Based On The Characteristics Of Power And SEMG Changes To Evaluate The Fatigue Of The Inferior Extensor Isotonic Extreme Endurance

Objective: Muscle fatigue is an inevitable physiological phenomenon in sports and daily work and life.Its appearance will affect sports performance,production efficiency and so on.Muscle fatigue can be directly measured by quantitatively detecting the output muscle strength or power retention ability,but the mechanical parameters lack a comprehensive analysis and evaluation of the functional state of the neuromuscular system.Therefore,this study is based on the direct measurement method of muscle fatigue,combined with electromyogram to compare the linear and non-linear change characteristics of sEMG signal to monitor the change characteristics of the output power during the dynamic work fatigue process,and then comprehensively evaluate the muscle fatigue state.It has important methodological significance for the basic research of neuromuscular system fatigue detection.Method: The experiment selected 40 healthy male college students who had never experienced systemic muscle strength training.Use BTE Primus equipment to perform isometric peak torque and isotonic peak endurance tests on the ankle joint plantar flexors,knee extensors and hip extensors.Synchronously collect and record the kinematics and dynamics parameters of the corresponding exercise links,as well as the sEMG signal data of the extensor muscle group,and then the sEMG signals are standardized.The linear regression power%prediction model is constructed by calculating the time domain,frequency domain,timefrequency parameters of each muscle sEMG,and the maximum output power change rate in each stretching phase.By calculating the average absolute value of sEMG,the rate of slope sign changes,the zero-crossing rate and the wavelength of the four time-frequency characteristic values,a nonlinear neural network prediction model is constructed to obtain the output power%prediction value.Finally,the F test is used to compare the difference of power% predicted by different methods.Results:(1)The order of a single joint IPT measurement results for the lower limbs extensor muscles is hip joint,knee joint,ankle joint,and there are significant differences in IPT between different joints(P<0.05).(2)The results of the ultimate muscle endurance holding test under the condition of 50% IPT resistance load show that the hip extensors persistence time is the longest,and the ankle joints are slightly second,the knee extensors is the shortest.Compared with the number of repetitions,the output power of the extensor muscle group and the timefrequency parameters of sEMG have a negative correlation in the process of extreme muscle endurance fatigue.(3)The power% of all extensors is negatively correlated with RMS% and MAV%,and positively correlated with MNF%,MDF%,IMNF%,and IMDF%.In addition,compared with traditional sEMG time-domain and frequency-domain parameters in tracking local muscle fatigue,time-frequency parameters provide higher accuracy to map power output loss.(4)Compared with using single or combined sEMG parameters to predict dynamic working muscle power%,the nonlinear neural network model shows a higher signal-to-noise ratio and coefficient of determination.Conclusion:(1)The isotonic peak muscle endurance maintenance ability of the lower limbs single joint can be tested by using the test plan of this research.The 50% IPT load intensity and the action frequency of the respective joints were used to perform rapid stretching exercises of 45 to 60 times per minute to record the exercise time or the number of stretches.(2)With the number of repetitions increases,the power and sEMG instantaneous frequency parameters of each joint extensor muscle group in the isotonic peak muscle endurance fatigue process show a decreasing trend.In the process of repetitive stretching exercises,dynamic muscles aggravate the non-stationarity of sEMG signals,and the accuracy of time-frequency parameters in predicting and evaluating local muscle fatigue is better than time-domain and frequency parameters,and the prediction performance of IMDF is better than IMNF.(3)The nonlinear neural network model and the combination of multiple EMG parameters to obtain multiple linear regression models have better goodness of fit and signal-to-noise ratio than using a single sEMG parameter to predict dynamic muscle power changes.Compared with the multiple linear regression model,the nonlinear neural network model predicts the loss of muscle power better.