Servo motor energy-saving control system and method

Abstract

The invention provides an energy-saving control system and method for a servo motor, and the system comprises an energy-saving control middle table, an obtaining unit, a state analysis unit, a difference analysis unit and a regulation and control unit, and the obtaining unit is used for obtaining the current working parameters of the servo motor in operation; the state analysis unit is used for inputting the current working parameters into a state analysis model to obtain a current working load state and optimal control parameters output by the state analysis model; the difference analysis unit is used for performing difference analysis based on the current working load state and the target working load state to obtain a working load difference variable; the target load state is a preset expected load state of the servo motor; and the regulation and control unit is used for carrying out trigger condition analysis based on the working load difference variable and regulating and controlling the servo motor according to a trigger condition analysis result in combination with the optimal control parameter. According to the invention, the energy consumption of the servo motor in the operation process is reduced.

Description

Technical Field

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

[0002] Servo motors play an important role in the field of industrial automation because they can precisely control speed and position accuracy. With the increasing emphasis on energy efficiency, how to optimize the control method of servo motors to achieve energy conservation and emission reduction has become a research hotspot.

[0003] Currently, the traditional PID control algorithm is widely used in servo motor control systems. However, due to its characteristics of continuous calculation and feedback error adjustment, it may cause the servo motor to frequently adjust its speed and torque during operation, thereby increasing energy consumption. For example, when there is a slight change in the servo motor load, the PID controller may over-adjust the output of the servo motor, resulting in the servo motor operating in an unnecessary high-energy consumption state. Summary of the Invention

[0004] The present invention provides a servo motor energy-saving control system and method to solve the problem of energy loss caused by the frequent adjustment of speed and torque of the servo motor during operation, thereby reducing the energy consumption of the servo motor during operation.

[0005] In a first aspect, the present invention provides a servo motor energy-saving control system, including: an energy-saving control center platform, an acquisition unit, a state analysis unit, a difference analysis unit, and a regulation unit. The energy-saving control center platform is respectively connected to the acquisition unit, the state analysis unit, the difference analysis unit, and the regulation unit to manage each unit;

[0006] The acquisition unit is used to acquire the current working parameters during the operation of the servo motor;

[0007] The state analysis unit is used to input the current working parameters into the state analysis model to obtain the current working load state and the optimal control parameters output by the state analysis model; the state analysis model is trained based on sample parameter data and corresponding sample label results; the sample label results are the label results of the current working load state and the optimal control parameters;

[0008] The difference analysis unit is used to perform difference analysis based on the current working load state and the target working load state to obtain a working load difference variable; the target load state is the expected load state preset for the servo motor;

[0009] The regulation unit is used to perform trigger condition analysis based on the working load difference variable and regulate the servo motor according to the trigger condition analysis result in combination with the optimal control parameters.

[0010] In a second aspect, the present invention further provides a servo motor energy-saving control method, which is applied to the servo motor energy-saving control system as described in the first aspect; the servo motor energy-saving control method includes:

[0011] Obtain the current working parameters during the operation of the servo motor;

[0012] Input the current working parameters into the state analysis model to obtain the current working load state and the optimal control parameters output by the state analysis model; the state analysis model is trained based on sample parameter data and corresponding sample label results; the sample label results are the label results of the current working load state and the optimal control parameters;

[0013] Perform difference analysis based on the current working load state and the target working load state to obtain a working load difference variable; the target load state is the expected load state preset for the servo motor;

[0014] Perform trigger condition analysis based on the working load difference variable, and adjust the servo motor according to the trigger condition analysis result in combination with the optimal control parameters.

[0015] In a third aspect, the present invention further provides an electronic device, including: a memory for storing a computer software program; a processor for reading and executing the computer software program, thereby implementing the servo motor energy-saving control method as described in any one of the above.

[0016] In a fourth aspect, the present invention further provides a non-transitory computer-readable storage medium, in which a computer software program is stored, and when the computer software program is executed by a processor, the servo motor energy-saving control method as described in any one of the above is implemented.

[0017] In a fifth aspect, the present invention further provides a computer program product, including a computer program, and when the computer program is executed by a processor, the servo motor energy-saving control method as described in any one of the above is implemented.

[0018] The servo motor energy-saving control system provided by the embodiments of the present invention analyzes the current working parameters through the state analysis model, so it can accurately analyze the current working load state and the optimal control parameters of the servo motor, and further can accurately analyze the working load difference variable between the current working load state and the target working load state. Finally, based on the judgment result obtained by analyzing the working load difference variable and the optimal control parameters, the servo motor can be accurately adjusted, realizing the accurate adjustment plan for triggering the operation of the servo motor as needed, avoiding the problem of energy loss caused by the servo motor frequently adjusting the speed and torque during operation, thereby reducing the energy consumption of the servo motor during operation. Description of the Drawings

[0019] Figure 1 is a schematic structural diagram of the servo motor energy-saving control system provided by an embodiment of the present invention;

[0020] Figure 2 is a schematic flow diagram of the servo motor energy-saving control method provided by an embodiment of the present invention;

[0021] Figure 3 is an embodiment diagram of the electronic device provided by an embodiment of the present invention;

[0022] Figure 4 is an embodiment diagram of the computer-readable storage medium provided by an embodiment of the present invention. Detailed implementation manners

[0023] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative efforts shall fall within the protection scope of the present invention.

[0024] In the description of the present invention, the terms "first" and "second" are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of the described features. In the description of the present invention, "a plurality of" means two or more, unless otherwise specifically defined.

[0025] In the description of the present invention, the term "for example" is used to mean "serving as an example, illustration, or explanation". Any embodiment described as "for example" in the present invention is not necessarily construed as being more preferred or having more advantages than other embodiments. In order for any person skilled in the art to implement and use the present invention, the following description is given. In the following description, details are set forth for purposes of explanation. It should be understood that those skilled in the art can recognize that the present invention can be implemented without using these specific details. In other instances, well-known structures and processes are not elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is to be accorded the widest scope consistent with the principles and features disclosed herein.

[0026] Optionally, refer to Figure 1 as shown Figure 1It is a schematic structural diagram of the servo motor energy-saving control system provided by the present invention. The servo motor energy-saving control system includes: an energy-saving control center platform, an acquisition unit, a state analysis unit, a difference analysis unit, and a regulation unit. The energy-saving control center platform is respectively connected to the acquisition unit, the state analysis unit, the difference analysis unit, and the regulation unit to manage each unit.

[0027] The energy-saving control center platform and its associated multiple important units play a key role. The energy-saving control center platform is like the intelligent brain of the entire system, comprehensively and efficiently managing the acquisition unit, the state analysis unit, the difference analysis unit, and the regulation unit to ensure the coordinated operation of each unit and achieve the optimal utilization of energy and high-efficiency energy conservation.

[0028] Optionally, the acquisition unit is used to acquire the current working parameters during the operation of the servo motor. A distributed sensor network is used to acquire the current working parameters during the operation of the servo motor in real time, including parameters such as torque, speed, current, voltage, temperature, and vibration frequency, etc. And to ensure the accuracy and real-time performance during the data transmission process, high-precision sensors are used in the distributed sensor network, which are respectively installed at key parts of the servo motor. For example, a strain torque sensor (such as HBMT408) is installed at the shaft end of the servo motor, a thermocouple is placed inside the stator winding for temperature measurement, and an encoder (such as Heidenhain ERN1387) is used to synchronously collect the speed. A current sensor is installed on the power supply line of the servo motor, etc., to achieve the synchronous acquisition of the real-time operation parameters of the servo motor.

[0029] Optionally, the state analysis unit is used to input the current working parameters into the state analysis model to obtain the current working load state and the optimal control parameters output by the state analysis model; the state analysis model is trained based on sample parameter data and corresponding sample label results; the sample label results are the label results of the current working load state and the optimal control parameters; A state analysis model is trained and constructed using a deep neural network (such as LSTM, CNN, or a hybrid model of LSTM+CNN). After the model training is completed, the current working parameters of a preset sliding window (such as 100ms) are input, and the output is the current working load state (such as light load, no load, medium load, and heavy load, etc.) and the optimal control parameters (i.e., corresponding to different parameter data in the current working parameters).

[0030] Furthermore, the state analysis model trained by the energy-saving control system is obtained by training based on a large number of sample parameter data and corresponding sample label results. Among them, the sample parameter data are the operating parameters of the servo motor collected under different working conditions, and the sample label results are the corresponding current working load state and the optimal control parameters. During the training process, the model will learn the mapping relationship between the sample parameter data and the sample label results, and adjust the parameters of the model through optimization algorithms (such as the gradient descent method) to enable it to accurately classify and predict the input data, so that it can output the corresponding current working load state and the optimal control parameters according to the input current working parameters. It is realized that the current working load state of the servo motor can be accurately identified through the state analysis model, and the optimal control parameters are given, providing a data basis for subsequent regulation, so as to achieve the purpose of energy saving.

[0031] Optionally, a difference analysis unit is configured to perform difference analysis based on the current working load state and the target working load state to obtain a working load difference variable; the target load state is the expected load state preset for the servo motor; for example, calculate the difference in load level, load change rate, etc. between the current working load state and the target working load state, so as to obtain the working load difference variable, specifically as described in steps 301 - 304. For example, on an automated production line, the target working load state of the servo motor is "medium load", while the current working load state obtained through the state analysis model is "heavy load". Through the set quantization standard, the difference between "heavy load" and "medium load" is quantified into a working load difference variable. For example, the difference variable is 20, indicating that the current working load is 20% higher than the target working load, so that the working load difference variable can intuitively reflect the gap between the current working load of the servo motor and the target working load, providing a clear direction for subsequent regulation, and helping to optimize the working load of the servo motor, thus achieving the effect of energy saving.

[0032] Optionally, a control unit is configured to perform trigger condition analysis based on the workload difference variable, and regulate the servo motor according to the trigger condition analysis result in combination with the optimal control parameters. Trigger conditions are preset (such as setting a first preset threshold and a second preset threshold). When the workload difference variable meets the corresponding conditions, the corresponding control strategy is triggered, specifically as described in steps 401 - 403. Taking the trigger threshold as 10% as an example, when the workload difference variable is 20%, which is greater than the trigger threshold, the regulation mechanism is triggered. According to the optimal control parameters, the rotational speed of the servo motor is adjusted from the current 1800 r / min to the optimal rotational speed of 1500 r / min, and at the same time, the torque of the servo motor is adjusted to make the workload of the servo motor gradually approach the target workload state. By realizing regulation through trigger condition analysis and combination with optimal control parameters, it is possible to timely adjust the operating parameters of the servo motor according to the actual workload situation on the premise of ensuring the normal operation of the servo motor, avoid the servo motor running under unnecessary high loads, and thus achieve the purpose of energy saving.

[0033] In the embodiment of the present invention, the current working parameters are analyzed through a state analysis model. Therefore, the current workload state and the optimal control parameters of the servo motor can be accurately analyzed. Further, the workload difference variable between the current workload state and the target workload state can be accurately analyzed. Finally, based on the judgment result obtained from the analysis of the workload difference variable and in combination with the optimal control parameters, the servo motor can be precisely regulated, realizing a precise regulation plan for triggering the operation of the servo motor as needed, avoiding the problem of energy loss caused by the frequent adjustment of the rotational speed and torque of the servo motor during operation, and thus reducing the energy consumption of the servo motor during operation.

[0034] Optionally, referring to Figure 2 , Figure 2 is a schematic flowchart of the servo motor energy-saving control method provided by the present invention. In the embodiment of the present invention, the execution subject of the servo motor energy-saving control is an energy-saving control system. Therefore, the energy-saving control method based on the servo motor includes:

[0035] Step 10, obtaining the current working parameters during the operation of the servo motor.

[0036] Optionally, the energy-saving control system uses a distributed sensor network to obtain the current operating parameters of the servo motor in real time, including parameters such as torque, speed, current, voltage, temperature, and vibration frequency. And to ensure the accuracy and real-time performance during data transmission, high-precision sensors are used in the distributed sensor network, which are respectively installed at key parts of the servo motor. For example, a strain torque sensor (such as HBMT408) is installed at the shaft end of the servo motor, a thermocouple is placed in the stator winding for temperature measurement, and an encoder (such as Heidenhain ERN1387) is used to synchronously collect the speed. A current sensor is installed on the power supply line of the servo motor, etc., to achieve synchronous acquisition of the real-time operating parameters of the servo motor.

[0037] Step 20: Input the current operating parameters into the state analysis model to obtain the current workload state and the optimal control parameters output by the state analysis model; the state analysis model is trained based on sample parameter data and corresponding sample label results; the sample label results are the label results of the current workload state and the optimal control parameters.

[0038] Optionally, the energy-saving control system uses a deep neural network (such as an LSTM, CNN, or a hybrid model of LSTM+CNN) to train and construct the state analysis model. After the model training is completed, input the current operating parameters of a preset sliding window (such as 100 ms), and the output is the current workload state (such as light load, no load, medium load, and heavy load, etc.) and the optimal control parameters (i.e., corresponding to different parameter data in the current operating parameters).

[0039] Furthermore, the state analysis model trained by the energy-saving control system is trained based on a large amount of sample parameter data and corresponding sample label results. Among them, the sample parameter data are the operating parameters of the servo motor collected under different working conditions, and the sample label results are the corresponding current workload state and the optimal control parameters. During the training process, the model will learn the mapping relationship between the sample parameter data and the sample label results, and adjust the parameters of the model through an optimization algorithm (such as the gradient descent method) to enable it to accurately classify and predict the input data, so that it can output the corresponding current workload state and the optimal control parameters according to the input current operating parameters. It is realized that the current workload state of the servo motor can be accurately identified through the state analysis model, and the optimal control parameters are given, providing a data basis for subsequent regulation, so as to achieve the purpose of energy saving.

[0040] Further, taking a neural network model as an example, during the model training process, 1000 sets of servo motor operating parameters under different working conditions are collected as sample parameter data, including current, voltage, speed, etc. At the same time, the current working load state (such as light load, medium load, heavy load) and the optimal control parameters (such as optimal speed, optimal torque, etc.) corresponding to each set of data are marked as sample label results. These data are divided into a training set and a test set, and the training set is used to train the neural network model, adjusting the weights and biases of the model until the model achieves good performance on the test set. When the current operating parameters are obtained, they are input into the trained neural network model, and the model outputs that the current working load state is "medium load", and the optimal control parameter is the optimal speed of 1500 r / min.

[0041] Step 30, based on the difference analysis between the current working load state and the target working load state, obtain the working load difference variable; the target load state is the expected load state preset for the servo motor.

[0042] Optionally, the energy-saving control system compares the current working load state obtained in step 20 with the preset target working load state, and defines some quantization indexes to measure the difference between the two, such as calculating the difference in load degree between the current working load state and the target working load state, the load change rate, etc., so as to obtain the working load difference variable, specifically as described in steps 301 - 304. For example, on an automated production line, the target working load state of the servo motor is "medium load", while the current working load state obtained through the state analysis model is "heavy load". Through the set quantization standard, the difference between "heavy load" and "medium load" is quantified as a working load difference variable. For example, the difference variable is 20, indicating that the current working load is 20% higher than the target working load, so that the working load difference variable can intuitively reflect the gap between the current servo motor working load and the target working load, provide a clear direction for subsequent regulation, help to optimize the servo motor working load, and thus achieve the energy-saving effect.

[0043] Step 40, based on the working load difference variable, conduct trigger condition analysis, and regulate the servo motor according to the trigger condition analysis result in combination with the optimal control parameter.

[0044] Optionally, the energy-saving control system pre-sets triggering conditions (such as setting a first preset threshold and a second preset threshold). When the workload difference variable meets the corresponding conditions, the corresponding control strategy is triggered, specifically as described in steps 401 - 403. Taking the triggering threshold as 10% as an example, when the workload difference variable is 20%, which is greater than the triggering threshold, the regulation mechanism is triggered. According to the optimal control parameters, the rotational speed of the servo motor is adjusted from the current 1800 r / min to the optimal rotational speed of 1500 r / min, and at the same time, the torque of the servo motor is adjusted to make the workload of the servo motor gradually approach the target workload state. By realizing regulation through triggering condition analysis and combining optimal control parameters, it is possible to adjust the operating parameters of the servo motor in a timely manner according to the actual workload situation on the premise of ensuring the normal operation of the servo motor, avoiding the servo motor from operating under unnecessary high loads, thereby achieving the purpose of energy saving.

[0045] Furthermore, during the process of setting the first preset threshold and the second preset threshold in the energy-saving control system, first determine the rated load of the servo motor and select based on the range of the rated load. For example, the first preset threshold is set around 30% - 50% of the rated load, and the second preset threshold may be around 70% - 90% of the rated load. At the same time, if the response speed of the servo motor is relatively fast, when setting the first preset threshold and the second preset threshold, they can also be made close to the load fluctuation range during actual operation for more timely regulation. It should be noted that the influence of working environment factors on the first preset threshold and the second preset threshold should also be considered. For example, in a high-temperature environment, the heat dissipation effect of the servo motor will deteriorate, and its performance may be affected. At this time, the preset threshold needs to be reduced to prevent the servo motor from overheating and being damaged. For example, when the ambient temperature is 40°C, compared with the normal temperature of 25°C, the first preset threshold and the second preset threshold may both be reduced by 10% - 20%. Further, if the servo motor has multiple operating modes, such as high-speed operation mode, low-speed operation mode, heavy-load mode, light-load mode, etc., set the preset threshold respectively according to the load characteristics in different modes. For example, in the high-speed operation mode, the servo motor may be more easily affected by dynamic loads, and the preset threshold may need to be adjusted according to the calculation results of dynamic loads; in the heavy-load mode, the second preset threshold can be appropriately increased to consider the long-term operation reliability of the servo motor. Therefore, in summary, when setting the first preset threshold and the second preset threshold, the comprehensive influence of rated load, response speed, environmental factors, and equipment operation model should be considered for setting to ensure that during the on-demand response process, the servo motor can be responded to and controlled in a timely manner, while avoiding the problem of energy loss caused by frequent adjustment of rotational speed and torque.

[0046] In the embodiment of the present invention, the current working parameters are analyzed through a state analysis model, so that the current working load state and the optimal control parameters of the servo motor can be accurately analyzed. Further, the working load difference variable between the current working load state and the target working load state can be accurately analyzed. Finally, based on the judgment result obtained by analyzing the working load difference variable and the optimal control parameters, the servo motor can be accurately regulated, realizing the accurate regulation plan for triggering the operation of the servo motor on demand, avoiding the problem of energy loss caused by the frequent adjustment of the rotation speed and torque of the servo motor during operation, and thus reducing the energy consumption of the servo motor during operation.

[0047] In one embodiment, the descriptions of steps 301 - 304 are as follows:

[0048] Step 301: Align the working parameters in the current working load state and the target working load state in time series to obtain the corresponding sequence of working parameters.

[0049] Optionally, after the energy-saving control system obtains the current working load state and the target working load state, in the actual operation scenario of the servo motor, the working parameters of the current working load state and the target working load state may be sampled at different time points, resulting in a time offset. Therefore, time series alignment is required (to synchronize them in the time dimension for subsequent analysis). For example, the dynamic time warping (DTW) algorithm is used. By finding the optimal matching path between two time series, the cumulative distance between the two series is minimized, and then the corresponding sequence of working parameters is obtained.

[0050] In one embodiment, let the sequence of working parameters of the current working load state be X 0 ={x 1 , x 2 , …, x m}, and the sequence of working parameters of the target working load state be Y 0 ={y 1 , y 2 , …, y n}. Construct an m*n cost matrix D, where D(i,j) represents the distance (such as Euclidean distance) between x i and y j . Then define a cumulative cost matrix C, and C(i,j) represents the cumulative cost of the optimal matching path from the starting point of sequence X 0 to x i and from the starting point of sequence Y 0 to y j . Then the recurrence formula for C(i,j) is: C(i,j) = d 0 (x i , y j) + min{C(i - 1, j), C(i, j - 1), C(i - 1, j - 1)}. After that, by backtracking the cumulative cost matrix C, the optimal matching path is found, thus realizing the time series alignment, and finally obtaining the corresponding sequence of working parameters. To enable subsequent analysis based on data with the same time scale, avoid analysis errors caused by time asynchrony, and make the analysis results more accurately reflect the actual workload differences.

[0051] Step 302: Perform grey relational analysis on the corresponding sequence of working parameters to obtain the grey relational degree.

[0052] Optionally, the energy-saving control system performs grey relational analysis on the corresponding sequence of working parameters to obtain the grey relational degree, that is, the energy-saving control system measures the similarity between the current workload state in the corresponding sequence of working parameters and the working parameter sequence of the target workload state. During the calculation of the grey relational degree, by mining the potential non-linear relationships in the data, using the concept of polynomial feature mapping, and after the mapping, fully considering the distribution and mutual relationships of the features in the high-dimensional space, further evaluate the relational degree between working parameters, and finally comprehensively consider the importance of different-dimensional features to obtain the final grey relational degree R i , as specifically described in Steps 3021 - 3022.

[0053] Step 303: Perform transition probability analysis based on the corresponding sequence of working parameters to obtain the state transition probability matrix.

[0054] Optionally, the energy-saving control system divides the states of the corresponding sequence of working parameters. For example, the parameter values are divided into multiple intervals, and each interval corresponds to a state. Let the state set be After that, count the number of times of transitioning from one state to another in the corresponding sequence of working parameters. Let N ij represent the number of times of transitioning from state s i to state s j . The state transition probability where represents the number of states in the state set S z . Finally, all state transitions P ij constitute the state transition probability matrix to reflect the dynamic change law of the workload state through the state transition probability matrix, and understand the transfer possibility of the workload between different states by analyzing the elements in the matrix.

[0055] Step 304: Construct a workload evaluation function based on the grey relational degree and the state transition probability matrix, and determine the workload difference variable based on the workload evaluation function.

[0056] Optionally, in the process of constructing the workload evaluation function of the energy-saving control system, the analysis of sequence similarity measurement by combining grey relational analysis and the analysis of the state transition law by Markov chain are integrated, where the workload evaluation function where γ and δ are weight coefficients, and γ + δ = 1, which are used to balance the proportion of grey relational degree and state transition probability difference in the comprehensive evaluation; P XX represents the state transition probability matrix corresponding to the current working parameter sequence; P YY represents the state transition probability matrix corresponding to the target working parameter sequence; D s represents the workload difference variable.

[0057] The embodiment of the present invention integrates multiple methods such as time series alignment, grey relational analysis, and transition probability analysis, and analyzes the differences between the current workload state and the target workload state from multiple perspectives, which can more comprehensively and accurately reflect the differences in the workload.

[0058] In one embodiment, the descriptions of steps 3021 - 3023 are as follows:

[0059] Step 3021, perform polynomial feature mapping based on the sequence corresponding to the working parameter to obtain a feature space. Optionally, in the process of performing polynomial feature mapping on the sequence corresponding to the working parameter by the energy-saving control system, first set the sequence corresponding to the working parameter as X X and Y Y , and then, in order to capture the non-linear relationship in the sequence, map them to the τ-dimensional polynomial space respectively. Specifically, for the sequence X X , by calculating the powers of X X X X X X 2 , …, X X τ a new feature vector is formed Similarly, for the sequence Y Y , a new feature vector is obtained where τ is a preset mapping dimension, which can be adjusted according to data characteristics and analysis requirements. Finally, the hidden non-linear information in the sequence is mined.

[0060] Step 3022, perform spatial correlation analysis based on the feature space to obtain a correlation matrix.

[0061] Optionally, after the energy-saving control system obtains the feature space and , in the process of calculating the correlation matrix R (τ) . For the element in the matrix , first calculate That is, the absolute value of the difference of the corresponding elements in the feature space is obtained, and then according to the formula the correlation degree is calculated, where minΔ (τ) and maxΔ (τ) respectively represent the minimum value and the maximum value among all , ρ represents the discrimination coefficient (generally taken as 0.5), which is used to adjust the sensitivity of the correlation degree, and finally the correlation degree matrix R (τ) is obtained. Finally, by analyzing the sequence correlation degree in the high-dimensional feature space, the multi-dimensional information after the sequence is mapped by the polynomial is considered, which can more accurately reflect the complex correlation between sequences than analyzing in the original space and can capture the subtle connections that cannot be found by linear analysis.

[0062] Step 3023, based on the correlation degree matrix and the preset dimension weights, perform non-linear correlation synthesis to obtain the grey correlation degree.

[0063] Optionally, after the energy-saving control system obtains the correlation degree matrix, the formula for the preset dimension weights is τ 0 represents the optimal dimension, c represents a constant, which is used to adjust the rate of weight change. Then, by synthesizing the correlation degrees of each dimension, the non-linear correlation degree synthesis is performed through the formula to finally obtain the grey correlation degree R i , where tr(R (τ) ) represents the trace of the correlation degree matrix R (τ) , that is, the sum of the main diagonal elements, and DD represents the total number of dimensions. Considering the importance differences of different dimensions in the correlation analysis, the correlation degrees of each dimension are weighted and summed through the dimension weights. The non-linear weight function can flexibly adjust the weights according to the distance between the dimension and the optimal dimension, making the final grey correlation degree more reasonably synthesize the information of each dimension and improving the accuracy and reliability of the analysis results.

[0064] Through polynomial feature mapping and high-dimensional space correlation analysis, the embodiments of the present invention can mine the non-linear connections in the sequences corresponding to the working parameters, and are more in-depth and comprehensive in analyzing the sequence relationship than the traditional methods; and by introducing dimension weights in the correlation synthesis process and dynamically adjusting the weights according to the relationship between the dimension and the optimal dimension, the calculation of the grey correlation degree is made more scientific and reasonable, and can better reflect the true correlation degree of the sequences.

[0065] In one embodiment, the descriptions of steps 401 - 403 are as follows:

[0066] Step 401, if the workload difference variable is less than the first preset threshold, and it is determined that the preset conditions are met, the servo motor is regulated based on the first adjustment strategy in combination with the optimal control parameters; the preset conditions are: whether the adjacent workload difference variables are continuous and whether the total value of the cumulative workload difference variables is greater than or equal to the first preset threshold.

[0067] Optionally, when setting the first preset threshold and the second preset threshold, the energy-saving control system first adjusts the magnitudes of the first preset threshold and the second preset threshold. Taking the second preset threshold being greater than or equal to the first preset threshold as an example, if the first preset threshold is 5% and the second preset threshold is 15%, when the workload difference variable is less than the first preset threshold, it indicates that the change in the workload difference variable is relatively small compared to the overall control. At this time, when the workload difference variable changes relatively little, the continuity of the workload difference variable is judged. Taking a time window as an example, the workload difference variables of k consecutive sampling periods (such as k = 5) are selected, and it is judged whether the absolute value of the difference between adjacent difference variables is less than the fluctuation threshold ∈ (such as ∈ = 0.1), which can be adjusted according to the actual fluctuation situation or determined based on historical fluctuation values. If it is satisfied, it is considered that the adjacent workload difference variables are continuous, indicating that the change in the workload shows a relatively stable trend, and the stable trend also indicates that the load change is not caused by accidental interference or noise, but may represent a certain potential and continuous load change situation. For example, on an automated production line, as the product production process progresses step by step, the load of the servo motor may increase continuously and slowly, and at this time, the adjacent workload difference variables will be in a continuous state. If it is not satisfied, it is considered that the adjacent workload difference variables are not continuous, indicating that the change in the workload is caused by accidental factors, such as momentary external interference, measurement error, etc. This discontinuous change usually does not represent the true long-term change trend of the load and may only be temporary and random fluctuations. For example, during the operation of the servo motor, it may be affected by the electromagnetic interference of nearby equipment, resulting in an abnormal workload difference variable at a certain moment, but this abnormality will not continuously affect the normal operation of the servo motor. Therefore, when it is not satisfied, its fluctuation is ignored and no subsequent adjustment is required.

[0068] Furthermore, when the energy-saving control system determines that the adjacent workload difference variables are continuous, it continues to calculate the total value of the cumulative workload difference variables, which is the sum of the absolute values of the workload difference variables within k consecutive cycles. where ΔZ iIt is expressed as the workload difference variable in the i-th cycle. After obtaining the total value of the cumulative workload difference variable, the total value of the cumulative workload difference variable is judged again with the first preset threshold. When the total value of the cumulative workload difference variable is greater than or equal to the first preset threshold, it indicates that the workload has exceeded the range tolerated by the system during the steady increase process, and it is necessary to take control measures in time, that is, based on the first adjustment strategy and the optimal control parameters, the servo motor is regulated to make the operating state of the servo motor return to the optimal state as soon as possible, ensuring the stability and energy-saving effect of the system;

[0069] Furthermore, since the workload is steadily increasing and the increase amplitude is small each time, when it exceeds the tolerance range, the servo motor should be regulated steadily following the steady increase of the workload to avoid problems such as large torque friction of the servo motor caused by too large regulation amplitude. The specific regulation process is as described in steps 4011 - 4015.

[0070] Step 402, if the workload difference variable is greater than or equal to the first preset threshold and less than or equal to the second preset threshold, the servo motor is adjusted based on the second adjustment strategy combined with the optimal control parameters.

[0071] Optionally, when the energy-saving control system receives that the workload difference variable is greater than or equal to the first preset threshold and less than or equal to the second preset threshold, it indicates that there is a certain difference between the current workload state of the servo motor and the target workload state, but the difference is not particularly large, belonging to a medium-level load change. In this case, the energy-saving control system first conducts a correlation analysis on the optimal control parameters and the current working parameters. At the same time, a dynamic characteristic model of the servo motor is constructed based on physical principles to predict the performance change trend of the servo motor when adjusting the control parameters. According to the performance change trend and the comprehensive adjustment parameters, the adjustment order of different control parameters is determined, and then the servo motor is adjusted in order and amplitude. In this way, on the premise of ensuring the stable operation of the servo motor, the servo motor can be gradually adapted to the load change, as specifically described in steps 4021 - 4025.

[0072] Step 403, if the workload difference variable is greater than the second preset threshold, the servo motor is adjusted based on the third adjustment strategy combined with the optimal control parameters.

[0073] Optionally, when the workload difference variable received by the energy-saving control system is greater than the second preset threshold, it indicates that the current workload state of the servo motor is significantly different from the target load state, and there is a large change in the load. This may be caused by reasons such as a sudden increase in production tasks or equipment failures. Therefore, global optimization adjustment is performed based on the third adjustment strategy combined with the optimal control parameters. During the global optimization adjustment process, actual constraint factors such as the temperature rise constraint and voltage range of the servo motor, as well as environmental impact factors, are considered to quickly enable the servo motor to adapt to the significantly changed load and ensure the efficient and stable operation of the servo motor, as specifically described in steps 4031 - 4036.

[0074] In the embodiment of the present invention, different adjustment strategies are adopted according to the magnitude of the workload difference variable, making the regulation more accurate and flexible. It can not only avoid frequent adjustments but also quickly respond to large load changes. In addition, during the adjustment process, on the premise of ensuring the stable operation of the servo motor, the energy consumption of the servo motor is reduced through a reasonable adjustment strategy, improving the comprehensive performance of the servo motor. Finally, it can adapt to different degrees of load changes. Whether it is a small fluctuation or a large mutation, corresponding regulation measures can be taken for timely and rapid adjustment, improving the reliability and stability of the system.

[0075] In one embodiment, the descriptions of steps 4011 - 4015 are as follows:

[0076] Step 4011: Perform a difference processing on the optimal control parameters and the current working parameters to obtain a multi-dimensional adjustment vector.

[0077] Optionally, after the energy-saving control system receives the optimal control parameters and the current working parameters, let the optimal control parameter vector be The current working parameter vector be Then the multi-dimensional adjustment vector For example, when controlling the servo motor's voltage V, frequency f, and torque T, if Then The multi-dimensional adjustment vector comprehensively reflects the differences between the optimal control parameters and the current working parameters in each dimension, providing a clear direction and quantitative basis for subsequent precise adjustment of the control parameters.

[0078] Step 4012: Determine the load fluctuation amplitude based on the workload difference variable within a preset time period.

[0079] Optionally, for n workload difference variables ΔZ within a preset time period by the energy-saving control system 1 , ΔZ 2 , …, ΔZ n , then for the load fluctuation amplitude Where Denoted as the mean of the workload difference variable within a preset time period, ΔZ i Denoted as the workload difference variable in the i-th cycle. Since the workload difference variable may have different values at different times, considering only a single value cannot comprehensively understand the fluctuation characteristics of the load. Therefore, calculating the load fluctuation amplitude through the standard deviation can comprehensively consider the dispersion degree of the workload difference variable within the preset time period and accurately reflect the load fluctuation situation.

[0080] Step 4013: Determine the reference range of the preset adjustment step size based on the magnitude of the load fluctuation amplitude.

[0081] Optionally, after the energy-saving control system obtains the load fluctuation amplitude, it pre-sets the mapping relationship between the load fluctuation amplitude and the reference range of the preset adjustment step size. For example, when the load fluctuation amplitude A f is in 0 - A 1 (such as A 1 = 1), the reference range of the preset adjustment step size is [α 1 , α 2 (such as α 1 = 0.05, α 2 = 0.1); when A f is in A 1 - A 2 (such as A 2 = 3), the reference range of the preset adjustment step size is [α 3 , α 4 (such as α 3 = 0.1, α 4 = 0.2), etc. Furthermore, by establishing a piecewise function, let B(A f ) be the reference range of the preset adjustment step size, then B(A f ) = Dynamically determining the reference range of the preset adjustment step size according to the load fluctuation amplitude can make the adjustment step size match the severity of the load change. When the load fluctuates greatly, appropriately increase the adjustment step size to accelerate the adjustment speed to adapt to the load change; when the load fluctuates slightly, reduce the adjustment step size to improve the adjustment accuracy and avoid over-adjustment.

[0082] Step 4014: Correct the reference range based on the current workload difference variable to obtain the target adjustment step size.

[0083] Optionally, after the energy-saving control system determines the range, it obtains the current workload difference variable ΔZ cur , and the reference range of the preset adjustment step size is [α min , α max . At this time, the correction can be performed within the reference range by the method of linear interpolation to obtain the target adjustment step size. For example, when ΔZcur When it is at the lower value of the load fluctuation amplitude range, the target adjustment step size approaches α min ; when ΔZ cur is at a higher value, the target adjustment step size approaches α max . Let the load fluctuation amplitude range be [A min , A max , then the target adjustment step size Thus, by combining the current workload difference variable, the reference range is corrected, further refining the determination process of the adjustment step size, making the adjustment step size more in line with the current actual load situation, and improving the accuracy of the adjustment.

[0084] Step 4015, based on the multi-dimensional adjustment vector and the target adjustment step size, gradually update the control parameters.

[0085] Optionally, after the energy-saving control system determines the target adjustment step size α, first set the number of control parameter updates to L, and initialize the control parameter vector as At the Lth update, the control parameter vector After each update, judge whether the satisfactory operating state is reached according to the feedback information of the servo motor. If not, continue the next update. Finally, by gradually updating the control parameters, the impact on the operation of the servo motor caused by the sudden change of the parameters is avoided, and the stability of the operation of the servo motor is ensured. At the same time, combined with the multi-dimensional adjustment vector and the target adjustment step size, precise adjustment can be carried out in the direction of the optimal control parameters on the premise of stable operation.

[0086] In the embodiment of the present invention, the adjustment direction and magnitude of each parameter are clarified through the multi-dimensional adjustment vector, and the target adjustment step size is determined by combining the load fluctuation amplitude and the current workload difference variable, realizing the precise adjustment of the control parameters, enabling the servo motor to better adapt to the change of the workload; and dynamically determining the preset adjustment step size reference range according to the load fluctuation situation, and correcting it by combining the current workload difference variable, making the adjustment strategy have good self-adaptability and being able to cope with load changes of different degrees and characteristics.

[0087] In one embodiment, the descriptions of steps 4021 - 4025 are as follows:

[0088] Step 4021, perform a correlation analysis on the optimal control parameters and the current working parameters to obtain the multi-parameter collaborative adjustment parameters.

[0089] Optionally, after the energy-saving control system judges that the workload difference variable is greater than or equal to the first preset threshold and less than or equal to the second preset threshold, use the Pearson correlation coefficient to analyze the optimal control parameters and the current working parameters The correlation between each dimension. The Pearson correlation coefficient formula is where both i and j represent parameter dimensions, and h represents the number of samples. and respectively represent the means of the i-th dimension of the optimal control parameter and the j-th dimension of the current working parameter. And through this coefficient, an n*n correlation coefficient matrix E can be obtained e =(r ij ). Then, based on the correlation coefficient matrix E e calculate the multi-parameter collaborative adjustment parameter

[0090] In one embodiment, taking three parameters of voltage V, frequency f, and torque T as an example, if r Vf represents the correlation coefficient between voltage and frequency, r VT represents the correlation coefficient between voltage and torque, etc., it can be calculated by a weighted method such as then V syn =V opt -C cur +ω 1 r Vf (f opt -f cur )+ω 2 r VT (T opt -T cur ), f syn =f opt -V cur +ω 3 r fV (V opt -V cur )+ω 4 r fT (T opt -T cur ), T syn =T opt -T cur +ω 5 r TV (V opt -V cur )+ω 6 r Tf (f opt -f cur ), where ω 1 -ω 6 represents the weight coefficient, which is determined according to the actual situation. For example, ω 1 =ω 2 =ω 3 =ω 4 =ω 5 =ω 6= 0.1, and finally obtain By considering the mutual influence among different control parameters, the multi-parameter collaborative adjustment parameters obtained can more comprehensively reflect the parameter adjustment requirements, avoid adjusting a single parameter alone and ignoring its impact on other parameters, thereby achieving more efficient and accurate control.

[0091] Step 4022: Construct a dynamic model of the servo motor based on physical principles, and predict the performance change trend of the servo motor caused by parameter adjustment based on the current working load state and the dynamic model of the servo motor.

[0092] Optionally, the energy-saving control system constructs a dynamic model of the servo motor based on physical principles. Taking the DC servo motor as an example, its voltage balance equation is The electromagnetic torque equation is T = K t I α , and the motion equation is where U represents the armature voltage, R α represents the armature resistance, L a represents the armature inductance, I α represents the armature current, E b represents the back electromotive force, K t represents the torque constant, J represents the moment of inertia, w represents the rotational speed, T represents the electromagnetic torque, T L represents the load torque, B w represents the viscous friction coefficient. After that, the energy-saving control system, according to the current working load state, such as knowing the current load torque T Lcur , rotational speed w cur , etc., substitutes them into the dynamic model. Assuming that the voltage U needs to be adjusted, by solving the dynamic model (such as using Laplace transform to convert the time-domain equation into a frequency-domain equation for analysis), the performance change trends of performance indicators such as the armature current I α , electromagnetic torque T, and rotational speed w after the voltage adjustment can be predicted. The dynamic model of the servo motor based on physical principles can accurately reflect the physical process inside the servo motor, and combine the current working load state to predict the performance change trend after parameter adjustment, providing a basis for determining the adjustment sequence and amplitude subsequently, making the adjustment more forward-looking and reasonable.

[0093] Step 4023: Determine the adjustment sequence of different control parameters based on the multi-parameter collaborative adjustment parameters and the influence amplitude of the performance change trend on the servo motor performance.

[0094] Optionally, the energy-saving control system analyzes the influence degree of each parameter in the multi-parameter collaborative adjustment parameters on the servo motor performance. For example, through sensitivity analysis, determine the influence coefficients S Vw , S VT , SVE , the influence coefficient S of frequency adjustment on these performances fw , S fT , S fE , etc. Combining with the performance change trend, if the rotational speed fluctuates greatly under the current workload state, and it is found according to the sensitivity analysis that the influence coefficient |S Vw | of voltage on rotational speed is large, and in the prediction of the performance change trend, voltage adjustment can effectively improve the rotational speed fluctuation, then the voltage parameter is preferentially adjusted.

[0095] Further, in an embodiment, the energy-saving control system can construct a comprehensive evaluation function to determine the adjustment order. Let the comprehensive evaluation function where θ represents the control parameter (such as voltage V, frequency f, torque T), represents the performance index (such as rotational speed w, torque T, energy consumption E), represents the weight coefficient of the performance index, which is determined according to actual needs. For example, in a scenario with high requirements for rotational speed stability, α 1 (corresponding to the weight of rotational speed) can be set to 0.5, α 2 = α 3 = 0.25. By calculating F V , F f , F T , determine the adjustment order in descending order. Determining the adjustment order according to the influence amplitude can preferentially adjust the parameters that are most effective in improving the performance of the servo motor under the current workload state, improve the adjustment efficiency, and make the servo motor reach a good operating state faster.

[0096] Step 4024, determine the adjustment amplitude based on the magnitude and direction of the workload difference variable.

[0097] Optionally, the energy-saving control system sets the workload difference variable as ΔZ. When ΔZ > 0, it indicates that the current workload is greater than the target workload; when ΔZ < 0, it indicates that the current workload is less than the target workload. The adjustment amplitude β is determined according to the absolute value of ΔZ. For example, establish a mapping relationship. When |ΔZ| is within 0 - Z 1 (such as Z 1 = 5), β = β 0 (such as β 0 = 0.1); when |ΔZ| is within Z 1 - Z 2 (such as Z 2 = 10), β = β 0 + q 1 (|ΔZ| - Z 1 ) (such as q 1= 0.05), etc. That is, the adjustment range is positively correlated with the absolute value of the workload difference variable. The greater the difference, the greater the adjustment range. It is possible to determine the adjustment range according to the magnitude and direction of the workload difference variable, closely combine the adjustment range with the load deviation, and enable the workload of the servo motor to approach the target load more quickly, improving the response speed and control accuracy of the system.

[0098] Step 4025: Adjust the parameters of the servo motor based on the adjustment sequence and the adjustment range.

[0099] Optionally, the energy-saving control system adjusts the control parameters in sequence according to the determined adjustment sequence. For example, if the adjustment sequence is voltage, frequency, and torque, first adjust the voltage according to the adjustment range β. Let the current voltage be V cur , the adjusted voltage V new = V cur + βV syn (V syn is the collaborative adjustment parameter of the voltage). After adjusting the voltage, according to the feedback information of the servo motor, such as speed, torque, etc., combined with the dynamic model, evaluate the performance of the servo motor again, and then adjust the frequency in the same way, f new = f cur + βf syn , and finally adjust the torque T new = T cur + βT syn . Continuously monitor the performance of the servo motor after each adjustment until the workload difference variable meets the requirements.

[0100] In the embodiment of the present invention, through correlation analysis, the multi-parameter collaborative adjustment parameters are obtained, taking into account the mutual influence between parameters, predicting the performance change in combination with the dynamic model, achieving precise collaborative control, and improving the comprehensive performance of the servo motor; and determining the adjustment sequence according to the influence amplitude, determining the adjustment range according to the workload difference, constructing a scientific and reasonable adjustment strategy, improving the adjustment efficiency, and quickly adapting to the load change; finally, the orderly adjustment process and reasonable amplitude control ensure the stability of the servo motor during the adjustment process, reduce the risk of system oscillation, and improve the system reliability.

[0101] In one embodiment, the descriptions of steps 4031 - 4036 are as follows:

[0102] Step 4031: Construct a local optimization model based on the optimal control parameters and physical principles.

[0103] Optionally, taking the DC servo motor as an example, the energy-saving control system starts from the physical principle and also involves the voltage balance equation of the servo motor The electromagnetic torque equation is T = K t I α , and the motion equation is Let the optimal control parameters include voltage U opt , current I opt , torque T opt , etc. The local optimization model takes the energy consumption E, rotational speed stability S w and torque response T r of the servo motor as the objective function. The energy consumption rotational speed stability S w can be represented by the variance of the rotational speed fluctuation , and the torque response T r can be measured by the time t T it takes for the torque to reach the target value from the initial value. The local optimization model is J 0 = w 1 E + w 2 S + w 3 T r , where w 1 , w 2 , w 3 are the weight coefficients of each objective, and w 1 + w 2 + w 3 = 1, and J 0 represents the objective function of the local optimization model.

[0104] Step 4032, perform a performance analysis on the current workload state to obtain the actual constraint factors for the servo motor.

[0105] Optionally, the energy-saving control system conducts a comprehensive analysis of the current workload state, including load torque T L , rotational speed w, current I, etc. The actual constraint factors include the temperature rise constraint of the servo motor, where T temp ≤ T max , where T temp represents the actual temperature of the servo motor, and T max represents the maximum allowable temperature of the servo motor; the voltage range constraint U min ≤ U ≤ U max , the current range constraint I min ≤ I ≤ I max , etc. By monitoring the operating data of the servo motor and combining with the technical parameters of the servo motor, these actual constraint factors are determined. Considering the actual constraint factors can ensure that the servo motor operates under safe and stable conditions and avoid damage or failure of the servo motor caused by excessive adjustment of control parameters.

[0106] Step 4033, analyze the environmental data where the servo motor is located to determine the environmental impact factors.

[0107] Optionally, the energy-saving control system obtains environmental data according to the real-time monitoring of the environment of the servo motor, including temperature T env , humidity H env , altitude h env , etc. These environmental factors will affect the performance of the servo motor. For example, a high-temperature environment will reduce the heat dissipation efficiency of the servo motor, and a high-humidity environment may affect the insulation performance of the servo motor. According to the current environmental monitoring data and the working load status, determine the degree of influence of environmental impact factors on the performance of the servo motor. A relationship model between environmental factors and servo motor performance indicators can be established through experimental data, such as the efficiency change curve η(T env ) of the servo motor at different temperatures. Then determine the environmental impact factors.

[0108] Step 4034, fuse the actual constraint factors and environmental impact factors with the local optimization model respectively to obtain the global optimization model.

[0109] Optionally, the energy-saving control system incorporates the actual constraint factors and environmental impact factors into the local optimization model J 0 . For actual constraint factors, such as the temperature rise constraint T temp ≤T max , it can be added to the objective function by introducing a penalty term. Let the penalty function where k 1 is the penalty coefficient. For environmental impact factors, such as the influence of temperature on efficiency η(T env ), the energy consumption objective function E can be modified to Then the global optimization model J quan =J 0 +P temp +P vol +P curr +…, where P vol , P curr , etc. are penalty terms for voltage, current and other constraint conditions, and J quan represents the objective function of the global optimization model after fusing the constraints.

[0110] Step 4035, perform multi-objective optimization on the global optimization model to obtain the optimal control parameter combination.

[0111] Optionally, the energy-saving control system uses the genetic algorithm to perform multi-objective optimization on the global optimization model J quan . The basic steps of the genetic algorithm include initializing the population, selection, crossover, mutation and iterative update. When initializing the population, a group of control parameter combinations are randomly generated as the initial population. In the selection operation, according to the objective function J quanEvaluate the individuals in the population based on the values, and select the individuals with high fitness as the parent generation. The crossover operation exchanges some genes of the parent individuals to generate offspring individuals. The mutation operation randomly mutates the genes of some offspring individuals. Through multiple iterative updates, continuously optimize the population until the termination condition is met (such as reaching the maximum number of iterations or the objective function value converges), and obtain the optimal control parameter combination. The genetic algorithm is used to perform multi-objective optimization on the global optimization model, which can take into account multiple performance indicators and constraint conditions simultaneously, and obtain the control parameter combination with the optimal comprehensive performance.

[0112] Step 4036, based on the optimal control parameter combination, use linear interpolation to regulate the servo motor.

[0113] Optionally, the energy-saving control system sets the current control parameter vector as The optimal control parameter combination is During the regulation process, use the method of linear interpolation to gradually adjust the control parameters. Let the regulation time be t e , divide the regulation process into N n time periods, and the time interval of each time period is In the φ-th time period, the control parameter vector where φ = 1, 2,..., N n . The linear interpolation regulation method can make the control parameters smoothly transition from the current value to the optimal value, avoid the impact on the servo motor caused by parameter mutation, and ensure the stability of the servo motor operation.

[0114] The embodiment of the present invention comprehensively considers the optimal control parameters, physical principles, actual constraint factors and environmental impact factors, constructs a global optimization model, and realizes the comprehensive optimization of various performances of the servo motor; and introduces actual constraint factors and environmental impact factors to ensure the servo motor operates under safe and stable conditions, improving the reliability and adaptability of the servo motor; finally, uses linear interpolation for regulation to make the control parameters transition smoothly, avoiding the impact on the servo motor caused by parameter mutation, and ensuring the stability of the servo motor operation.

[0115] Please refer to Figure 3 , Figure 3 is the embodiment diagram of the electronic device provided by the embodiment of the present invention. As Figure 3 shown, the embodiment of the present invention provides an electronic device 300, including a memory 310, a processor 320, and a computer program 311 stored on the memory 310 and executable on the processor 320. When the processor 320 executes the computer program 311, the following steps are implemented:

[0116] An acquisition unit: used to acquire the current working parameters during the operation of the servo motor;

[0117] Status analysis unit: configured to input current working parameters into a status analysis model to obtain the current workload status and the optimal control parameters output by the status analysis model; the status analysis model is trained based on sample parameter data and corresponding sample label results; the sample label results are the label results of the current workload status and the optimal control parameters.

[0118] Difference analysis unit: configured to perform difference analysis based on the current workload status and the target workload status to obtain a workload difference variable; the target load status is the expected load status preset for the servo motor.

[0119] Regulation unit: configured to perform trigger condition analysis based on the workload difference variable, and regulate the servo motor according to the trigger condition analysis result in combination with the optimal control parameters.

[0120] Please refer to Figure 4 , Figure 4 which is the embodiment diagram of the computer-readable storage medium provided by the embodiment of the present invention. As Figure 4 shown, this embodiment provides a computer-readable storage medium 400, on which a computer program 311 is stored. When the computer program 311 is executed by a processor, the following steps are implemented:

[0121] Acquisition unit: configured to acquire the current working parameters during the operation of the servo motor.

[0122] Status analysis unit: configured to input current working parameters into a status analysis model to obtain the current workload status and the optimal control parameters output by the status analysis model; the status analysis model is trained based on sample parameter data and corresponding sample label results; the sample label results are the label results of the current workload status and the optimal control parameters.

[0123] Difference analysis unit: configured to perform difference analysis based on the current workload status and the target workload status to obtain a workload difference variable; the target load status is the expected load status preset for the servo motor.

[0124] Regulation unit: configured to perform trigger condition analysis based on the workload difference variable, and regulate the servo motor according to the trigger condition analysis result in combination with the optimal control parameters.

[0125] On the other hand, the present invention also provides a computer program product. The computer program product includes a computer program, the computer program can be stored on a non-transitory computer-readable storage medium, and when the computer program is executed by a processor, the computer can execute the servo motor energy-saving control method provided by the above-mentioned methods. The method includes:

[0126] Acquisition unit: used to acquire the current working parameters during the operation of the servo motor;

[0127] State analysis unit: used to input the current working parameters into the state analysis model to obtain the current working load state and the optimal control parameters output by the state analysis model; the state analysis model is trained based on the sample parameter data and the corresponding sample label results; the sample label results are the label results of the current working load state and the optimal control parameters;

[0128] Difference analysis unit: used to perform difference analysis based on the current working load state and the target working load state to obtain the working load difference variable; the target load state is the expected load state preset for the servo motor;

[0129] Regulation unit: used to perform trigger condition analysis based on the working load difference variable, and regulate the servo motor according to the trigger condition analysis result in combination with the optimal control parameters.

[0130] The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative labor.

[0131] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus the necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions to make a computer device (which can be a personal computer, server, or network device, etc.) execute the methods of each embodiment or some parts of the embodiments.

[0132] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of each embodiment of the present invention.

Claims

1. A servo motor energy-saving control system, characterized in that: include: An energy-saving control center, an acquisition unit, a state analysis unit, a difference analysis unit, and a control unit. The energy-saving control center is connected to the acquisition unit, the state analysis unit, the difference analysis unit, and the control unit, respectively, to manage each unit; An acquisition unit, used for acquiring current working parameters of the servo motor during operation; A state analysis unit, used for inputting the current working parameters into a state analysis model to obtain the current workload state and the optimal control parameters output by the state analysis model; the state analysis model is trained based on the sample parameter data and the corresponding sample label results; the sample label results are the label results of the current workload state and the optimal control parameters; a difference analysis unit, configured to perform a difference analysis based on the current workload state and a target workload state to obtain a workload difference variable; the target load state is a desired load state preset by the servo motor; A control unit is used to perform a trigger condition analysis based on the workload difference variable, and to control the servo motor according to the trigger condition analysis result in combination with the optimal control parameter.

2. A servo motor energy-saving control method, implemented based on the servo motor energy-saving control system according to claim 1, characterized in that: The servo motor energy-saving control method comprises: Used to obtain the current working parameters of the servo motor during operation; Used to input the current working parameters into the state analysis model to obtain the current workload state and optimal control parameters output by the state analysis model; the state analysis model is trained based on sample parameter data and corresponding sample label results; the sample label results are label results of the current workload state and optimal control parameters; Used to perform difference analysis based on the current workload state and the target workload state to obtain a workload difference variable; the target load state is a desired load state preset for the servo motor; It is used to perform trigger condition analysis based on the workload difference variable, and regulate the servo motor according to the trigger condition analysis result combined with the optimal control parameters.

3. The servo motor energy-saving control method according to claim 2, characterized in that: The trigger condition includes a first preset threshold and a second preset threshold, and the trigger condition analysis is performed based on the workload difference variable, and the servo motor is regulated according to the trigger condition analysis result combined with the optimal control parameter, including: If the workload difference variable is less than a first preset threshold, the servo motor is regulated based on the first adjustment strategy combined with the optimal control parameter when it is determined that the preset condition is met; the preset condition is: whether the adjacent workload difference variables are continuous and whether the total value of the accumulated workload difference variables is greater than or equal to the first preset threshold; If the workload difference variable is greater than or equal to a first preset threshold and less than or equal to a second preset threshold, adjusting the servo motor based on a second adjustment strategy in combination with the optimal control parameters; If the workload difference variable is greater than a second preset threshold, the servo motor is adjusted based on a third adjustment strategy in combination with the optimal control parameters.

4. The servo motor energy-saving control method according to claim 3, characterized in that: The regulating the servo motor based on the first adjustment strategy in combination with the optimal control parameters includes: Performing difference processing on the optimal control parameter and the current working parameter to obtain a multi-dimensional adjustment vector; determining a load fluctuation amplitude based on the workload difference variable within a preset time period; Determining a reference range of a preset adjustment step length based on the magnitude of the load fluctuation amplitude; Modifying the reference range based on the current workload difference variable to obtain a target adjustment step size; Based on the multi-dimensional adjustment vector and the target adjustment step, the control parameters are gradually updated.

5. The servo motor energy-saving control method according to claim 3, characterized in that: The step of adjusting the servo motor based on the second adjustment strategy in combination with the optimal control parameters includes: Performing a correlation analysis on the optimal control parameter and the current working parameter to obtain a multi-parameter coordinated adjustment parameter; Constructing a servo motor dynamic model based on physical principles, and adjusting the performance change trend of the servo motor based on the current workload state and the predicted parameters of the servo motor dynamic model; Determining the adjustment order of different control parameters based on the multi-parameter coordinated adjustment parameters and the magnitude of the impact of the performance change trend on the servo motor performance; determining an adjustment magnitude based on the magnitude and direction of the workload difference variable; The parameters of the servo motor are adjusted based on the adjustment sequence and the adjustment amplitude.

6. The servo motor energy-saving control method according to claim 3, characterized in that: The step of adjusting the servo motor based on the third adjustment strategy in combination with the optimal control parameters includes: Constructing a local optimization model based on the optimal control parameters and physical principles; Performing performance analysis on the current workload state to obtain actual constraints on the servo motor; Analyze the environmental data of the servo motor and determine the environmental influencing factors; The actual constraint factors and the environmental influencing factors are respectively integrated with the local optimization model to obtain the global optimization model; Performing multi-objective optimization on the global optimization model to obtain an optimal control parameter combination; The servo motor is regulated by linear interpolation based on the optimal control parameter combination.

7. The servo motor energy-saving control method according to claim 2, characterized in that: The performing difference analysis based on the current workload state and the target workload state to obtain a workload difference variable includes: Performing time series alignment on the working parameters in the current workload state and the target workload state to obtain a corresponding sequence of the working parameters; Performing grey relational analysis on the corresponding sequence of the working parameters to obtain a grey relational degree; Performing a transition probability analysis based on the corresponding sequence of the working parameters to obtain a state transition probability matrix; A workload evaluation function is constructed based on the grey relational degree and the state transition probability matrix, and a workload difference variable is determined based on the workload evaluation function.

8. The servo motor energy-saving control method according to claim 7, characterized in that: The grey correlation analysis is performed on the sequence corresponding to the working parameters to obtain the grey correlation degree, including: Perform polynomial feature mapping based on the corresponding sequence of the working parameters to obtain a feature space; Perform spatial correlation analysis based on the feature space to obtain a correlation matrix; Based on the correlation matrix and the preset dimension weights, nonlinear correlation synthesis is performed to obtain the grey correlation degree.

9. An electronic device, comprising: Memory for storing computer software programs; A processor is used to read and execute the computer software program, wherein when the processor executes the computer software program, it implements the servo motor energy-saving control method as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium having a computer software program stored therein, characterized in that: When the computer software program is executed by a processor, the servo motor energy-saving control method according to any one of claims 1 to 7 is implemented.

Images