Overload detection method and system for wire feeding motor

Abstract

The invention discloses an overload detection method and system for a wire feeding motor, and relates to the technical field of overload detection of motors. Comprising the following steps: establishing an electromagnetic-thermal-mechanical stress multi-field coupling model of the wire feeding motor, inputting first input data, calculating a magnetic field distortion degree and a temperature rise gradient, and forming a composite criterion; based on the composite criterion, analyzing current fluctuation characteristics in the wire feeding process, and extracting a wire feeding resistance fluctuation spectrum by utilizing spectrum analysis; according to the wire feeding resistance fluctuation frequency spectrum, load dynamic decoupling is conducted, and the equivalent friction coefficient is reversely deduced in real time; the overload state of the wire feeding motor is judged according to the composite criterion and the dynamic change of the equivalent friction coefficient; and adjusting parameters in the electromagnetic-thermal-mechanical stress multi-field coupling model based on the judgment result of the overload state. According to the method, the friction coefficient is dynamically reversely deduced by using the wire feeding resistance fluctuation spectrum, and the overload state is effectively judged through the composite criterion, so that the problem that a single parameter is easy to misjudge in a traditional overload detection method is avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of overload detection of motors, and particularly to an overload detection method and system for a wire feeding motor. Background Art

[0002] Wire feeding motors are widely used in fields such as welding equipment, automated production lines, and additive manufacturing. Their stability and reliability directly affect welding quality and production efficiency. With the continuous improvement of the requirements for high-precision and high-efficiency welding processes in the manufacturing industry, the working performance and intelligent control level of wire feeding motors are also continuously improving. The existing overload detection methods for wire feeding motors mainly rely on the monitoring of single parameters such as current, voltage, or speed. For example, the overload state is judged through current thresholds or load abnormalities are judged based on voltage fluctuations.

[0003] However, these traditional methods are difficult to accurately reflect the complex working conditions during the wire feeding process. Especially under the coupled action of multiple factors such as high-frequency current fluctuations, non-linear friction forces, and temperature changes, false judgments or missed judgments are likely to occur. In addition, some studies have tried to obtain load information through mechanical stress sensors, but due to complex installation, high cost, and significant changes to the equipment structure, their practical applications are limited. Summary of the Invention

[0004] In view of the problems existing in the existing technologies in the overload detection of wire feeding motors, such as low accuracy, poor real-time performance, and high false judgment rate, the present invention is proposed.

[0005] Therefore, the problem to be solved by the present invention is how to effectively improve the accuracy of overload detection.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: In the first aspect, the present invention provides an overload detection method for a wire feeding motor, which includes establishing an electromagnetic-thermal-mechanical stress multi-field coupling model of the wire feeding motor, inputting first input data, calculating the magnetic field distortion degree and the temperature rise gradient, and forming a composite criterion; the first input data includes the real-time current, temperature, and stress data of the wire feeding motor; based on the composite criterion, analyzing the current fluctuation characteristics during the wire feeding process, and using spectrum analysis to extract the wire feeding resistance fluctuation spectrum; according to the wire feeding resistance fluctuation spectrum, performing load dynamic decoupling and real-time back-calculating the equivalent friction coefficient; judging the overload state of the wire feeding motor through the dynamic changes of the composite criterion and the equivalent friction coefficient; based on the judgment result of the overload state, adjusting the parameters in the electromagnetic-thermal-mechanical stress multi-field coupling model and feeding back to the link of extracting the wire feeding resistance fluctuation spectrum.

[0007] As a preferred solution of the overload detection method for the wire feeding motor of the present invention, wherein: the establishment of the electromagnetic-thermal-mechanical stress multi-field coupling model includes: based on current data, using Maxwell's equations to establish an electromagnetic field model inside the wire feeding motor and initializing the electromagnetic field parameters; according to the electromagnetic field model, calculating the heat source generated by electromagnetic loss and establishing a temperature field model through Fourier's heat conduction equation, initializing the temperature field parameters; based on the temperature field gradient and electromagnetic force, using Hooke's law and thermal stress coupling analysis to establish a mechanical stress field model and initializing the mechanical stress field parameters; through the sequential establishment of the electromagnetic-thermal-mechanical stress fields and establishing data sharing and parameter transfer relationships between each field, enabling coupling relationships between the fields and forming a complete multi-field coupling model.

[0008] As a preferred solution of the overload detection method for the wire feeding motor of the present invention, wherein: the composite criterion is formed by weighting the magnetic field distortion degree and the temperature rise gradient.

[0009] As a preferred solution of the overload detection method for the wire feeding motor of the present invention, wherein: analyzing the current fluctuation characteristics during the wire feeding process and using spectrum analysis to extract the wire feeding resistance fluctuation spectrum includes: according to the preprocessed current fluctuation data, performing spectrum analysis, calculating the spectrum density of the current fluctuation, and obtaining the spectrum characteristics of the current fluctuation; according to the spectrum characteristics of the current fluctuation, combining the information of the magnetic field distortion degree and the temperature rise gradient in the composite criterion, performing spectrum feature selection, and constructing the wire feeding resistance fluctuation spectrum; the spectrum feature selection includes: using a feature selection algorithm based on mutual information to calculate the mutual information amount between the magnetic field distortion degree, the temperature rise gradient and each spectrum feature, measuring the dependence relationship, and performing spectrum feature selection; optimizing and dimension reduction of the selected spectrum features.

[0010] As a preferred solution of the overload detection method for the wire feeding motor of the present invention, wherein: performing load dynamic decoupling according to the wire feeding resistance fluctuation spectrum includes the following steps: based on the wire feeding resistance fluctuation spectrum, using the wavelet transform method to decompose the spectrum, dividing the spectrum signal into sub-signals of different scales and frequency bands; calculating the spectrum energy of each sub-signal and combining the wire feeding current fluctuation characteristics to screen the sensitive frequency band; based on the sensitive frequency band, using a dynamic decoupling method based on the generalized least squares method to separate the frequency band related to the friction coefficient and other interference frequency bands, and obtaining the net friction influence signal.

[0011] As a preferred solution of the overload detection method for the wire feeding motor of the present invention, wherein: the generation of the equivalent friction coefficient includes: according to the decoupled net friction influence signal, establishing a non-linear inverse model of the equivalent friction coefficient and using the recursive least squares method for parameter identification to dynamically calculate the real-time equivalent friction coefficient.

[0012] As a preferred embodiment of the overload detection method for the wire feeding motor of the present invention, wherein: determining the overload state of the wire feeding motor includes: constructing a composite criterion model based on the equivalent friction coefficient, in combination with the magnetic field distortion degree and the temperature rise gradient, and calculating the composite criterion value. Setting upper and lower limit determination thresholds according to the operating characteristics of the wire feeding motor and the typical parameter change trends under overload conditions. And .

[0013] As a preferred embodiment of the overload detection method for the wire feeding motor of the present invention, wherein: determining the overload state of the wire feeding motor further includes: based on the calculated composite criterion value And the set determination threshold, performing an overload state determination: when , it is determined that the wire feeding motor is in an overload state; when , it is determined that the wire feeding motor is in a normal state; when , it is determined that the wire feeding motor is in a critical state, and further analysis is carried out in combination with the dynamic change trend of the equivalent friction coefficient: if the equivalent friction coefficient shows a continuous upward trend and the difference between the composite criterion and the upper limit threshold is less than the difference threshold , then an early warning of potential overload risk is given; otherwise, the current state judgment is maintained.

[0014] In a second aspect, the present invention provides an overload detection system for a wire feeding motor, which includes: a composite criterion calculation module, configured to establish an electromagnetic-thermal-mechanical stress multi-field coupling model of the wire feeding motor, input first input data, calculate the magnetic field distortion degree and the temperature rise gradient, and form a composite criterion; the first input data includes the real-time current, temperature, and stress data of the wire feeding motor; a spectrum analysis module, configured to analyze the current fluctuation characteristics during the wire feeding process based on the composite criterion, and extract the wire feeding resistance fluctuation spectrum using spectrum analysis; a decoupling and friction coefficient inverse deduction module, configured to perform load dynamic decoupling according to the wire feeding resistance fluctuation spectrum and inversely deduce the equivalent friction coefficient in real time; an overload state determination module, configured to determine the overload state of the wire feeding motor through the composite criterion and the dynamic change of the equivalent friction coefficient; an adjustment and feedback module, configured to adjust the parameters in the electromagnetic-thermal-mechanical stress multi-field coupling model based on the determination result of the overload state and feedback to the link of extracting the wire feeding resistance fluctuation spectrum.

[0015] In a third aspect, the present invention provides a computer device, including a memory and a processor, where the memory stores a computer program, wherein: when the computer program instructions are executed by the processor, the steps of the overload detection method for the wire feeding motor as described in the first aspect of the present invention are implemented.

[0016] Fourthly, the present invention provides a computer-readable storage medium, on which a computer program is stored, wherein: when the computer program instructions are executed by a processor, the steps of the overload detection method of the wire feeding motor as described in the first aspect of the present invention are implemented.

[0017] The beneficial effects of the present invention are as follows: By combining the multi-field coupling model with the online identification of the friction coefficient, the present invention realizes the accurate detection and real-time response to the overload state of the wire feeding motor; dynamically deduces the friction coefficient by using the wire feeding resistance fluctuation spectrum, and effectively determines the overload state through a composite criterion, avoiding the problem of easy misjudgment of a single parameter in the traditional overload detection method. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0019] Figure 1 It is a schematic flowchart of the overload detection method for the wire feeding motor.

[0020] Figure 2 It is a schematic diagram for determining the overload state.

[0021] Figure 3 It is a structural diagram of the overload detection system for the wire feeding motor. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0022] In order to make the above objects, features and advantages of the present invention more obvious and understandable, the following will make a detailed description of the specific embodiments of the present invention with reference to the drawings in the specification.

[0023] In the following description, many specific details are set forth in order to fully understand the present invention. However, the present invention can also be implemented in other ways different from those described herein. Those skilled in the art can make similar promotions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0024] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure or characteristic that can be included in at least one implementation manner of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment that excludes other embodiments.

[0025] Embodiment 1 Refer to Figures 1 to 3, which is the first embodiment of the present invention. This embodiment provides an overload detection method for a wire feeding motor, as shown in Figure 1 and includes: S1: Establish an electromagnetic-thermal-mechanical stress multi-field coupling model for the wire feeding motor, input the first input data, calculate the magnetic field distortion degree and the temperature rise gradient, and form a composite criterion.

[0026] Among them, the first input data includes the real-time current, temperature, and stress data of the wire feeding motor.

[0027] After collecting the first input data, perform filtering and normalization processing on the first input data to remove noise and unify the data scale, providing a high-quality data basis for subsequent calculations.

[0028] In an embodiment of the present application, establishing an electromagnetic-thermal-mechanical stress multi-field coupling model for the wire feeding motor specifically includes the following operations: Based on historical input data, establish an electromagnetic-thermal-mechanical stress multi-field coupling model and initialize the coupling parameters, including electromagnetic field parameters, temperature field parameters, and mechanical stress field parameters, so that there is a coupling relationship between each field; According to the coupling parameters of the electromagnetic-thermal-mechanical stress multi-field coupling model, use an adaptive algorithm to calculate the coupling coefficient, dynamically adjust the mutual influence between the electromagnetic field, temperature field, and mechanical stress field, and ensure the accuracy and stability of the multi-field coupling model.

[0029] Specifically, based on the current data, use Maxwell's equations to establish an electromagnetic field model inside the wire feeding motor and initialize the electromagnetic field parameters, including magnetic permeability, eddy current loss coefficient, and current density distribution; According to the electromagnetic field model, calculate the heat source generated by electromagnetic loss, and establish a temperature field model through Fourier's heat conduction equation, and initialize the temperature field parameters, including heat conduction coefficient, heat capacity, and heat dissipation boundary conditions; Based on the temperature field gradient and electromagnetic field force, use the generalized Hooke's law and thermal stress coupling analysis to establish a mechanical stress field model and initialize the mechanical stress field parameters, including elastic modulus, Poisson's ratio, and thermal expansion coefficient.

[0030] By sequentially establishing the electromagnetic-thermal-mechanical stress fields and establishing data sharing and parameter transfer relationships between each field, there is a coupling relationship between each field, forming a complete multi-field coupling model.

[0031] Furthermore, the calculation process of the coupling coefficient includes the following operation steps: Perform a sensitivity analysis on the electromagnetic field parameters, temperature field parameters, and mechanical stress field parameters, determine the influence weights of each parameter on the coupling effect, and adaptively adjust the initial parameter values through the gradient descent algorithm. The calculation formula is as follows: Among them, is the coupling parameter for the sensitivity to the state change amount ; is the change gradient of the state variable in the electromagnetic-thermal-mechanical stress multi-field coupling model, is the coupling parameter for the tiny change amount.

[0032] In the above process, calculating the sensitivity of each coupling parameter to the multi-field state change amount to obtain the parameter sensitivity can be used for the adaptive adjustment of the coupling coefficient in the subsequent steps; it can be seen that the above calculation process improves the traditional local sensitivity analysis method, introduces the parameter change gradient and the state change gradient, can reflect the dynamic influence of each parameter on different state variables, and effectively improves the dynamic response ability of the coupling model.

[0033] Adopting an adaptive algorithm, based on the sensitivity analysis results, dynamically calculate the coupling coefficients between the electromagnetic field, the temperature field and the mechanical stress field, and update the coupling coefficients according to the real-time change of the working state of the wire feeding motor. The calculation formula is as follows: Among them, is the coupling coefficient between the electromagnetic field, the temperature field and the mechanical stress field, is the coupling coefficient in the previous iteration, is the adaptive learning rate, used to control the update amplitude of the coupling coefficient, is the state change amount obtained by experimental measurement, is the state change amount calculated by the model.

[0034] In the above process, the formula is based on the parameter sensitivity analysis results, adopts an adaptive incremental adjustment strategy to dynamically update the coupling coefficient, and realizes the dynamic optimization and adaptive adjustment of the coupling coefficient by fusing the adaptive learning rate and the sensitivity and introducing a state difference feedback mechanism, overcoming the error accumulation problem caused by the fixed or linearized processing of the coupling coefficient in the traditional model.

[0035] Considering the non-linear coupling effect between each field, a generalized regression neural network (GRNN) is used to establish a non-linear coupling relationship, and the network weights are updated online through an adaptive algorithm, so that the coupling model has the ability of non-linear adaptive adjustment.

[0036] Preferably, using the generalized regression neural network (GRNN) to establish a non-linear coupling relationship includes: Among them, is the total state quantity of the state variables in the magneto-thermal-mechanical stress multi-field coupling model . is the non-linear activation function of the coupling parameter , adopting the Sigmoid function form, is the slope parameter of the Sigmoid function, controlling the non-linearity degree of the activation function, is the model error term, and N is the sum of the parameters of all coupling fields.

[0037] In the above process, through three steps of sensitivity analysis, dynamic coupling coefficient calculation and non-linear coupling relationship modeling, the adaptive adjustment and non-linear regulation of the coupling parameters in the multi-field coupling model are realized, and the error accumulation problem caused by the fixed or linearized treatment of the coupling coefficient in the traditional model is overcome.

[0038] Preferably, based on the electromagnetic-thermal-mechanical stress multi-field coupling model and the adaptive coupling coefficient, calculate the magnetic field distortion degree and the temperature rise gradient, and construct a composite criterion through logical operations, including the following operation steps: According to the obtained current, temperature and stress, apply the electromagnetic-thermal-mechanical stress multi-field coupling model to calculate the magnetic field strength, temperature gradient and mechanical stress distribution; Calculate the magnetic field distortion degree and the temperature rise gradient : Among them, the magnetic field distortion degree The calculation formula is as follows: Among them, is the measured value, is the ideal value; Among them, is the temperature change amount, is the corresponding spatial distance; the temperature rise gradient is used to evaluate the rate of change of the thermal field, so as to predict possible overload and thermal damage risks.

[0039] Preferably, based on the adaptive coupling coefficient, dynamically adjust the weights of the magnetic field distortion degree and the temperature rise gradient to ensure the reasonable contribution of the two to overload detection.

[0040] Finally, construct a composite criterion through logical operations, and generate an overload determination result according to the relationship between the magnetic field distortion degree and the temperature rise gradient: This composite criterion not only considers the influence of each physical field, but also can adapt to different working conditions and load conditions.

[0041] S2: Based on the composite criterion, analyze the current fluctuation characteristics during the wire feeding process, and use spectrum analysis to extract the spectrum of the wire feeding resistance fluctuation.

[0042] Preferably, the spectrum of the wire feeding resistance fluctuation is used as the input parameter for the on-line identification of the friction coefficient.

[0043] Optionally, by collecting the current fluctuation data of the wire feeding motor in real time, use the adaptive noise filtering algorithm to preprocess the collected current fluctuation data to filter out environmental noise and high-frequency interference, ensuring the accuracy of subsequent spectrum analysis.

[0044] Among them, the adaptive noise filtering algorithm adaptively adjusts the filtering parameters according to the time-frequency characteristics of the current fluctuation data to improve the noise suppression effect and the retention degree of the fluctuation characteristics.

[0045] In this embodiment, in order to more accurately extract the spectrum characteristics of the current fluctuation during the wire feeding process, this step is based on the preprocessed current fluctuation data, performs spectrum analysis, calculates the spectral density of the current fluctuation, and obtains the main frequency components and their amplitudes, specifically including the following operations: First, use the short-time Fourier transform to perform spectrum analysis on the preprocessed current fluctuation data to calculate the spectral density of the current fluctuation. In order to achieve a balance between time-frequency resolution, an adaptive window function is introduced in this embodiment, that is, according to the frequency distribution characteristics of the current fluctuation, adaptively adjust the window length and overlap rate of the STFT.

[0046] When the frequency components are concentrated in the lower frequency band, a longer window length is used to obtain higher frequency resolution; when the frequency components are distributed in the higher frequency band, a shorter window length is used to improve the time resolution, so as to better capture the transient fluctuation characteristics.

[0047] In addition, in order to further improve the accuracy of the spectral density calculation, this step introduces a multi-window function superposition strategy, that is, calculate the spectral density by superimposing multiple different types of window functions (such as Hanning window, Hamming window, and Blackman window), and take the weighted average to reduce the spectral leakage effect and fluctuation error.

[0048] According to the spectral density result, extract the main frequency components and their amplitudes of the current fluctuation to form the spectral characteristics of the current fluctuation.

[0049] In this embodiment, peak detection is performed on the spectral density by setting an adaptive threshold, and the frequency and amplitude corresponding to each peak are extracted; among them, the setting of the adaptive threshold is based on the fundamental frequency and harmonic characteristics of the current fluctuation data, making the peak detection more sensitive and the false detection rate lower.

[0050] It should be noted that the spectral features extracted in this step not only include the main frequency components and their amplitudes, but also key information such as spectral energy distribution, harmonic characteristics, and phase differences, so as to provide a richer data basis for the construction of the wire feeding resistance fluctuation spectrum in the subsequent steps.

[0051] In this embodiment, in order to more precisely construct the wire feeding resistance fluctuation spectrum, in this step, according to the spectral features of the current fluctuation, combined with the information of the magnetic field distortion degree and the temperature rise gradient in the composite criterion, spectral feature selection is carried out to eliminate environmental interference and irrelevant frequency components, and finally the wire feeding resistance fluctuation spectrum is formed, which specifically includes the following operations: According to the spectral features of the current fluctuation, combined with the information of the magnetic field distortion degree and the temperature rise gradient in the composite criterion, spectral feature screening is carried out.

[0052] In this embodiment, a feature selection algorithm based on mutual information is used to calculate the mutual information amount between the magnetic field distortion degree, the temperature rise gradient and each spectral feature to measure the dependence relationship between them, so as to screen out the spectral features strongly correlated with the wire feeding resistance fluctuation.

[0053] It should be noted that in order to improve the accuracy and robustness of feature selection, this step introduces a recursive feature elimination (RFE) strategy, that is, gradually removes the frequency components with less influence on the spectral features, and dynamically updates the mutual information calculation, and finally retains the spectral features most relevant to the wire feeding resistance fluctuation.

[0054] In addition, in order to avoid misselection of features caused by data fluctuations of the magnetic field distortion degree and the temperature rise gradient, this step introduces a multi-scale feature selection method, that is, calculates the mutual information amount respectively under different time windows and frequency scales and performs weighted averaging to improve the stability and accuracy of feature selection.

[0055] For the selected spectral features, in order to further improve the recognition degree and calculation efficiency of the features, this step optimizes and reduces the dimension of the spectral features.

[0056] In this embodiment, an adaptive principal component analysis (APCA) method is used to reduce the dimension of the spectral features, that is, according to the variance contribution rate of each spectral feature, adaptively determine the number of principal components, and perform linear mapping and dimension reduction on the high-dimensional features to reduce redundant features while retaining the main information.

[0057] Finally, according to the optimized and dimension-reduced spectral features, the wire feeding resistance fluctuation spectrum is constructed.

[0058] It can be seen that the present invention classifies the spectral features according to different frequency bands and amplitude sizes, and uses an adaptive weighted algorithm to calculate the contribution degree of the spectral components in each frequency band, and finally constructs a spectral model closely related to the wire feeding resistance fluctuation.

[0059] In addition, to further improve the accuracy of the fluctuation spectrum, this step introduces a multi-channel spectrum fusion strategy, that is, by fusing and calculating the spectrum features of different measurement channels to eliminate the influence of single-channel measurement errors and environmental interference, thereby constructing a more accurate wire feeding resistance fluctuation spectrum.

[0060] S3: According to the wire feeding resistance fluctuation spectrum, perform load dynamic decoupling and inversely deduce the equivalent friction coefficient in real time.

[0061] S3.1: According to the wire feeding resistance fluctuation spectrum, separate the friction influencing factors and extract the sensitive frequency band of the equivalent friction coefficient.

[0062] In this embodiment, to more accurately inversely deduce the equivalent friction coefficient, this step separates the friction influencing factors according to the wire feeding resistance fluctuation spectrum and extracts the sensitive frequency band of the equivalent friction coefficient, which specifically includes the following operations: First, based on the wire feeding resistance fluctuation spectrum, use the wavelet transform method to decompose the spectrum, divide the spectrum signal into sub-signals of different scales and frequency bands for further analysis of the correlation between different frequency bands and the friction coefficient.

[0063] Next, calculate the spectrum energy of each sub-signal, and combine the wire feeding current fluctuation characteristics to screen out the sensitive frequency band highly correlated with the friction coefficient.

[0064] Exemplarily, the spectrum energy can be calculated through the following process: Perform Fourier transform on each sub-signal to obtain its spectrum representation; calculate the square of the spectrum amplitude and integrate over the entire spectrum range to obtain the spectrum energy of the sub-signal.

[0065] Perform correlation analysis on the spectrum energy of each sub-signal and the wire feeding current fluctuation characteristics to screen out the sensitive frequency band highly correlated with the friction coefficient; use the Pearson correlation coefficient to calculate the correlation between the spectrum energy and the current fluctuation characteristics, and screen out the frequency band with a higher correlation coefficient.

[0066] S3.2: Based on the sensitive frequency band, perform load dynamic decoupling and establish an equivalent friction coefficient inverse deduction model.

[0067] S3.2.1: Based on the sensitive frequency band, use the dynamic decoupling method based on the generalized least squares method to separate the frequency band related to the friction coefficient and other interference frequency bands to obtain the net friction influence signal.

[0068] S3.2.2: During the decoupling process, combine the current fluctuation characteristics and the mechanical stress state, and use the generalized least squares method to dynamically adjust the influence coefficients of different frequency bands to improve the accuracy and stability of decoupling.

[0069] Optionally, a Kalman filter is used to smooth the decoupling result.

[0070] S3.2.3: Based on the decoupled net friction influence signal, establish a nonlinear backstepping model for the equivalent friction coefficient, and use the recursive least squares method for parameter identification to dynamically calculate the real-time equivalent friction coefficient.

[0071] Since the existing friction coefficient calculation methods are prone to interference from multiple factors such as current fluctuations and mechanical stress states under non-stationary working conditions, resulting in low calculation accuracy and difficulty in reflecting the dynamic changes of the friction coefficient in real time; through sensitive frequency band screening, generalized least squares dynamic decoupling, and recursive least squares parameter identification, the present invention not only effectively separates the friction-related frequency band and the interference frequency band, but also improves the accuracy and stability of decoupling by dynamically adjusting the influence coefficient, thereby obtaining a more accurate and real-time updated equivalent friction coefficient; compared with traditional frequency domain filtering, this solution can dynamically adapt to different working conditions in the case of complex current fluctuation characteristics and multiple interference frequency bands, reduce error accumulation, and make the friction coefficient calculation more stable and reliable.

[0072] Optionally, substitute the equivalent friction coefficient into the electromagnetic-thermal-mechanical stress multi-field coupling model to update the friction parameters in the mechanical stress state, make the multi-field coupling model closer to the actual situation, and based on the updated friction parameters, recalculate the mechanical stress state, including tangential stress and normal stress, and feedback the calculation results to the optimization and adjustment process in S5 to achieve dynamic compensation and adjustment of the friction factors during the wire feeding process.

[0073] S4: Determine the overload state of the wire feeding motor through the composite criterion and the dynamic change of the equivalent friction coefficient.

[0074] Preferably, when the combination of the magnetic field distortion degree, the temperature rise gradient, and the equivalent friction coefficient exceeds the set threshold, it is determined that the wire feeding motor is in an overload state.

[0075] Specifically, based on the equivalent friction coefficient obtained in S3, combined with the magnetic field distortion degree and the temperature rise gradient obtained in S2, construct a composite criterion model, and the construction of the composite criterion model is based on the weighted combination of the magnetic field distortion degree, the temperature rise gradient, and the equivalent friction coefficient.

[0076] Set the determination threshold according to the working characteristics of the wire feeding motor and the typical parameter change trend under the overload state.

[0077] Preferably, the determination threshold can be adaptively updated through statistical analysis or machine learning methods based on the motor historical working data and fault records to further improve the accuracy and robustness of the determination.

[0078] Optionally, to avoid misjudgment, upper and lower limit determination thresholds can be set and 。

[0079] Further, based on the calculated composite criterion value and the set threshold, the overload state is determined, as Figure 2 shown below: When , it is determined that the wire feeding motor is in an overload state; When , it is determined that the wire feeding motor is in a normal state; When , it is determined that the wire feeding motor is in a critical state, and further analysis is carried out in combination with the dynamic change trend of the equivalent friction coefficient: if the equivalent friction coefficient shows a continuous upward trend and the difference between the composite criterion and the upper limit threshold is less than the difference threshold , potential overload risks are pre-warned in advance; otherwise, the current state judgment is maintained.

[0080] Through dynamic determination, not only the accuracy of overload state identification is improved, but also misjudgment and lag response can be pre-warned and avoided.

[0081] S5: Based on the determination result of the overload state, adjust the parameters in the electromagnetic-thermal-mechanical stress multi-field coupling model, and feedback them to the link of extracting the wire feeding resistance fluctuation spectrum to optimize the accuracy and response speed of subsequent overload detection.

[0082] For the determined overload state, generate corresponding parameter adjustment strategies, including parameters in multiple dimensions such as electromagnetic, thermal, and mechanical stress. Among them, the parameter adjustment strategy includes the following content: According to the magnetic field distortion degree in the overload state, dynamically adjust the electromagnetic coupling coefficient to reduce the influence of electromagnetic interference on the fluctuation spectrum; according to the change of the temperature rise gradient, correct the heat conduction coefficient and temperature compensation coefficient to more accurately reflect the influence of temperature on mechanical stress; in combination with the dynamic change of the equivalent friction coefficient, update the friction attenuation coefficient and elastic modulus to optimize the calculation accuracy of the mechanical stress state.

[0083] According to the parameter adjustment strategy, key parameters such as the electromagnetic coupling coefficient, heat conduction coefficient, friction attenuation coefficient, and elastic modulus are corrected in real time; the updated parameters are used to recalculate the mechanical stress state and compared with the calculation results of the previous cycle to verify the update effect.

[0084] In an embodiment of the present invention, the parameters in the updated electromagnetic-thermal-mechanical stress multi-field coupling model are feedback to the wire feeding resistance fluctuation spectrum extraction link in step S2: Based on the updated parameters, recalculate the resistance fluctuation spectrum to more accurately reflect the actual working state of the wire feeding motor; perform a secondary screening on the spectrum energy concentration and sensitive frequency bands to eliminate the interference frequency bands introduced by parameter errors and improve the accuracy of spectrum analysis. During the feedback process, adopt an incremental update method to reduce the interference to the spectrum extraction link, and continuously optimize the accuracy and response speed of overload detection through cyclic iteration; at the same time, compare the spectrum extraction results before and after feedback. If the difference between the two is significant, trigger the regeneration of the parameter adjustment strategy to further optimize the parameter settings.

[0085] Furthermore, this embodiment also provides an overload detection system for a wire feeding motor, including A composite criterion calculation module, which is used to establish an electromagnetic-thermal-mechanical stress multi-field coupling model of the wire feeding motor, input the first input data, calculate the magnetic field distortion degree and temperature rise gradient, and form a composite criterion; the first input data includes the real-time current, temperature and stress data of the wire feeding motor; A spectrum analysis module, which is used to analyze the current fluctuation characteristics during the wire feeding process based on the composite criterion, and extract the wire feeding resistance fluctuation spectrum by using spectrum analysis; A decoupling and friction coefficient inverse deduction module, which is used to perform load dynamic decoupling according to the wire feeding resistance fluctuation spectrum and inversely deduce the equivalent friction coefficient in real time; An overload state determination module, which is used to determine the overload state of the wire feeding motor through the composite criterion and the dynamic change of the equivalent friction coefficient; An adjustment and feedback module, which is used to adjust the parameters in the electromagnetic-thermal-mechanical stress multi-field coupling model based on the determination result of the overload state and feedback it to the link of extracting the wire feeding resistance fluctuation spectrum.

[0086] This embodiment also provides a computer device, which is applicable to the situation of the overload detection method of the wire feeding motor, including a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions to implement the overload detection method of the wire feeding motor proposed in the above embodiment.

[0087] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a carrier network, NFC (Near Field Communication), or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or may be a button, a trackball, or a touchpad provided on the housing of the computer device, or may also be an external keyboard, touchpad, or mouse, etc.

[0088] This embodiment also provides a storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements the overload detection method of the wire feeding motor as proposed in the above embodiment.

[0089] In summary, the present invention combines a multi-field coupling model with online identification of the friction coefficient to achieve precise detection and real-time response to the overload state of the wire feeding motor; uses the spectrum of the wire feeding resistance fluctuation to dynamically deduce the friction coefficient, and effectively determines the overload state through a composite criterion, avoiding the problem of easy misjudgment of a single parameter in the traditional overload detection method.

[0090] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. A method for detecting an overload of a wire feeding motor, characterized in that: include: Establish an electromagnetic-thermal-mechanical stress multi-field coupling model of the wire feeding motor, input the first input data, calculate the magnetic field distortion and temperature rise gradient, and form a composite criterion; The first input data includes real-time current, temperature and stress data of the wire feeding motor; Based on the composite criterion, the current fluctuation characteristics in the wire feeding process are analyzed, and the wire feeding resistance fluctuation spectrum is extracted by spectrum analysis; According to the wire feeding resistance fluctuation spectrum, dynamic load decoupling is performed to reversely estimate the equivalent friction coefficient in real time; Determining the overload state of the wire feeding motor by means of the composite criterion and the dynamic change of the equivalent friction coefficient; Based on the determination result of the overload state, the parameters in the electromagnetic-thermal-mechanical stress multi-field coupling model are adjusted and fed back to the link of extracting the wire feeding resistance fluctuation spectrum.

2. The overload detection method for a wire feeding motor according to claim 1, characterized in that: The establishment of the electromagnetic-thermal-mechanical stress multi-field coupling model includes: Based on the current data, the electromagnetic field model inside the wire feeding motor is established using Maxwell equations, and the electromagnetic field parameters are initialized; According to the electromagnetic field model, the heat source generated by electromagnetic loss is calculated, and the temperature field model is established through the Fourier heat conduction equation to initialize the temperature field parameters; According to the temperature field gradient and electromagnetic field force, the generalized Hooke's law and thermal stress coupling analysis are used to establish the mechanical stress field model and initialize the mechanical stress field parameters; By establishing electromagnetic, thermal and mechanical stress fields in sequence and establishing data sharing and parameter transfer relationships between each field, a coupling relationship is established between the fields to form a complete multi-field coupling model.

3. The overload detection method of the wire feeding motor according to claim 2, characterized in that: The composite criterion is formed by weighting the magnetic field distortion degree and the temperature rise gradient.

4. The overload detection method for a wire feeding motor according to claim 1, characterized in that: The analyzing of the current fluctuation characteristics during the wire feeding process and the extraction of the wire feeding resistance fluctuation spectrum by spectrum analysis include: According to the pre-processed current fluctuation data, spectrum analysis is performed to calculate the spectrum density of the current fluctuation and obtain the spectrum characteristics of the current fluctuation; According to the spectrum characteristics of the current fluctuation, combined with the information of magnetic field distortion and temperature rise gradient in the composite criterion, spectrum characteristics are selected to construct the wire feeding resistance fluctuation spectrum; The spectrum feature selection includes: The mutual information-based feature selection algorithm is used to calculate the mutual information between the magnetic field distortion, temperature rise gradient and each spectrum feature, measure the dependency relationship, and select spectrum features. The screened spectral features are optimized and dimensionally reduced.

5. The overload detection method for a wire feeding motor according to claim 4, characterized in that: According to the wire feeding resistance fluctuation spectrum, performing load dynamic decoupling includes the following steps: Based on the wire feeding resistance fluctuation spectrum, the spectrum is decomposed by using a wavelet transform method, and the spectrum signal is divided into sub-signals of different scales and frequency bands; Calculate the spectrum energy of each sub-signal and select the sensitive frequency band based on the fluctuation characteristics of wire feeding current; Based on the sensitive frequency band, a dynamic decoupling method based on generalized least squares method is adopted to separate the friction coefficient related frequency band and other interference frequency bands to obtain the net friction influence signal.

6. The overload detection method for a wire feeding motor according to claim 5, characterized in that: The generation of the equivalent friction coefficient includes: According to the net friction influence signal after decoupling, a nonlinear back-stepping model of the equivalent friction coefficient is established, and the recursive least squares method is used for parameter identification to dynamically calculate the real-time equivalent friction coefficient.

7. The overload detection method for a wire feeding motor according to claim 1, characterized in that: Determining the overload state of the wire feeding motor includes: According to the equivalent friction coefficient, combined with the magnetic field distortion and temperature rise gradient, a composite criterion model is constructed to calculate the composite criterion value. ; Set the upper and lower limit judgment thresholds according to the working characteristics of the wire feeding motor and the typical parameter change trend under overload conditions and .

8. The overload detection method for a wire feeding motor according to claim 7, characterized in that: The determining the overload state of the wire feeding motor further comprises: According to the calculated composite criterion value And the set judgment threshold, to judge the overload status: when When , it is determined that the wire feeding motor is in an overload state; when When , it is determined that the wire feeding motor is in normal state; when When the wire feeding motor is in a critical state, further analysis is performed in combination with the dynamic change trend of the equivalent friction coefficient: if the equivalent friction coefficient shows a continuous upward trend and the difference between the composite criterion and the upper threshold is less than the difference threshold , then an early warning of potential overload risk will be issued; otherwise, the current status judgment will be maintained.

9. An overload detection system for a wire feeding motor, based on the overload detection method for a wire feeding motor according to any one of claims 1 to 8, characterized in that: Also includes: A composite criterion calculation module is used to establish an electromagnetic-thermal-mechanical stress multi-field coupling model of the wire feeding motor, input first input data, calculate the magnetic field distortion and temperature rise gradient, and form a composite criterion; the first input data includes real-time current, temperature and stress data of the wire feeding motor; A spectrum analysis module, used to analyze the current fluctuation characteristics during the wire feeding process based on the composite criterion, and extract the wire feeding resistance fluctuation spectrum by spectrum analysis; A decoupling and friction coefficient reverse estimation module is used to perform load dynamic decoupling and reverse estimation of the equivalent friction coefficient in real time according to the wire feeding resistance fluctuation spectrum; An overload state determination module, used to determine the overload state of the wire feeding motor through the composite criterion and the dynamic change of the equivalent friction coefficient; The adjustment and feedback module is used to adjust the parameters in the electromagnetic-thermal-mechanical stress multi-field coupling model based on the judgment result of the overload state, and feed back to the link of extracting the wire feeding resistance fluctuation spectrum.

Images