A servo motor energy-saving control system and method

Abstract

The present invention discloses a servo motor energy-saving control system and method, which relates to the technical field of servo motor energy saving. It includes: based on the magnetic field intensity matrix and the temperature distribution matrix, obtaining the magnetothermal coupling coefficient, identifying the temperature anomaly area, and generating a hot spot marker; based on the magnetothermal coupling coefficient and the hot spot marker, encoding the magnetothermal coupling coefficient into a pulse sequence through a pulsed neural network, and using the synaptic weight update mechanism of the hidden layer neurons to generate an energy recovery threshold and an efficiency threshold; based on the energy recovery threshold and the efficiency threshold, performing dynamic energy recovery on the servo motor, collecting the vibration signal after energy recovery, calculating the maximum Lyapunov exponent, and updating the pulsed neural network. The present invention realizes the precise correlation mapping between magnetic field distortion and local temperature rise through the calculation of the magnetothermal coupling coefficient and the hot spot marker, breaks through the limitations of traditional independent analysis, and provides a physical basis for energy recovery.

Description

Technical Field

[0001] The present invention relates to the technical field of servo motor energy saving, and in particular to a servo motor energy saving control system and method. Background Art

[0002] Servo motor energy saving control technology has received extensive attention in the field of industrial automation in recent years. Traditional methods mainly rely on PID regulation of the current loop and speed loop, combined with vector control to achieve energy efficiency optimization. With the development of sensing technology, the collaborative monitoring of magnetic field intensity and temperature distribution has become a research hotspot. Existing technologies can already collect motor operation parameters through Hall sensors and thermocouple arrays, and use methods such as Kalman filtering and wavelet transform for data processing. In terms of energy recovery, technologies such as regenerative braking and supercapacitor energy storage have achieved partial energy reuse, and intelligent control algorithms based on neural networks have also been gradually applied to motor energy efficiency optimization. However, the existing technologies have insufficient dynamic characterization of the magnetic field-temperature coupling effect, making it difficult to accurately identify the correlation between local overheating areas and energy losses, resulting in a lack of physical basis for setting the energy recovery threshold, which restricts the further improvement of energy saving efficiency.

[0003] The existing technologies have the following limitations: First, the processing of magnetic field and temperature data is usually carried out independently, lacking quantitative analysis of the magnetothermal coupling effect and unable to establish a mapping relationship between magnetic field distortion and local temperature rise; Second, the setting of the energy recovery threshold mostly relies on empirical formulas or static parameters, without considering the dynamic nonlinear characteristics of the motor operation state, resulting in over-recovery or under-recovery problems. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a servo motor energy saving control method to solve the problems of the lack of correlation between magnetic field and temperature and the static energy recovery threshold in the existing technologies.

[0006] To solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a method for energy-saving control of a servo motor, which includes collecting magnetic field intensity data and temperature distribution data of the servo motor and performing preprocessing, and simultaneously obtaining a magnetic field intensity matrix and a temperature distribution matrix; based on the magnetic field intensity matrix and the temperature distribution matrix, obtaining a magneto-thermal coupling coefficient, identifying a temperature anomaly region, and generating a hot spot marker; based on the magneto-thermal coupling coefficient and the hot spot marker, encoding the magneto-thermal coupling coefficient into a pulse sequence through a spiking neural network, and using the synaptic weight update mechanism of the hidden layer neurons to generate an energy recovery threshold and an efficiency threshold; based on the energy recovery threshold and the efficiency threshold, performing dynamic energy recovery on the servo motor, collecting the vibration signal after energy recovery, calculating the maximum Lyapunov exponent, and updating the spiking neural network.

[0008] As a preferred solution of the servo motor energy-saving control method of the present invention, wherein: the preprocessing includes noise filtering, outlier removal, normalization, spatio-temporal alignment, and data standardization.

[0009] As a preferred solution of the servo motor energy-saving control method of the present invention, wherein: the steps of obtaining the magnetic field intensity matrix and the temperature distribution matrix are as follows

[0010] Based on the preprocessed magnetic field intensity data, a magnetic field intensity matrix is constructed by the grid interpolation method;

[0011] Based on the preprocessed temperature distribution data, a temperature distribution matrix is constructed by the Kriging interpolation method.

[0012] As a preferred solution of the servo motor energy-saving control method of the present invention, wherein: the steps of obtaining the magneto-thermal coupling coefficient are as follows

[0013] Through the collaborative filtering algorithm, the magnetic field intensity eigenvector in the magnetic field intensity matrix and the temperature distribution eigenvector in the temperature distribution matrix are extracted;

[0014] The singular value decomposition method is used to decompose the magnetic field intensity eigenvector and the temperature distribution eigenvector into eigenvectors, and combined with the tensor product and the non-linear activation function, the magneto-thermal coupling coefficient is obtained.

[0015] As a preferred solution of the servo motor energy-saving control method of the present invention, wherein: the steps of identifying the temperature anomaly region and generating the hot spot marker are as follows

[0016] Based on the magneto-thermal coupling coefficient, the temperature anomaly region of the servo motor is identified by the local outlier factor method, and the coupling threshold is calculated through the fuzzy membership function;

[0017] According to the temperature anomaly region and the coupling threshold, a hot spot marker is generated by the fuzzy logic discrimination method.

[0018] As a preferred embodiment of the servo motor energy-saving control method of the present invention, the method includes: based on the magneto-thermal coupling coefficient and the hot spot marker, encoding the magneto-thermal coupling coefficient into a pulse sequence through a spiking neural network, and generating an energy recovery threshold and an efficiency threshold by using the synaptic weight update mechanism of the hidden layer neurons. The steps are as follows:

[0019] Based on the hot spot marker, convert the magneto-thermal coupling coefficient into a pulse sequence through a pulse encoder;

[0020] Input the pulse sequence into the hidden layer neurons, and each neuron updates its membrane potential according to the input pulse sequence;

[0021] Define the membrane potential threshold of the neuron based on the historical pulse firing frequency and the dynamic range of the membrane potential;

[0022] Calculate the pulse firing frequency based on the updated membrane potential of the neuron;

[0023] By analyzing the historical energy recovery data, use the linear regression method to fit the relationship between the pulse firing frequency and the energy recovery value, and construct an energy recovery function by using a smoothing factor;

[0024] By analyzing the historical efficiency data, use the non-linear fitting method to capture the non-linear influence of the pulse firing frequency on the servo motor efficiency, and construct an efficiency function by combining a saturation factor;

[0025] Generate an energy recovery threshold and an efficiency threshold according to the pulse firing frequency, the energy recovery function, and the efficiency function.

[0026] As a preferred embodiment of the servo motor energy-saving control method of the present invention, the method includes: based on the energy recovery threshold and the efficiency threshold, perform dynamic energy recovery on the servo motor, collect the vibration signal after energy recovery, calculate the maximum Lyapunov exponent, and update the spiking neural network. The steps are as follows:

[0027] Real-time monitor the real-time energy parameters of the servo motor. When the real-time energy parameters exceed the energy recovery threshold, perform energy recovery, and adjust the intensity of energy recovery according to the efficiency threshold;

[0028] Based on the vibration signal of the servo motor, calculate the maximum Lyapunov exponent through phase space reconstruction and the small data method;

[0029] Update the spiking neural network according to the maximum Lyapunov exponent.

[0030] In a second aspect, the present invention provides a servo motor energy-saving control system, including a data acquisition module, a hot spot marking module, a threshold generation module, and an energy recovery module; the data acquisition module is configured to collect magnetic field intensity data and temperature distribution data of the servo motor, perform preprocessing, and obtain a magnetic field intensity matrix and a temperature distribution matrix; the hot spot marking module is configured to obtain a magnetothermal coupling coefficient based on the magnetic field intensity matrix and the temperature distribution matrix, identify a temperature anomaly region, and generate a hot spot mark; the threshold generation module is configured to encode the magnetothermal coupling coefficient into a pulse sequence through a pulsed neural network based on the magnetothermal coupling coefficient and the hot spot mark, and generate an energy recovery threshold and an efficiency threshold by using a synaptic weight update mechanism of hidden layer neurons; the energy recovery module is configured to perform dynamic energy recovery on the servo motor based on the energy recovery threshold and the efficiency threshold, collect vibration signals after energy recovery, calculate the maximum Lyapunov exponent, and update the pulsed neural network.

[0031] In a third aspect, the present invention provides a computer device, including a memory and a processor, where the memory stores a computer program, and: when the computer program is executed by the processor, any step of the servo motor energy-saving control method described in the first aspect of the present invention is implemented.

[0032] In a fourth aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, and: when the computer program is executed by the processor, any step of the servo motor energy-saving control method described in the first aspect of the present invention is implemented.

[0033] The beneficial effects of the present invention are as follows: Through the calculation of the magnetothermal coupling coefficient and hot spot marking, the precise correlation mapping between magnetic field distortion and local temperature rise is realized, breaking through the limitations of traditional independent analysis, and providing a physical basis for energy recovery; by encoding the magnetothermal coupling coefficient into a pulse sequence through a pulsed neural network and using a synaptic weight update mechanism of hidden layer neurons, the real-time response ability of the servo motor to energy recovery and efficiency optimization during operation is improved. Description of the Drawings

[0034] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for description in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings without creative efforts based on these drawings.

[0035] Figure 1 It is a flowchart of the servo motor energy-saving control method in Embodiment 1.

[0036] Figure 2 It is a schematic diagram of the servo motor energy-saving control system in Embodiment 1.

[0037] Figure 3 It is a flowchart for generating hotspot markers in Embodiment 1.

[0038] Figure 4 It is a flowchart for the operation of the spiking neural network in Embodiment 1. Specific Embodiments

[0039] To make the above objects, features, and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings of the specification.

[0040] Many specific details are set forth in the following description to facilitate a thorough understanding of the present invention. However, the present invention may be practiced in other ways different from those described herein. Those skilled in the art can make similar generalizations without departing from the spirit of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0041] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The appearances of "in one embodiment" in different places in this specification do not all refer to the same embodiment, nor are they separate or alternative embodiments that exclude each other.

[0042] Embodiment 1, referring to Figures 1 to 4 This embodiment provides a servo motor energy-saving control method, including the following steps:

[0043] S1. Collect the magnetic field intensity data and temperature distribution data of the servo motor and perform preprocessing, and at the same time obtain the magnetic field intensity matrix and temperature distribution matrix;

[0044] The magnetic field intensity data of the servo motor is collected by a Hall sensor, and the temperature distribution data is collected by a thermocouple array.

[0045] The preprocessing includes noise filtering, outlier removal, normalization, spatio-temporal alignment, and data standardization for the magnetic field intensity data and temperature distribution data.

[0046] Noise filtering uses a low-pass filter to process the magnetic field intensity data collected by the Hall sensor and the temperature distribution data collected by the thermocouple array, removing high-frequency interference signals and retaining the effective data components.

[0047] Outlier removal screens the magnetic field intensity data and temperature distribution data through the 3σ criterion, removing data points outside the range of three times the standard deviation from the mean to ensure data reliability.

[0048] Normalization linearly scales the magnetic field intensity data based on a preset magnetic saturation threshold (0.5T - 1.8T), and normalizes the temperature distribution data according to the ambient temperature reference value (25°C) and the maximum allowable temperature rise (150°C), so as to unify the data to the same dimension.

[0049] Spatio-temporal alignment synchronizes the magnetic field intensity data and the temperature distribution data by timestamp matching, ensuring that the magnetic field intensity value and the temperature value at the same time point correspond, and eliminating the time deviation.

[0050] Data standardization converts the processed magnetic field intensity data and temperature distribution data into a 4×3 dimensional magnetic field intensity matrix and a temperature distribution matrix that can be directly used for analysis, completing the preprocessing process.

[0051] Based on the preprocessed magnetic field intensity data, a magnetic field intensity matrix is constructed by the grid interpolation method.

[0052] Furthermore, the preprocessed magnetic field intensity data contains discrete sampling points at different spatial positions. The grid interpolation method first divides the spatial region of the servo motor into a regular grid of 4 rows and 3 columns; then, based on the magnetic field intensity values of the sampling points, the bilinear interpolation algorithm is used to calculate the magnetic field intensity values of the unsampled points within the grid, ensuring that each grid cell is filled with magnetic field intensity data. Finally, the magnetic field intensity values of all grid cells are arranged according to their positions to generate a 4×3 dimensional magnetic field intensity matrix.

[0053] Based on the preprocessed temperature distribution data, a temperature distribution matrix is constructed by the Kriging interpolation method.

[0054] Furthermore, first perform spatial correlation analysis. Using the spatial autocorrelation analysis method commonly used in Geographic Information System (GIS), calculate the spatial distance matrix and temperature difference matrix between each sampling point. Evaluate the global spatial autocorrelation by calculating Moran's I, and use the Local Indicators of Spatial Association (LISA) to identify local clustering patterns. This step requires setting a reasonable neighborhood search radius, usually using the adaptive bandwidth determination method. Then perform temperature value estimation, using the ordinary Kriging method in spatial interpolation technology. When establishing the spatial weight matrix, use the inverse distance weighting method to determine the influence weights of neighboring points. For each point to be estimated, select the optimal number of neighboring sampling points to participate in the calculation, usually using the Akaike Information Criterion (AIC) to determine the optimal number of neighborhood points. During the calculation process, it is necessary to solve the Kriging equations and use the LU decomposition method for matrix inversion. Finally, perform grid processing, arranging the estimated temperature values in a regular grid. The grid division uses the equal-spacing division method, evenly divided into 3 columns in the horizontal direction and 4 rows in the vertical direction. The temperature value of each grid cell is determined according to its center point coordinates, and the nearest neighbor method is used for coordinate matching. The finally preprocessed temperature distribution data contains discrete sampling points at different spatial positions. The Kriging interpolation method first analyzes the spatial correlation between sampling points through GIS. Then, based on the spatial correlation between sampling points, use the Akaike Information Criterion to determine the optimal number of neighborhood points, estimate the temperature values of unsampled points according to the optimal number of neighborhood estimated points. Finally, arrange the temperature estimation values of all grid cells in a regular grid. The grid division uses the equal-spacing division method, evenly divided into 3 columns in the horizontal direction and 4 rows in the vertical direction. The temperature value of each grid cell is determined according to its center point coordinates, and the nearest neighbor method is used for coordinate matching. Generate a 4×3-dimensional temperature distribution matrix.

[0055] S2. Based on the magnetic field intensity matrix and the temperature distribution matrix, obtain the magnetothermal coupling coefficient, identify the temperature anomaly region, and generate hot spot markers;

[0056] Extract the magnetic field intensity eigenvector in the magnetic field intensity matrix and the temperature distribution eigenvector in the temperature distribution matrix through the collaborative filtering algorithm;

[0057] It should be noted that first perform dimensionality reduction processing on the magnetic field intensity matrix using Principal Component Analysis (PCA), and retain the principal components with a cumulative contribution rate exceeding 95% as the magnetic field eigenvectors; at the same time, use the Non-Negative Matrix Factorization (NMF) technology to process the temperature distribution matrix, and obtain the temperature eigenvectors through iterative decomposition by the Alternating Least Squares (ALS). Subsequently, use the Maximal Information Coefficient (MIC) to evaluate the correlation between the two groups of eigenvectors, and screen out the feature pairs with a correlation coefficient greater than 0.8.

[0058] Using the singular value decomposition method, the feature vectors based on the magnetic field intensity and temperature distribution are decomposed into eigenvectors to capture the implicit relationship between them. Combining the tensor product and the non-linear activation function, the magneto-thermal coupling coefficient is obtained, and the expression is as follows:

[0059] ;

[0060] Among them, represents the magneto-thermal coupling coefficient, represents the -th magnetic field intensity eigenvector in the magnetic field intensity distribution matrix, represents the -th temperature distribution eigenvector in the temperature distribution matrix, represents the tensor product operator, represents the weight matrix for adjusting the tensor product operation of the magnetic field intensity eigenvector and the temperature eigenvector, represents the total number of magnetic field intensity vectors, represents the total number of temperature distribution eigenvectors;

[0061] It should be noted that first, the magnetic field intensity distribution matrix is eigen-decomposed to extract orthogonal magnetic field intensity eigenvectors ; at the same time, the temperature distribution matrix is eigen-decomposed to extract orthogonal temperature distribution eigenvectors ; then, according to the requirements of the expression, all possible combinations of eigenvectors (a total of × pairs) are subjected to the tensor product operation to generate high-order interaction features; then each tensor product result is dot-multiplied with the learnable weight matrix , and the ReLU non-linear activation function is applied to the summation result of all and ; the finally output scalar value is the required magneto-thermal coupling coefficient. Based on the magneto-thermal coupling coefficient, the temperature abnormal area of the servo motor is identified by the local outlier factor method;

[0062]

[0063] ​It should be noted that first, the magnetothermal coupling coefficient is correlated with the temperature values in the temperature distribution matrix to generate comprehensive characteristic data reflecting the relationship between temperature distribution and magnetic field strength. Then, the local outlier factor method is used to calculate the comprehensive characteristic data to evaluate the degree of difference between each temperature region and its surrounding regions. For temperature regions with a significantly higher degree of difference than the surrounding regions, they are marked as temperature anomaly regions. At the same time, combined with the actual temperature values in the temperature distribution matrix, the abnormal characteristics of the marked regions are further verified. Finally, the temperature anomaly regions of the servo motor are output.

[0064] Furthermore, the specific process of the correlation analysis:

[0065] The correlation analysis is carried out using the Pearson Correlation Coefficient method. First, the magnetothermal coupling coefficient matrix and the temperature distribution matrix are spatially aligned to ensure that the magnetic field characteristics and temperature values of each grid cell correspond one by one. By calculating the ratio of the covariance to the standard deviation of the magnetothermal coupling coefficient and the temperature value at each spatial position, a correlation coefficient matrix is obtained. The analysis results show that under normal operating conditions, there is a significant positive correlation between the two (for example, >0.7), while in the abnormal region, the correlation weakens (for example, <0.3) or shows a negative correlation characteristic. This correlation coefficient matrix is the comprehensive characteristic data for subsequent analysis.

[0066] The calculation process of the local outlier factor:

[0067] The local outlier factor (LOF) calculation is implemented using the k-Nearest Neighbors algorithm. Taking each temperature region as the center, its nearest neighbor regions are found in the comprehensive characteristic data. Calculate the ratio of the local reachability density of this region to its neighborhood, and this ratio is the LOF value. Specifically, when calculating, first determine the k-distance neighborhood of each point, then calculate the local reachability density (LRD), and finally obtain the LOF value by comparing the LRD ratio of the target point to its neighborhood.

[0068] The evaluation criteria for the degree of difference:

[0069] The degree of difference is quantitatively evaluated through the statistical distribution of the LOF value. Set the difference threshold K1. When the LOF value exceeds the mean value K1 of the same-layer regions, it is determined to be a significant difference. This difference is manifested as: 1) the destruction of spatial continuity, and the magnetothermal correlation in the abnormal region suddenly decreases; 2) indicating a sudden change in the temperature gradient.

[0070] Based on the magnetothermal coupling coefficient, through the fuzzy membership function, the coupling threshold is calculated, and the expression is:

[0071] ;

[0072] Among them, is the coupling threshold, is the magnetothermal coupling coefficient is the membership degree belonging to the hot spot, ranging between [0, 1], represents the slope parameter of the fuzzy membership function, represents the offset parameter of the fuzzy membership function;

[0073] It should be noted that first, the magnetothermal coupling coefficient is input into the fuzzy membership function, and according to the correlation degree between the magnetothermal coupling coefficient and the hot spot area, its membership value is calculated. The membership value reflects the possibility that the magnetothermal coupling coefficient belongs to the hot spot area, ranging from zero to one. Then, according to the characteristics of the membership function, the critical point at which the magnetothermal coupling coefficient reaches the hot spot area, that is, the coupling threshold, is determined. The coupling threshold is used to judge whether the magnetothermal coupling coefficient belongs to the hot spot area, providing a basis for subsequent hot spot marking.

[0074] Based on the temperature anomaly area and the coupling threshold, a hot spot mark is generated through the fuzzy logic discrimination method.

[0075] Furthermore, first analyze whether the temperature value of each unit in the temperature distribution matrix is located in the temperature anomaly area. If it is located in the anomaly area, further check whether the magnetothermal coupling coefficient of this unit exceeds the coupling threshold calculated by the fuzzy membership function; if both of these conditions are met at the same time, mark this unit as a hot spot. For example, if the temperature value of a certain unit is located in the anomaly area and the magnetothermal coupling coefficient exceeds the threshold, the hot spot mark is generated as 1, otherwise it is marked as 0.

[0076] S3. Based on the magnetothermal coupling coefficient and the hot spot mark, encode the magnetothermal coupling coefficient into a pulse sequence through a spiking neural network, and use the synaptic weight update mechanism of the hidden layer neurons to generate an energy recovery threshold and an efficiency threshold;

[0077] Based on the hot spot mark, through the pulse encoder, the magnetothermal coupling coefficient and is converted into a pulse sequence;

[0078] It should be noted that first, the pulse encoder receives the magnetothermal coupling coefficient as an input signal, and at the same time, feature weighting processing is carried out in combination with the hot spot marker. The frequency and intensity of pulse emission are determined according to the numerical characteristics of the magnetothermal coupling coefficient and the status of the hot spot marker. The pulse encoder adopts an encoding rule based on amplitude-frequency conversion to convert the continuous magnetothermal coupling coefficient value into a discrete pulse signal. During the conversion process, the numerical size of the magnetothermal coupling coefficient and the hot spot marker jointly determine the density of pulse emission. For the area marked as a hot spot, the corresponding magnetothermal coupling coefficient will generate a higher-frequency pulse sequence, while the area without a hot spot marker maintains the conventional encoding method. The finally output pulse sequence completely retains the characteristic information of the original magnetothermal coupling coefficient, and reflects the spatial characteristics of the hot spot marker through the pulse frequency difference, providing a standardized signal form for the input of the subsequent pulse neural network.

[0079] The pulse sequence is input into the neurons in the hidden layer. Each neuron updates the membrane potential of each neuron according to the input pulse sequence. The expression is:

[0080] ;

[0081] Where, represents the membrane potential of the th neuron updated at the current time , represents the membrane potential of the th neuron at the previous time , represents the synaptic weight of the th input pulse, represents the command value indicating whether a pulse is emitted at time , is the total number of input pulses, is the th precise time point when the

[0082] th input pulse actually arrives at the neuron synapse; It should be noted that first, the neurons in the hidden layer receive the input signal from the pulse sequence, and each input pulse affects the membrane potential of the neuron. The neuron adjusts the degree of change of its membrane potential according to the synaptic weight of the input pulse, and the pulse with a larger weight has a greater impact on the membrane potential. At the same time, the neuron judges whether the membrane potential needs to be updated at the current moment according to the time characteristics of the input pulse. If the input pulse arrives at a specific moment (the specific moment refers to the precise time point when the input pulse signal actually arrives at the neuron synapse ), the neuron will cumulatively update the membrane potential according to the weight and arrival time of the pulse. Finally, the membrane potential of the neuron reflects the comprehensive influence of all input pulses.

[0083] Define the membrane potential threshold of a neuron based on the historical impulse firing frequency and the dynamic range of the membrane potential ;

[0084] It should be noted that the dynamic range of the membrane potential refers to the change interval of the neuron membrane potential within a specific time, reflecting the activity degree of the neuron's response to input impulses. The upper limit of the dynamic range of the membrane potential is the maximum value of the neuron membrane potential, representing the potential level when the neuron is stimulated by strong input impulses; the lower limit is the minimum value of the neuron membrane potential, representing the potential level when the neuron is not stimulated by input impulses or the input impulses are weak. By analyzing the dynamic range of the membrane potential, the sensitivity and response ability of the neuron to input impulses can be understood, thus providing a basis for defining the membrane potential threshold. The size of the dynamic range of the membrane potential directly affects the setting of the membrane potential threshold. A larger dynamic range usually corresponds to a higher membrane potential threshold, while a smaller dynamic range corresponds to a lower membrane potential threshold.

[0085] Based on the updated membrane potential of the neuron Calculate the impulse firing frequency , and the expression is:

[0086] ;

[0087] By analyzing the historical energy recovery data, use the linear regression method to fit the relationship between the impulse firing frequency and the energy recovery value, and use a smoothing factor to construct an energy recovery function , and the expression is:

[0088] ;

[0089] Among them, is the fitting coefficient, reflecting the influence degree of the impulse firing frequency on the energy recovery, is the offset;

[0090] It should be noted that the process of obtaining the fitting coefficient is as follows: First, collect the impulse firing frequency data of the servo motor under different working conditions, and record the corresponding energy recovery efficiency at the same time. Use the least squares method to perform curve fitting on the relationship between the impulse firing frequency and the energy recovery efficiency, and obtain a linear regression equation reflecting the correlation between the two. The slope of this equation is the fitting coefficient , and its numerical value directly reflects the influence degree of the impulse firing frequency on the energy recovery efficiency.

[0091] The offset is obtained as follows: On the basis of determining the fitting coefficient , analyze the energy recovery efficiency reference value when the impulse firing frequency is zero. Measure the energy recovery efficiency of the servo motor in the state of no impulse input through experiments, and combine the fitting coefficient Based on the calculation result, the intercept value of the regression equation is determined as the offset . The offset reflects the basic influence of other factors on the energy recovery efficiency except the pulse emission frequency.

[0092] The specific other factors refer to the environmental temperature change during the operation of the servo motor, the friction loss of mechanical transmission components, the electromagnetic field strength fluctuation, the power supply voltage stability, the instantaneous change characteristics of the mechanical load, the thermal loss of the winding resistance, the permanent magnet demagnetization effect, the bearing lubrication state, and the efficiency characteristics of the power electronic converter, etc.

[0093] By analyzing the historical efficiency data, a non - linear fitting method is used to capture the non - linear influence of the pulse emission frequency on the servo motor efficiency, and an efficiency function is constructed by combining with the saturation factor , and the expression is:

[0094] ;

[0095] where, is the saturation factor, which controls the stable value of the function, is the non - linear coefficient, which is used to control the influence rate of the pulse emission frequency on the efficiency, is the base of the natural logarithm;

[0096] It should be noted that first, the historical efficiency data of the servo motor at different pulse emission frequencies are collected, and the change law between the pulse emission frequency and the efficiency is analyzed by the non - linear fitting method. According to the fitting result, the key characteristic parameters affecting the efficiency change are determined, including the saturation factor that controls the stable value of the function and the non - linear coefficient that controls the influence rate. These two characteristic parameters are substituted into a mathematical relationship with asymptotic characteristics to form a function that can accurately describe the influence of the pulse emission frequency on the efficiency.

[0097] The determination process of the saturation factor and the non - linear coefficient is as follows: First, based on the measured data of the servo motor under various working conditions, a corresponding relationship database between the pulse emission frequency and the energy recovery efficiency is established. The measured data in the database are subjected to non - linear regression analysis, and the least - squares method is used to fit the curve to find the mathematical model that best reflects the relationship between the pulse emission frequency and the efficiency. During the fitting process, the model parameters are adjusted by the iterative optimization algorithm to minimize the error between the predicted value and the measured value. In the finally converged optimization result, the parameter that controls the stable value of the curve is the saturation factor, and the parameter that determines the curve change rate is the non - linear coefficient.

[0098] Based on the pulse emission frequency and the energy recovery function , the energy recovery threshold is calculated, and the expression is:

[0099] ;

[0100] Wherein, is the energy recovery threshold, represents the energy recovery value at time . The pulse firing frequency is mapped to the energy recovery value. represents the command value indicating whether a pulse is fired at time . represents the current time, represents the specific time point of pulse firing;

[0101] It should be noted that first, a complete record of the change of the pulse firing frequency with time is obtained. The pulse firing frequency at each moment is input into the energy recovery function to obtain the corresponding energy recovery value. Then, the specific time points of all pulse firings are identified, and the values output by the energy recovery function are extracted at these time points. Finally, the energy recovery values at these time points are accumulated and summed to obtain the final energy recovery threshold. This threshold reflects the total amount of energy recovery that can be achieved at the critical pulse firing moments.

[0102] Based on the pulse firing frequency and the energy recovery function , the efficiency threshold is calculated, and the expression is:

[0103] ;

[0104] Wherein, is the efficiency threshold, represents the efficiency value at time .

[0105] It should be noted that first, the time series data of the pulse firing frequency is collected. The frequency value at each sampling point is input into the efficiency function to calculate the corresponding efficiency value at each moment. Then, the actual firing time points of the pulse signal are detected, and the output results of the efficiency function are extracted at these specific moments to form the efficiency threshold. This efficiency threshold characterizes the overall efficiency level that can be achieved when the pulse event occurs.

[0106] S4. Based on the energy recovery threshold and the efficiency threshold, perform dynamic energy recovery on the servo motor, collect the vibration signal after energy recovery, calculate the maximum Lyapunov exponent, and update the pulse neural network.

[0107] Real-time monitor the real-time energy parameters of the servo motor. When the real-time energy parameters exceed the energy recovery threshold , perform energy recovery, and at the same time adjust the intensity of energy recovery according to the efficiency threshold , and the expression is:

[0108] ;

[0109] Among them, represents the energy recovery power at time ; represents the command value indicating whether to issue a pulse at time ; represents the real-time power of the servo motor, represents the start time point of energy recovery;

[0110] It should be noted that the real-time energy parameters of the servo motor include real-time feedback power and real-time output power.

[0111] During the energy recovery process, the vibration signal of the servo motor is collected by a vibration sensor;

[0112] Based on the vibration signal of the servo motor, the maximum Lyapunov exponent is calculated by phase space reconstruction and small data method, and the expression is:

[0113] ;

[0114] Among them, is the maximum Lyapunov exponent, is the phase space trajectory reconstructed based on the vibration signal at time ; is the phase space trajectory reconstructed based on the vibration signal at time ; is the phase space trajectory reconstructed based on the vibration signal at time ; represents the time interval of adjacent trajectory evolution time.

[0115] It should be noted that first, the data of the vibration signal changing with time is collected. Then, the vibration signal is delayed at a fixed time interval to generate multiple time series, and the time series are combined into a multi-dimensional vector, that is, a trajectory point in the phase space is formed. By repeating this process, a complete phase space trajectory is formed.

[0116] According to the maximum Lyapunov exponent , the pulse neural network is updated.

[0117] It should be noted that, first, the current device operating state is identified according to the largest Lyapunov exponent. When the largest Lyapunov exponent exhibits chaotic characteristics, the learning ability of the spiking neural network is immediately enhanced to accelerate the adjustment process of the neuron connection weights; when the largest Lyapunov exponent shows stable characteristics, the existing learning intensity is maintained. Subsequently, the plasticity regulation mechanism based on spike timing is started to precisely regulate the connection relationships of all neurons, focusing on strengthening the neuron connections with clear timing correlations, and at the same time appropriately weakening the connections with insignificant timing correlations. This regulation process fully integrates real-time vibration characteristics and long-term evolution laws, and through continuous optimization, ensures that the parameters of the spiking neural network are optimally matched with the device operating state. Finally, the update of the spiking neural network is completed.

[0118] This embodiment also provides a servo motor energy-saving control system, including: a data acquisition module, a hot spot marking module, a threshold generation module, and an energy recovery module; the data acquisition module is used to collect the magnetic field intensity data and temperature distribution data of the servo motor and perform preprocessing, and at the same time obtain the magnetic field intensity matrix and temperature distribution matrix; the hot spot marking module is used to obtain the magnetothermal coupling coefficient based on the magnetic field intensity matrix and temperature distribution matrix, identify the temperature anomaly area, and generate a hot spot mark; the threshold generation module is used to encode the magnetothermal coupling coefficient into a pulse sequence through a spiking neural network based on the magnetothermal coupling coefficient and the hot spot mark, and use the synaptic weight update mechanism of the hidden layer neurons to generate an energy recovery threshold and an efficiency threshold; the energy recovery module is used to perform dynamic energy recovery on the servo motor based on the energy recovery threshold and the efficiency threshold, collect the vibration signal after energy recovery, calculate the largest Lyapunov exponent, and update the spiking neural network.

[0119] This embodiment also provides a computer device, applicable to the situation of the servo motor energy-saving control method, 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 servo motor energy-saving control method proposed in the above embodiment.

[0120] 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 can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, a trackball, or a touchpad provided on the housing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0121] This embodiment also provides a storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the servo motor energy-saving control method proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, abbreviated as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, abbreviated as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, abbreviated as EPROM), programmable read-only memory (Programmable Red-Only Memory, abbreviated as PROM), read-only memory (Read-Only Memory, abbreviated as ROM), magnetic memory, flash memory, a magnetic disk, or an optical disc.

[0122] In summary, the present invention realizes the precise correlation mapping between magnetic field distortion and local temperature rise through magnetic-thermal coupling coefficient calculation and hot spot marking, breaks through the limitations of traditional independent analysis, and provides a physical basis for energy recovery; encodes the magnetic-thermal coupling coefficient into a pulse sequence through a pulsed neural network, and uses the synaptic weight update mechanism of hidden layer neurons to improve the real-time response ability of the servo motor to energy recovery and efficiency optimization during operation.

[0123] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than 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 by the scope of the claims of the present invention.

Claims

1. A servo motor energy-saving control method, characterized in that: Including, Collecting the magnetic field intensity data and temperature distribution data of the servo motor and performing preprocessing, and simultaneously obtaining the magnetic field intensity matrix and the temperature distribution matrix; Based on the magnetic field intensity matrix and the temperature distribution matrix, obtaining the magnetothermal coupling coefficient, identifying the temperature anomaly region, and generating a hot spot marker; Based on the magnetothermal coupling coefficient and the hot spot marker, encoding the magnetothermal coupling coefficient into a pulse sequence through a spiking neural network, and using the synaptic weight update mechanism of the hidden layer neurons to generate an energy recovery threshold and an efficiency threshold; Based on the energy recovery threshold and the efficiency threshold, performing dynamic energy recovery on the servo motor, collecting the vibration signal after energy recovery, calculating the maximum Lyapunov exponent, and updating the spiking neural network; The steps of obtaining the magnetothermal coupling coefficient are as follows. Through the collaborative filtering algorithm, extracting the magnetic field intensity eigenvector in the magnetic field intensity matrix and the temperature distribution eigenvector in the temperature distribution matrix; Using the singular value decomposition method to decompose the tensor product of the magnetic field intensity eigenvector and the temperature distribution eigenvector, and combining with a non-linear activation function to obtain the magnetothermal coupling coefficient; The steps of encoding the magnetothermal coupling coefficient into a pulse sequence through a spiking neural network based on the magnetothermal coupling coefficient and the hot spot marker, and using the synaptic weight update mechanism of the hidden layer neurons to generate an energy recovery threshold and an efficiency threshold are as follows. Based on the hot spot marker, encoding the magnetothermal coupling coefficient into a pulse sequence through a pulse encoder; Inputting the pulse sequence into the hidden layer neurons, and each neuron updates the membrane potential of each neuron according to the input pulse sequence; Defining the membrane potential threshold of the neuron based on the historical pulse firing frequency and the membrane potential dynamic range; Calculating the pulse firing frequency based on the updated membrane potential of the neuron; By analyzing the historical energy recovery data, using the linear regression method to fit the relationship between the pulse firing frequency and the energy recovery value, and constructing an energy recovery function using a smoothing factor; By analyzing the historical efficiency data, using the non-linear fitting method to capture the non-linear influence of the pulse firing frequency on the efficiency of the servo motor, and constructing an efficiency function in combination with a saturation factor; Generating an energy recovery threshold and an efficiency threshold according to the pulse firing frequency, the energy recovery function, and the efficiency function.

2. The servo motor energy-saving control method according to claim 1, characterized in that: The preprocessing includes noise filtering, outlier removal, normalization, spatio-temporal alignment, and data standardization.

3. The servo motor energy-saving control method according to claim 1, wherein: The steps of obtaining the magnetic field intensity matrix and the temperature distribution matrix are as follows. Based on the preprocessed magnetic field intensity data, constructing the magnetic field intensity matrix through the grid interpolation method; Based on the preprocessed temperature distribution data, constructing the temperature distribution matrix through the Kriging interpolation method.

4. The servo motor energy-saving control method according to claim 1, characterized in that: The steps of identifying the temperature anomaly region and generating a hot spot marker are as follows. Based on the magnetothermal coupling coefficient, identifying the temperature anomaly region of the servo motor through the local outlier factor method, and calculating the coupling threshold through the fuzzy membership function; Generating a hot spot marker through the fuzzy logic discrimination method according to the temperature anomaly region and the coupling threshold.

5. The servo motor energy-saving control method according to claim 1, characterized in that: The steps of performing dynamic energy recovery on the servo motor based on the energy recovery threshold and the efficiency threshold, collecting the vibration signal after energy recovery, calculating the maximum Lyapunov exponent, and updating the spiking neural network are as follows. Monitor the real-time energy parameters of the servo motor in real time. When the real-time energy parameters exceed the energy recovery threshold, perform energy recovery, and at the same time adjust the intensity of energy recovery according to the efficiency threshold; Based on the vibration signal of the servo motor, calculate the maximum Lyapunov exponent through phase space reconstruction and the small data method; Update the pulse neural network according to the maximum Lyapunov exponent.

6. A servo motor energy-saving control system, based on the servo motor energy-saving control method according to any one of claims 1 to 5, characterized in that: It includes a data acquisition module, a hot spot marking module, a threshold generation module, and an energy recovery module; The data acquisition module is used to collect the magnetic field intensity data and temperature distribution data of the servo motor and perform preprocessing, and at the same time obtain the magnetic field intensity matrix and temperature distribution matrix; The hot spot marking module is used to obtain the magnetothermal coupling coefficient based on the magnetic field intensity matrix and temperature distribution matrix, identify the temperature abnormal area, and generate a hot spot mark; The threshold generation module is used to encode the magnetothermal coupling coefficient into a pulse sequence through a pulse neural network based on the magnetothermal coupling coefficient and the hot spot mark, and use the synaptic weight update mechanism of the hidden layer neurons to generate an energy recovery threshold and an efficiency threshold; The energy recovery module is used to perform dynamic energy recovery on the servo motor based on the energy recovery threshold and the efficiency threshold, collect the vibration signal after energy recovery, calculate the maximum Lyapunov exponent, and update the pulse neural network.

7. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that: When the processor executes the computer program, the steps of the servo motor energy-saving control method according to any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, the steps of the servo motor energy-saving control method according to any one of claims 1 to 5 are implemented.

Images