time weights for each cognitive activity.On this basis,a simulated scheduling monitoring experiment under the guidance of different automation levels was designed.Evaluation verification.On the basis of the collected data,the EEG signal characteristics reflecting the changes in the cognitive load of high-speed rail dispatching and monitoring work were extracted,screened and analyzed.Based on this feature,a backward method extreme learning machine recognition model is designed.The experimental verification and analysis results show that(1)The main cognitive activities of high-speed rail dispatch monitoring are perception and judgment,and the power characteristics of EEG signal frequency bands in the right occipital lobe and left frontal lobe of the brain corresponding to visual and cognitive cognitive functions,especially the 4-8Hz frequency band for high-speed rail dispatching.Cognitive load changes in monitoring work respond significantly.The neurophysiological phenomenon provides strong evidence for the cognitive load analysis of high-speed rail dispatching and monitoring work,and achieves the unity of objective task measurement,subjective task evaluation and physiological response,and the information perception processing density calculation model determined based on task analysis and cognitive pattern analysis is correct.The lack of taskguided monitoring work patterns showed neurophysiological evidence that non-task cognitive activities occupy working memory capacity.(2)The back-off method extreme learning machine classification and recognition model based on feature importance ranking well retains the main information in the EEG signal features reflecting the cognitive load changes of high-speed rail dispatchers,and achieves a higher accuracy with fewer feature dimensions.The overall accuracy rate is 86.2%.The model can be used to study the cognitive load changes during monitoring work of dispatch,and to optimize the human-computer interaction mode of dispatch.