The Study Of Model Based On EEG Emotional Judgment

China's machinery manufacturing is the foundation of all industries,in machinery manufacturing,workers in harsh environments,such as noise and dust,and big working strength,easy to cause strong emotion.Based on the findings of a study of emotion and safety accidents,put forward the real-time detection of workers emotion model,we can know the emotional state of every worker,and in a timely manner to adjust.Research shows that people's behavioral decision depends in part on people's emotional state.Using eeg signals for judgment of emotion is an important research in the field of intelligent computing in recent years.The purpose of this paper is to study the mechanical manufacturing process of noise generated by the impact on staff mood,emotion to make use of eeg signals to judgment.Emotional judgment model based on eeg studies is a blend of many discipline theory knowledge is an important topic.On the basis of in-depth studies and researches of varieties of eeg feature extraction methods and classification algorithms,for the study of model based on eeg emotional judgment.We use the noise and music clips as stimulus material manufacturing,different emotions induced by volunteers.First,we use single electrode Mindwave equipment acquisition eeg signals,the FIR filter is adopted to brain electrical signal filtering,make the brain electrical signal to achieve application level;Secondly,the use of approximate entropy and wavelet energy entropy of eeg feature extraction,innovation is the first to approximate entropy and wavelet energy entropy to form the characteristics of the combination judgment;Again,using the BP neural network,ELM limit learning,probabilistic neural network(PNN)and generalized regression neural network(GRNN),four kinds of artificial neural network model to simulate the brain electrical signal classification judgment;Finally,in the emotional judgment model is established on MATLAB and verify.Through research,it is concluded that the model can achieve ideal result,recognition rate can reach more than eighty percent,this method is expected to get a good application in the field of machinery manufacturing.