Upper Limb Muscle Fatigue Detection And Analysis Based On Surface Electromyography Signals

The single rehabilitation plan of rehabilitation robots and excessive rehabilitation training can easily lead to muscle fatigue of patients in rehabilitation treatment,thereby causing secondary injury to the patient’s limbs.Therefore,the patient’s limbs are fatigued during the rehabilitation training of stroke patients.Testing is extremely important.Aiming at the problem of muscle fatigue detection,an experimental method for real-time detection and analysis of elbow joint EMG signals was established;the characteristics of muscle fatigue based on EMG signals were studied;a mathematical model for evaluating muscle fatigue status was established.Based on this method,the Plux company in Portugal was selected.The 8-channel Ergo Plux wireless surface electromyography measurement device collects,obtains the surface electromyography signal of the upper limb of the human body,preprocesses it,and obtains the denoised surface electromyography signal for analysis.The method of identifying the fatigue state of human upper limb muscles is systematically discussed.The physical therapist can intervene,control and adjust the rehabilitation treatment plan in real time based on such information to ensure the effectiveness and safety of the rehabilitation training.The research content and results of this article are as follows:(1)According to the analysis needs of upper limb surface EMG signal,carry out the human upper limb EMG signal acquisition experiment.The surface EMG signal of the upper limbs of the human body was selected as the research basis.Six boys were selected as the research objects.The 8-channel Ergo Plux wireless surface EMG measurement equipment produced by Portugal Plux Company was used for EMG signal acquisition.The 50 Hz power frequency notch and Savitzky·Golay were used.Methods The signal is preprocessed,and a wavelet threshold denoising method is proposed,and the pros and cons of the three wavelet threshold denoising methods are compared and analyzed.Research shows that the default threshold denoising method based on wavelet decomposition has the best preprocessing effect on s EMG.(2)In response to the need for quantitative analysis of the elbow joint muscle fatigue evaluation algorithm,the surface EMG signal acquisition experiment of the elbow joint flexion motion under different loads was carried out.By analyzing the elbow joint surface EMG signal of 8 test subjects,the average frequency was compared and analyzed.(MNF),Spectral Distance(SMR),Wavelet Method WIRM1551,Fuzzy Approximate Entropy(f Ap En)and Recursive Quantitative Analysis(RQA%DET)five methods,verifying the ability of the five methods to distinguish the degree of fatigue and anti-interference ability,calculation The results show that the spectral distance SMR is superior to other algorithms in terms of fatigue discrimination and anti-interference.(3)Put forward the idea of studying the classification of upper limb muscle fatigue state.After processing the collected human upper limb surface electromyographic signal,the feature extraction is performed,and the wavelet coefficients in the time and frequency domain are selected as the signal feature.Through the analysis of support vector machine(SVM)and The BP neural network algorithm realizes the recognition efficiency of upper limb muscle fatigue state classification,that is,the accuracy rate,which verifies the feasibility of using support vector machine(SVM)to classify upper limb muscle fatigue state.The calculation results show that the accuracy rate of support vector machine(SVM)in classifying upper limb muscle fatigue state can reach 89.1%.(4)Put forward the idea of designing the human upper limb muscle fatigue state detection system.By analyzing the design function requirements of the muscle fatigue state detection system,proposed the processing flow of the muscle fatigue state detection system,combined with the My SQL database to realize the management of the personal information of registered users,and passed Matlab GUI interface interactive mode realizes real-time recognition of muscle fatigue state.The feasibility of the muscle fatigue detection system is verified.