Muscle fatigue prediction method and system based on BP neural network model

Abstract

The invention relates to the technical field of muscle fatigue prediction, in particular to a muscle fatigue prediction method and system based on a BP neural network model, and the method comprises the steps: carrying out the filtering processing of original electromyographic signal data, and obtaining the electromyographic signal data; according to the maximum equal-length shrinkage corresponding to each individual and the root-mean-square value of each individual, obtaining standard electromyographic signal data of each individual; the integral myoelectricity value and the wavelet packet entropy of each muscle are obtained; optimizing the BP neural network configuration to obtain an optimized BP neural network structure; training the optimized BP neural network structure according to the integral myoelectricity value of each muscle, the wavelet packet entropy, the myoelectricity signal data set and the scoring result of the fatigue degree of each muscle to obtain a trained BP neural network structure; and performing muscle fatigue prediction through the trained BP neural network structure. According to the invention, the accuracy of muscle fatigue prediction is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of muscle fatigue prediction, and particularly to a method and system for predicting muscle fatigue based on a BP neural network model. Background Art

[0002] In traditional muscle fatigue prediction and analysis, subjective scoring methods are often used; however, traditional subjective scoring methods (such as questionnaires) lack objectivity and real-time performance, and it is difficult to accurately quantify the muscle fatigue state in dynamic operations (such as overhead maintenance operations); secondly, as a one-dimensional time series, surface electromyography (SEMG) signals have problems such as large noise interference and significant individual differences, and the raw data is messy and cannot be directly used for model input; finally, traditional BP (Backpropagation) neural networks are sensitive to the setting of initial weights and thresholds, are prone to falling into local optima or having a slow convergence speed, which affects the prediction accuracy. Summary of the Invention

[0003] The present invention provides a method and system for predicting muscle fatigue based on a BP neural network model, which solves the problems of subjectivity, poor real-time performance, large noise interference, non-standard data, and insufficient accuracy caused by the convergence problem of traditional BP neural networks in traditional muscle fatigue prediction methods.

[0004] The object of the present invention can be achieved by the following technical solutions: In the first aspect of the present invention, a method for predicting muscle fatigue based on a BP neural network model is provided, including: Determining the range of the number of individual samples for electromyophysiological tests; Obtaining a number of groups of raw electromyography signal data of the subjects; filtering the raw electromyography signal data to obtain the finally filtered signal data, denoted as electromyography signal data; obtaining the maximum isometric contraction corresponding to each individual in the electromyography signal data and the root mean square value of each individual; obtaining the standard electromyography signal data of each individual according to the maximum isometric contraction corresponding to each individual and the root mean square value of each individual; obtaining the integrated electromyogram value and wavelet packet entropy of each muscle according to the standard electromyography signal data of each individual. Optimizing the configuration of the BP neural network to obtain an optimized BP neural network structure; obtaining the scoring results of the fatigue degree of each muscle; determining an electromyography signal data set through the range of the number of individual samples for electromyophysiological tests; training the optimized BP neural network structure according to the integrated electromyogram value, wavelet packet entropy, electromyography signal data set, and scoring results of the fatigue degree of each muscle to obtain a trained BP neural network structure. Performing muscle fatigue prediction through the trained BP neural network structure.

[0005] Further, the step of determining the range of the individual sample size for the electromyophysiological test includes:

[0006] In the formula, represents the degree of freedom of the probability of rejecting a true null hypothesis, represents the degree of freedom of the probability of accepting a false null hypothesis, represents the standard deviation of the integrated electromyogram value of the muscle electrical signal, represents the basic value of the individual sample size for the surface electromyophysiological test, represents the difference in the integrated electromyogram values between the experimental group and the control group, represents rounding up; By using the basic value of the individual sample size for the surface electromyophysiological test, the range of the individual sample size for the electromyophysiological test is determined, that is, the individual sample size for the electromyophysiological test must be greater than or equal to the basic value of the individual sample size for the surface electromyophysiological test.

[0007] Further, the step of filtering the original electromyogram signal data to obtain the finally filtered signal data, denoted as electromyogram signal data, includes: Filter the original electromyogram signal data by first using a band-pass filtering algorithm, then using a notch filtering algorithm, and finally using a wavelet filtering algorithm; denote the finally filtered signal data as electromyogram signal data.

[0008] Further, the step of obtaining the maximum isometric contraction corresponding to each individual in the electromyogram signal data and the root mean square value of each individual; and obtaining the standard electromyogram signal data for each individual according to the maximum isometric contraction corresponding to each individual and the root mean square value of each individual, includes: Obtain the maximum isometric contraction of each individual through a force sensor; At a preset equal interval , obtain the data of several moments in the electromyogram signal of each individual;

[0009] In the formula, represents the data of the th moment in the electromyogram signal of each individual, represents the number of all moments in the electromyogram signal of each individual, represents the root mean square value of each individual.

[0010] Further, the step of obtaining the integrated electromyogram value and the wavelet packet entropy of each muscle according to the standard electromyogram signal data of each individual, includes: According to the standard electromyogram signal data of each individual, the integrated electromyogram value of each muscle is obtained through integral calculation; according to the standard electromyogram signal data of each individual, the wavelet packet entropy of each muscle is obtained through the wavelet packet transform algorithm.

[0011] Further, for the optimized BP neural network configuration, the optimized BP neural network structure is obtained; the scoring results of the fatigue degree of each muscle are obtained, including: For the BP neural network configuration optimized by the genetic algorithm, the optimized BP neural network structure is obtained; Among them, the scoring results of the fatigue degree of each muscle are evaluated manually.

[0012] Further, the prediction of muscle fatigue degree by the trained BP neural network structure includes: Taking the electromyogram signal data as input, through the trained BP neural network structure, the scoring results of muscle fatigue degree are output; there are 10 values for the scoring results of muscle fatigue degree, which are 0 to 10 respectively; among them, 0 to 3 indicates no fatigue, 4 to 6 indicates some fatigue, and above 6 points indicates very fatigue; Taking the scoring results of muscle fatigue degree as the muscle fatigue degree prediction results.

[0013] The second aspect of the present invention is to provide a muscle fatigue degree prediction system based on a BP neural network model, including: Sample range determination module: used to determine the range of the sample size of the electromyophysiological test individuals; Data processing module: used to obtain several groups of original electromyogram signal data of the subjects; filter the original electromyogram signal data to obtain the finally filtered signal data, denoted as electromyogram signal data; obtain the maximum isometric contraction corresponding to each individual and the root mean square value of each individual in the electromyogram signal data; according to the maximum isometric contraction corresponding to each individual and the root mean square value of each individual, obtain the standard electromyogram signal data of each individual; according to the standard electromyogram signal data of each individual, obtain the integrated electromyogram value and wavelet packet entropy of each muscle; Neural network optimization and training module: used for the optimized BP neural network configuration to obtain the optimized BP neural network structure; obtain the scoring results of the fatigue degree of each muscle; determine the electromyogram signal data set through the range of the sample size of the electromyophysiological test individuals; train the optimized BP neural network structure according to the integrated electromyogram value, wavelet packet entropy, electromyogram signal data set and scoring results of the fatigue degree of each muscle to obtain the trained BP neural network structure; Prediction module: used to predict the muscle fatigue degree through the trained BP neural network structure.

[0014] The third aspect of the present invention is to provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the method for predicting muscle fatigue degree based on a BP neural network model is implemented.

[0015] The fourth aspect of the present invention is to provide a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for predicting muscle fatigue degree based on a BP neural network model is implemented.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: the original electromyogram signal data is filtered to obtain electromyogram signal data, reducing the influence of noise on the electromyogram signal data; the maximum isometric contraction corresponding to each individual and the root mean square value of each individual in the electromyogram signal data are obtained, reducing the influence between individual differences; according to the maximum isometric contraction corresponding to each individual and the root mean square value of each individual, the standard electromyogram signal data of each individual is obtained; according to the standard electromyogram signal data of each individual, the integrated electromyogram value and wavelet packet entropy of each muscle are obtained, improving the calculation speed; the optimized BP neural network configuration is obtained, and the optimized BP neural network structure is obtained, reducing the problems that the BP neural network is sensitive to the setting of initial weights and thresholds and is prone to falling into local optimum or having a slow convergence speed; the scoring result of the fatigue degree of each muscle is obtained; the training set is determined through the range of the individual sample size of the electromyophysiological test; according to the integrated electromyogram value, wavelet packet entropy, electromyogram signal data set of each muscle, and the scoring result of the fatigue degree of each muscle, the optimized BP neural network structure is trained to obtain the trained BP neural network structure; the muscle fatigue degree is predicted through the trained BP neural network structure, improving the accuracy of muscle fatigue degree prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0018] Figure 1 It is a schematic flow chart of the steps of the method for predicting muscle fatigue degree based on a BP neural network model provided by the present invention; Figure 2 It is a schematic module flow chart of the system for predicting muscle fatigue degree based on a BP neural network model provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] To enable those skilled in the art to better understand the solution of the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the scope of protection of the present invention.

[0020] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily need to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0021] In view of the problems existing in the background technology, a method and system for predicting muscle fatigue degree based on the BP neural network model are designed through research, which has important practical significance.

[0022] As Figure 1 shown, the first aspect of the present invention is to provide a method for predicting muscle fatigue degree based on the BP neural network model, including the following steps: Step S001: Determine the range of the individual sample size of the electromyophysiological test.

[0023] It should be noted that the surface electromyogram signal is a one-dimensional time series signal obtained by amplifying, guiding, displaying and recording the bioelectric changes generated during neuromuscular activity through the skin surface. It can reflect the true muscle fatigue degree of the human body and has the characteristics of non-invasive, real-time monitoring, multi-target fixed-point monitoring, and being unaffected by posture and environment. For its typical upper limb muscles, a multi-channel physiological polygraph is used to test the responses of different muscles under the same maintenance actions, collect and process the time-domain and frequency-domain signal values, and analyze the relationships between the data to study the fatigue characteristics of the upper limbs during personnel maintenance operations.

[0024] Specifically, the basic value of the individual sample size of the surface electromyophysiological test is estimated according to the random design individual sample size estimation formula based on the t-distribution in statistics; among them, the random design individual sample size estimation formula is specifically expressed as:

[0025] In the formula, represents the degree of freedom of the probability of false rejection, represents the degree of freedom of the probability of false acceptance, represents the standard deviation of the integrated electromyogram value of the electromyographic signal, represents the basic value of the sample size of the surface electromyophysiological test individuals, represents the difference in the integrated electromyogram values between the experimental group and the control group, represents rounding up. Among them, the standard deviation of the integrated electromyogram value of the electromyographic signal and the difference in the integrated electromyogram values between the experimental group and the control group are both determined by convention based on literature or the experience of experts in the field; among them, = 0.05, .

[0026] Among them, in this embodiment, the probability of false rejection is 0.05 and the probability of false acceptance is 0.1. In this embodiment, the probability of false rejection and the probability of false acceptance are not specifically limited, and the implementer can determine according to the specific situation.

[0027] Among them, is the corresponding degree of freedom 1.645 when the probability of false rejection is 0.05, is the corresponding degree of freedom 1.282 when the probability of false acceptance is 0.1.

[0028] According to the random design individual sample size estimation formula, the basic value of the electromyophysiological test individual sample size obtained is:

[0029] Up to this point, the basic value of the electromyophysiological test individual sample size is obtained through the above method; therefore, when performing subsequent analysis, it is necessary to ensure that the range of the electromyophysiological test individual sample size is greater than or equal to .

[0030] Step S002: Filter the original electromyographic signal data to obtain the electromyographic signal data denoted; obtain the maximum isometric contraction corresponding to each individual in the electromyographic signal data and the root mean square value of each individual; obtain the standard electromyographic signal data of each individual according to the maximum isometric contraction corresponding to each individual and the root mean square value of each individual; obtain the integrated electromyogram value and wavelet packet entropy of each muscle according to the standard electromyographic signal data of each individual.

[0031] It should be noted that, for the sake of rigor in the experimental design, eight channels were collected during the collection of EMG data, covering all major muscles of the upper limb. However, after collecting these data and conducting preliminary analysis, the researchers found that compared to the resting state and the data of the other limb, only channels 2, 3, and 4, namely the data of the right biceps, deltoid, and trapezius muscles, showed significant fluctuations and a high muscle contribution rate. This is consistent with the observation during the experiment that antenna array installers usually use their right hand (dominant hand) to perform almost all overhead operations. Therefore, in the subsequent model implementation process, in order to avoid the impact of noise in other channel data on the accuracy of the model and the impact of data volume on the model running speed, the EMG data of other channels were eliminated, and only the EMG data of the right biceps, deltoid, and trapezius muscles were used as EMG signals for subsequent analysis.

[0032] It should be further explained that, due to the interference of noise when collecting electromyographic signal data, the collected electromyographic signal has a certain deviation, so the collected electromyographic signal is first subjected to denoising and filtering processing by a filtering method.

[0033] Specifically, several groups of original electromyographic signal data of the subjects are obtained; the original electromyographic signal data are first filtered using a bandpass filtering algorithm, then filtered using a notch filtering algorithm, and finally filtered using a wavelet filtering algorithm; the final filtered signal data is recorded as electromyographic signal data.

[0034] The bandpass filtering algorithm, notch filtering algorithm, and wavelet filtering algorithm are all well-known technologies and will not be described in detail here. In this embodiment, the frequency range of the bandpass filtering algorithm is 50 to 480 Hz, the specific frequency of the notch filtering algorithm is 50 Hz, and the frequency range of the wavelet filtering algorithm is 50 to 450 Hz. In this embodiment, neither the frequency range nor the specific frequency is specifically limited, and the implementer may determine the specific frequency based on the specific situation.

[0035] So far, the electromyographic signal data has been obtained through the above method.

[0036] It should be noted that due to the large differences between different individuals, the signal values corresponding to the electromyographic signals of different individuals under the same muscle fatigue are different. Therefore, there will be large errors when training the neural network model directly based on the individual electromyographic signals. Therefore, in order to eliminate the differences between different individuals, the maximum isometric contraction (MVC, Maximum Voluntary Contraction) corresponding to each individual is introduced to eliminate the difference problem.

[0037] Specifically, the maximum isometric contraction of each individual is obtained by a force sensor (force plate or tensiometer). , obtain the data at several moments in the electromyogram signal of each individual; according to the data at several moments in the electromyogram signal of each individual, obtain the root mean square value (RMS, Root Mean Square) of each individual; the root mean square value of each individual is specifically expressed by the formula:

[0038] In the formula, represents the data at the -th moment in the electromyogram signal of each individual, represents the number of all moments in the electromyogram signal of each individual, represents the root mean square value of each individual. Among them, in this embodiment, a preset equal interval of seconds is set. Among them, in this embodiment, the preset equal interval is not specifically limited, and the implementer can determine it according to the specific situation.

[0039] According to the root mean square value of each individual and the maximum isometric contraction of each individual, obtain the standard electromyogram signal data of each individual; the standard electromyogram signal data of each individual is specifically expressed by the formula:

[0040] In the formula, represents the root mean square value of each individual, represents the maximum isometric contraction of each individual, represents the standard electromyogram signal data of each individual.

[0041] It should be noted that common characteristic indexes include relatively traditional ones such as integrated electromyogram value (IEMG, Integrated Electromyography), average power frequency, median frequency, etc., and relatively new ones such as wavelet packet energy, energy ratio, energy entropy, etc.; in the previous data processing, researchers found that as the experiment progresses, while the fatigue of the subjects increases, among the traditional indexes, IEMG (time-domain electromyogram signal) has a negative correlation change trend, and among the new indexes, wavelet packet entropy (in the frequency domain) also has a strong change correlation; because IEMG has a strong correlation in the time domain and wavelet packet entropy has a strong correlation in the frequency domain, the model simultaneously selects IEMG and wavelet packet entropy as the secondary indexes of the electromyogram signal, so as to better characterize the fatigue change trend of the subjects during the whole experiment and provide more valuable features for the neural network to learn.

[0042] Specifically, according to the standard electromyogram signal data of each individual, the integrated electromyogram value of each muscle is obtained through integral calculation; according to the standard electromyogram signal data of each individual, the wavelet packet entropy of each muscle is obtained through the wavelet packet transform algorithm. Among them, the calculation process of the integrated electromyogram value is a well-known technology, and the process of calculating the wavelet packet entropy by the wavelet packet transform algorithm is also a well-known technology, so no specific elaboration will be made here.

[0043] So far, the integrated electromyogram value and wavelet packet entropy of each muscle are obtained through the above method.

[0044] Step S003: Optimize the BP neural network configuration to obtain the optimized BP neural network structure; according to the integrated electromyogram value, wavelet packet entropy, electromyogram signal data set of each muscle and the scoring result of the fatigue degree of each muscle, train the optimized BP neural network structure to obtain the trained BP neural network structure.

[0045] It should be noted that using the neural network for large sample training and prediction of the number of experimental times and time lapse changes can obtain prediction results with ten times the sample quantity. In this embodiment, the data is trained based on the BP neural network model, and the purpose is to obtain more reliable data results.

[0046] Furthermore, it should be noted that the BP neural network includes an input layer, a hidden layer, and an output layer; the BP neural network is very sensitive to the initial connection weights between the input layer and the hidden layer neurons and the initial thresholds between the hidden layer and the output layer. Therefore, unreasonable parameter settings will lead to slow network convergence speed and getting stuck in local optima.

[0047] Specifically, through the BP neural network configuration optimized by GA (Genetic Algorithm, genetic algorithm), the optimized BP neural network structure is obtained; among them, the optimal values of all parameters in the BP neural network configuration are obtained during the optimization process. Among them, the process of optimizing the BP neural network configuration by the genetic algorithm is a well-known technology, so no specific elaboration will be made here.

[0048] The electromyogram signal data set is determined through the range of the individual sample size of the electromyophysiological test; according to the integrated electromyogram value, wavelet packet entropy, electromyogram signal data set of each muscle and the scoring result of the fatigue degree of each muscle, train the optimized BP neural network structure to obtain the trained BP neural network structure. Among them, the scoring result of the fatigue degree of the muscle is evaluated manually.

[0049] Among them, during the training process, the electromyogram signal data set is divided into a training set and a test set according to a ratio of 8:2. Among them, after multiple trainings and improving the network structure according to the training results, the model hidden layer uses The structure has three hidden layers. The ReLU (Rectified Linear Unit) is used as the activation function in the first two layers. The learning rate is 0.005. When training for 100 rounds, it can basically meet the usage requirements while ensuring a certain generalization ability. After storing the trained model parameters as a weight file, the training process of the model is completely over.

[0050] Among them, during the training process of the BP neural network, a computer equipped with an x86 processor (with a main frequency of not less than 1 GHz and a memory of not less than 1 GB) and equipped with the Python 3.11.10 runtime library is used.

[0051] The third-party libraries called during the model implementation process include: Pytorch (neural network construction framework), Pandas (external file reading, processing, and writing), Openpyxl (Excel file reading support), TQDM (program execution progress display), Numpy (matrix operation support framework), Matplotlib (visualization drawing support framework), Scipy (scientific computing and filter construction library), Pywavelets (wavelet transform support library), OS (partial system interface library), and SYS (Python interpreter interaction support library).

[0052] Step S004: Predict the muscle fatigue degree through the trained BP neural network structure.

[0053] Take the electromyogram signal data as the input, and through the trained BP neural network structure, output the scoring result of the muscle fatigue degree; there are 10 values for the scoring result of the muscle fatigue degree, which are from 0 to 10; among them, 0 to 3 indicates not fatigued, 3 to 6 indicates somewhat fatigued, and above 6 points indicates very fatigued. Take the scoring result of the muscle fatigue degree as the muscle fatigue degree prediction result.

[0054] So far, the prediction of the muscle fatigue degree is completed.

[0055] As Figure 2 shown, the second aspect of the present invention is to provide a muscle fatigue degree prediction system based on a BP neural network model, including the following modules: Sample range determination module 101: used to determine the range of the sample size of the electromyophysiological test individual; Data processing module 102: It is used to obtain several groups of original electromyogram signal data of the subject; filter the original electromyogram signal data to obtain the finally filtered signal data, denoted as electromyogram signal data; obtain the maximum isometric contraction corresponding to each individual in the electromyogram signal data and the root mean square value of each individual; obtain the standard electromyogram signal data of each individual according to the maximum isometric contraction corresponding to each individual and the root mean square value of each individual; obtain the integrated electromyogram value and wavelet packet entropy of each muscle according to the standard electromyogram signal data of each individual. Neural network optimization and training module 103: It is used to optimize the configuration of the BP neural network to obtain the optimized BP neural network structure; obtain the scoring results of the fatigue degree of each muscle; determine the electromyogram signal data set through the range of the individual sample size of the electromyogram physiological test; train the optimized BP neural network structure according to the integrated electromyogram value, wavelet packet entropy, electromyogram signal data set and the scoring results of the fatigue degree of each muscle to obtain the trained BP neural network structure. Prediction module 104: It is used to predict the muscle fatigue degree through the trained BP neural network structure.

[0056] The third aspect of the present invention is to provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements a method for predicting muscle fatigue degree based on a BP neural network model.

[0057] The fourth aspect of the present invention is to provide a computer-readable storage medium. The computer-readable storage medium stores a computer program. When the computer program is executed by a processor, it implements a method for predicting muscle fatigue degree based on a BP neural network model.

[0058] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, optical storage, etc.) containing computer-usable program code.

[0059] The present invention is described with reference to the flowcharts and / or block diagrams of methods, systems, and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in the Figure 1 one or more flows and / or blocks Figure 1 one or more blocks.

[0060] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means that implement the functions specified in the Figure 1 one or more flows and / or blocks Figure 1 one or more blocks.

[0061] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, so that the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in the Figure 1 one or more flows and / or blocks Figure 1 one or more blocks.

[0062] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications or equivalent replacements can still be made to the specific implementation manners of the present invention, and any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the protection scope of the present invention.

Claims

1. A method for predicting muscle fatigue degree based on BP neural network model, characterized in that, Including: Determining the range of the individual sample size for electromyophysiological testing; Obtaining a number of groups of original electromyography signal data of the subjects; Filtering the original electromyography signal data to obtain the finally filtered signal data, denoted as electromyography signal data; Obtaining the maximum isometric contraction corresponding to each individual and the root mean square value of each individual in the electromyography signal data; Obtaining the standard electromyography signal data of each individual according to the maximum isometric contraction corresponding to each individual and the root mean square value of each individual; obtaining the integrated electromyography value and wavelet packet entropy of each muscle according to the standard electromyography signal data of each individual; Optimizing the BP neural network configuration to obtain the optimized BP neural network structure; obtaining the scoring results of the fatigue degree of each muscle; Determining an electromyography signal data set through the range of the individual sample size for electromyophysiological testing; Training the optimized BP neural network structure according to the integrated electromyography value, wavelet packet entropy, electromyography signal data set of each muscle and the scoring results of the fatigue degree of each muscle to obtain the trained BP neural network structure; Predicting the muscle fatigue degree through the trained BP neural network structure.

2. The muscle fatigue prediction method based on the BP neural network model according to claim 1, characterized in that The determining the range of the individual sample size for electromyophysiological testing includes: In the formula, represents the degree of freedom of the probability of false rejection, represents the degree of freedom of the probability of false acceptance, represents the standard deviation of the integrated electromyogram value of the electromyogram signal, represents the basic value of the individual sample size of the surface electromyogram physiological test, represents the difference in the integrated electromyogram values between the experimental group and the control group, represents rounding up for Determining the range of the individual sample size for electromyophysiological testing through the basic value of the individual sample size for surface electromyophysiological testing, that is, the individual sample size for electromyophysiological testing must be greater than or equal to the basic value of the individual sample size for surface electromyophysiological testing.

3. A method for predicting muscle fatigue degree based on a BP neural network model according to claim 1, characterized in that, The filtering the original electromyography signal data to obtain the finally filtered signal data, denoted as electromyography signal data, includes: Filtering the original electromyography signal data by first using a band-pass filtering algorithm, then using a notch filtering algorithm, and finally using a wavelet filtering algorithm; denoting the finally filtered signal data as electromyography signal data.

4. A method for predicting muscle fatigue degree based on a BP neural network model according to claim 1, characterized in that, The obtaining the maximum isometric contraction corresponding to each individual and the root mean square value of each individual in the electromyography signal data; Obtaining the standard electromyography signal data of each individual according to the maximum isometric contraction corresponding to each individual and the root mean square value of each individual includes: Obtaining the maximum isometric contraction of each individual through a force sensor; At a preset equal interval , obtain data at several moments in the electromyogram signal of each individual; In the formula, represents the data at the -th moment in the EMG signal of each individual, represents the number of data at all moments in the EMG signal of each individual, represents the root mean square value of each individual.

5. A method for predicting muscle fatigue degree based on a BP neural network model according to claim 1, characterized in that, The obtaining the integrated electromyography value and wavelet packet entropy of each muscle according to the standard electromyography signal data of each individual includes: Obtaining the integrated electromyography value of each muscle through integral calculation according to the standard electromyography signal data of each individual; obtaining the wavelet packet entropy of each muscle through a wavelet packet transform algorithm according to the standard electromyography signal data of each individual.

6. The muscle fatigue degree prediction method based on the BP neural network model according to claim 1, characterized in that The optimizing the BP neural network configuration to obtain the optimized BP neural network structure; obtaining the scoring results of the fatigue degree of each muscle includes: Optimizing the BP neural network configuration through a genetic algorithm to obtain the optimized BP neural network structure; Among them, the scoring results of the fatigue degree of each muscle are evaluated manually.

7. A method for predicting muscle fatigue degree based on a BP neural network model according to claim 1, characterized in that, The predicting the muscle fatigue degree through the trained BP neural network structure includes: Taking the electromyography signal data as the input, through the trained BP neural network structure, output the scoring result of muscle fatigue degree; there are 10 values for the scoring result of muscle fatigue degree, which are from 0 to 10 respectively; among them, 0 to 3 indicates no fatigue, 4 to 6 indicates some fatigue, and above 6 points indicates very fatigue; Taking the scoring result of muscle fatigue degree as the muscle fatigue degree prediction result.

8. A muscle fatigue prediction system based on a BP neural network model, characterized in that, Including: Sample range determination module: used to determine the range of the sample size of the electromyophysiological test individuals; Data processing module: used to obtain several groups of original electromyography signal data of the subjects; Perform filtering processing on the original electromyography signal data to obtain the finally filtered signal data, denoted as electromyography signal data; Obtain the maximum isometric contraction corresponding to each individual in the electromyography signal data and the root mean square value of each individual; Obtain the standard electromyography signal data of each individual according to the maximum isometric contraction corresponding to each individual and the root mean square value of each individual; obtain the integrated electromyography value and wavelet packet entropy of each muscle according to the standard electromyography signal data of each individual; Neural network optimization and training module: used to optimize the configuration of the BP neural network to obtain the optimized BP neural network structure; obtain the scoring result of the fatigue degree of each muscle; Determine the electromyography signal data set through the range of the sample size of the electromyophysiological test individuals; Train the optimized BP neural network structure according to the integrated electromyography value, wavelet packet entropy, electromyography signal data set of each muscle and the scoring result of the fatigue degree of each muscle to obtain the trained BP neural network structure; Prediction module: used to predict the muscle fatigue degree through the trained BP neural network structure.

9. An electronic device, characterized in that, Including a memory, a processor, and a computer program stored in the memory and executable on the processor, when the processor executes the computer program, it implements the method for predicting muscle fatigue degree based on the BP neural network model according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, when the computer program is executed by the processor, it implements the method for predicting muscle fatigue degree based on the BP neural network model according to any one of claims 1-7.

Images