Research On Muscle Fatigue State Analysis System Based On SEM

The detection and prevention of muscle fatigue is very important for the weight-bearing and training groups.This article aims to study the muscle fatigue analysis system based on the surface EMG signal and the muscle p H value when the muscle is fatigued.In the process of muscle weight-bearing exercise,energy conversion occurs in muscle cells,and the muscles gradually become acidic and gradually fatigued.In this paper,the muscle force analysis software Any Body is used to analyze the force of the biceps brachii,which solves the shortcomings of the previous research that cannot analyze the force of a single muscle.Due to the trauma of the muscle p H test,this topic uses the system dynamics model to calculate the muscle p H value calculation method,calculates the muscle p H value according to the muscle force value,and divides the muscle fatigue into five grades,which are in turn based on the muscle p H.Determine the state of muscle fatigue and solve the problem of measuring the degree of fatigue according to Borg subjective evaluation in previous studies.After determining the muscle fatigue standard,the surface EMG signal is collected in the experiment,the eigenvalues are analyzed and the dimensionality reduction is processed,and the eigenvalues are used as input.The muscle fatigue level is used as the output to train the DP＿PSO＿SVR fatigue regression model.The final result shows that the model has an accuracy of more than 95% in analyzing the fatigue degree.The research ideas mainly include the following four parts:(1)According to the characteristics of the surface EMG signal,design a signal conditioning circuit,complete the collection and processing of the surface EMG signal,and analyze and process the characteristic value of the surface EMG signal.(2)Use Any Body software to establish a personalized human body model,import the tensile force value in the experiment,and obtain the biceps muscle strength data under different loads through inverse dynamic analysis.(3)Establish a muscle p H calculation formula,calculate the muscle p H of the biceps brachii under different forces,and classify muscle fatigue levels to provide a benchmark for fatigue analysis model training.(4)The DP＿PSO＿SVR model is used to establish a non-linear mapping relationship,with the characteristic value of the EMG signal as the input and the fatigue level as the output to complete the training of the muscle fatigue prediction model.The comparison with the prediction effect of the BP neural network model and the random forest regression model verifies the validity and accuracy of the DP＿PSO＿SVR model.Through the EMG analysis of 100 samples of 15 subjects,the results show that the accuracy of the biceps fatigue level analysis is more than 95%,which is compared with the accuracy of the BP neural network model and the accuracy of the random forest regression model.The 90% accuracy is higher,and the fatigue state analysis is superior.The research results show that the combination of s EMG and muscle p H can accurately complete the analysis of muscle fatigue.The muscle fatigue analysis system completed in this article can analyze and predict the muscle fatigue status of rehabilitation patients,and also provides human rehabilitation medicine and scientific training.Important scientific basis.