Research On Fatigue Detection Method Based On Portable EEG Acquisition Equipment With Its System Implementation

The work of this paper is from the health and safety monitoring technology research project of Guangdong Power Grid,in which the fatigue detection module is the core content of the project.Globally,mental fatigue has become one of the important causes of emergencies in work scenario,for instance long-term fatigue driving,casualty accidents of steeplejacks,palpitations caused by insufficient rest and so on.The effective detection towards human mental fatigue can help people find the fatigue state in time and reduce the unexpected safety accidents in the working scene.Among many physiological indexes of human body,EEG signal,as the central nervous signal,can objectively reflect people’s mental state.Based on EEG signals and portable EEG acquisition equipment,this paper studies the detection of human mental fatigue by studies on fatigue detection based on EEG signal have the following problems.At present,studies on fatigue detection based on EEG signal have the following problems: Firstly,most of the existing fatigue detection schemes are based on the whole-brain or multi-channel EEG in the laboratory environment,and are mostly wet electrode cable acquisition equipment.The preparation before detection is cumbersome and complex,can not meet the portable,simple operation and so on under the outdoor scene of fatigue testing requirements.Also,it can not meet the practical application.Secondly,most of the existing fatigue detection schemes are single-machine running devices,without real-time transmission of fatigue detection results and closed-loop management of fatigue detection data from the client to the management end.Aiming at the problems above,the research contents of this paper mainly include:(1)The eeg acquisition scheme based on a portable 4-channel EEG acquisition device was studied,and the event-related potential experiment was designed,which was compared with the 64-256 channel EEG acquisition device of Neuroscan Syn Amps2,to verify the availability of the portable device and lay a hardware foundation for subsequent research and system development.(2)In this study,the mental fatigue calculation methods based on different frequency domain characteristics were compared.Through the design and implementation of simulated fatigue driving experiment and comparison experiment,it was found that(α+θ)/β could more significantly represent the mental fatigue state,and a complete set of mental fatigue detection scheme was proposed.It was also found that the values of(α+θ)/β tended to be stable after 3 hours of simulated fatigue driving.(3)The research compares the mental fatigue calculation methods based on different frequency domain features,and through the design and implementation of simulated fatigue driving experiments and comparative experiments,it is found that(α+θ)/β can more significantly represent the mental fatigue state,and based on this,the proposed A complete set of mental fatigue detection programs.At the same time,it is found that the value of(α+θ)/β tends to be stable in the late stage of simulated fatigue driving for up to 3 hours.