Mechanical clearance prediction method, device and system and medium

Abstract

The invention is suitable for the technical field of mechanical engineering, and provides a mechanical gap prediction method, device and system and a medium, and the method comprises the steps: obtaining a first mechanical gap value of mechanical equipment, and obtaining operation data and environment data when the mechanical equipment works; wherein the first mechanical clearance value is obtained through measurement when the mechanical equipment works under the standard condition; predicting a second mechanical gap value of the mechanical equipment through a pre-trained prediction model according to the operation data and the environment data; constructing a mechanical parameter calculation function according to the operation data, and adjusting the first mechanical gap value according to the mechanical parameter calculation function to obtain a third mechanical gap value; and performing weighted fusion on the first mechanical gap value, the second mechanical gap value and the third mechanical gap value to obtain a mechanical gap value of the mechanical equipment. According to the method, the mechanical gap of the mechanical equipment in the automatic calibration system of the inductive linear displacement sensor can be predicted in real time, and the prediction accuracy is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of mechanical engineering, and particularly relates to a method, device, system and medium for predicting mechanical clearance. Background Art

[0002] The automatic calibration system of a high-precision inductive linear displacement sensor uses automatic control technology to automatically complete operations such as loading and unloading, displacement monitoring and data recording of the inductive linear displacement sensor, so as to realize the automatic calibration of the inductive linear displacement sensor.

[0003] In the automatic calibration system of the inductive linear displacement sensor, the mechanical clearance of mechanical equipment is a key factor affecting the calibration accuracy and system stability. In the past, the traditional method of measuring mechanical clearance mostly used manual operation of measuring tools. Common measuring tools include dial indicators, laser rangefinders, etc. Such measuring methods can usually only be carried out during the initial commissioning of mechanical equipment, and it is difficult to track the dynamic changes of mechanical clearance. In addition, mechanical clearance will be affected by many factors. Changes in environmental conditions such as temperature and humidity, as well as mechanical vibrations, etc., will cause changes in mechanical clearance. Traditional measuring methods do not fully consider these factors and cannot provide real-time and accurate predictions in a changing working environment, which may have an adverse impact on the performance of mechanical equipment and its long-term operation stability. Summary of the Invention

[0004] In view of this, embodiments of the present invention provide a method, device, system and medium for predicting mechanical clearance to predict the mechanical clearance of mechanical equipment in the automatic calibration system of an inductive linear displacement sensor in real time and improve the accuracy of the prediction.

[0005] The first aspect of the embodiments of the present invention provides a method for predicting mechanical clearance, including:

[0006] Obtaining a first mechanical clearance value of mechanical equipment, and obtaining operation data and environmental data when the mechanical equipment is working; wherein, the first mechanical clearance value is measured when the mechanical equipment works under standard conditions;

[0007] Predicting a second mechanical clearance value of the mechanical equipment through a pre-trained prediction model according to the operation data and the environmental data;

[0008] Constructing a mechanical parameter calculation function according to the operation data, and adjusting the first mechanical clearance value according to the mechanical parameter calculation function to obtain a third mechanical clearance value;

[0009] Performing weighted fusion on the first mechanical clearance value, the second mechanical clearance value and the third mechanical clearance value to obtain the mechanical clearance value of the mechanical equipment.

[0010] In a possible implementation, the operating data includes the direction, pulses, and rotational speed of the servo motor of the mechanical equipment, the environmental data includes temperature and humidity, and the prediction model is a deep learning model;

[0011] Predicting the second mechanical clearance value of the mechanical equipment based on the operating data and the environmental data through a pre-trained prediction model includes:

[0012] Inputting the direction, the pulses, the rotational speed, the temperature, and the humidity into the deep learning model to obtain the second mechanical clearance value of the mechanical equipment.

[0013] In a possible implementation, the operating data includes the mechanical transmission radius, mechanical stiffness coefficient, reverse acceleration, load torque, and rotational speed of the servo motor of the mechanical equipment;

[0014] Constructing a mechanical parameter calculation function based on the operating data includes:

[0015] Construct a mechanical parameter calculation function:

[0016] where Δx mach is the third mechanical clearance value; T l is the load torque; r is the mechanical transmission radius; k s is the mechanical stiffness coefficient; a is the reverse acceleration; t lag is the mechanical hysteresis time, Δx gap is the first mechanical clearance value, v motor is the rotational speed of the servo motor.

[0017] In a possible implementation, the environmental data includes temperature and humidity;

[0018] Before weighted fusion of the first mechanical clearance value, the second mechanical clearance value, and the third mechanical clearance value, it further includes:

[0019] Determining the weights of the first mechanical clearance value, the second mechanical clearance value, and the third mechanical clearance value according to the environmental data;

[0020] where the weights of the first mechanical clearance value and the third mechanical clearance value are negatively correlated with the temperature, and the weight of the second mechanical clearance value is positively correlated with the temperature; the weights of the second mechanical clearance value and the third mechanical clearance value are negatively correlated with the humidity, and the weight of the first mechanical clearance value is positively correlated with the humidity; the sum of the weights of the first mechanical clearance value, the second mechanical clearance value, and the third mechanical clearance value is equal to 1.

[0021] In a possible implementation manner, determining weights of the first mechanical clearance value, the second mechanical clearance value, and the third mechanical clearance value according to the environmental data includes:

[0022] According to determine the weight of the first mechanical clearance value;

[0023] According to determine the weight of the second mechanical clearance value;

[0024] According to determine the weight of the third mechanical clearance value;

[0025] Wherein, temperature correction factor α T = 1 + K T (T - T0), where T0 is the standard temperature, K T is the temperature sensitivity coefficient, and T is the real-time temperature; humidity correction factor α H = 1 + K H (H - H0), where H0 is the standard humidity, K H is the humidity sensitivity coefficient, and H is the real-time humidity.

[0026] In a possible implementation manner, before performing weighted fusion on the first mechanical clearance value, the second mechanical clearance value, and the third mechanical clearance value, it further includes:

[0027] Perform data cleaning on the first mechanical clearance value, the second mechanical clearance value, and the third mechanical clearance value through a Kalman filter to filter out abnormal data values.

[0028] In a possible implementation manner, the method further includes:

[0029] Obtain historical operation data of the mechanical equipment at preset intervals;

[0030] According to the historical operation data, correct the calibration value of the preset parameters of the mechanical equipment.

[0031] A second aspect of the embodiments of the present invention provides a mechanical clearance prediction device, including:

[0032] An acquisition module, configured to acquire a first mechanical clearance value of a mechanical equipment, and acquire operation data and environmental data when the mechanical equipment is working; wherein, the first mechanical clearance value is measured when the mechanical equipment works under standard conditions;

[0033] A prediction module, configured to predict a second mechanical clearance value of the mechanical equipment according to the operation data and the environment data through a pre-trained prediction model;

[0034] A calculation module, configured to construct a mechanical parameter calculation function according to the operation data, and adjust the first mechanical clearance value according to the mechanical parameter calculation function to obtain a third mechanical clearance value;

[0035] A fusion module, configured to perform weighted fusion on the first mechanical clearance value, the second mechanical clearance value, and the third mechanical clearance value to obtain a mechanical clearance value of the mechanical equipment.

[0036] In a possible implementation manner, the operation data includes the direction, pulses, and rotational speed of a servo motor of the mechanical equipment, the environment data includes temperature and humidity, and the prediction model is a deep learning model;

[0037] The prediction module is configured to:

[0038] Input the direction, the pulses, the rotational speed, the temperature, and the humidity into the deep learning model to obtain a second mechanical clearance value of the mechanical equipment.

[0039] In a possible implementation manner, the operation data includes the mechanical transmission radius, mechanical stiffness coefficient, reverse acceleration, load torque, and rotational speed of a servo motor of the mechanical equipment;

[0040] The calculation module is configured to:

[0041] Construct a mechanical parameter calculation function:

[0042] where Δx mach is the third mechanical clearance value; T l is the load torque; r is the mechanical transmission radius; k s is the mechanical stiffness coefficient; a is the reverse acceleration; t lag is the mechanical lag time, Δx gap is the first mechanical clearance value, v motor is the rotational speed of the servo motor.

[0043] In a possible implementation manner, the environment data includes temperature and humidity;

[0044] Before performing weighted fusion on the first mechanical clearance value, the second mechanical clearance value, and the third mechanical clearance value, the fusion module is further configured to:

[0045] Determine the weights of the first mechanical clearance value, the second mechanical clearance value, and the third mechanical clearance value according to the environmental data;

[0046] Among them, the weights of the first mechanical clearance value and the third mechanical clearance value are negatively correlated with the temperature, and the weight of the second mechanical clearance value is positively correlated with the temperature; the weights of the second mechanical clearance value and the third mechanical clearance value are negatively correlated with the humidity, and the weight of the first mechanical clearance value is positively correlated with the humidity; the sum of the weights of the first mechanical clearance value, the second mechanical clearance value, and the third mechanical clearance value is equal to 1.

[0047] In a possible implementation manner, the fusion module is specifically configured to:

[0048] According to Determine the weight of the first mechanical clearance value;

[0049] According to Determine the weight of the second mechanical clearance value;

[0050] According to Determine the weight of the third mechanical clearance value;

[0051] Among them, Temperature correction factor α T = 1 + K T (T - T0), where T0 is the standard temperature, K T is the temperature sensitivity coefficient, and T is the real-time temperature; humidity correction factor α H = 1 + K H (H - H0), where H0 is the standard humidity, K H is the humidity sensitivity coefficient, and H is the real-time humidity.

[0052] In a possible implementation manner, before performing weighted fusion on the first mechanical clearance value, the second mechanical clearance value, and the third mechanical clearance value, the fusion module is further configured to:

[0053] Perform data cleaning on the first mechanical clearance value, the second mechanical clearance value, and the third mechanical clearance value through a Kalman filter to filter out abnormal data values.

[0054] In a possible implementation manner, the acquisition module is further configured to:

[0055] Obtain the historical operation data of the mechanical equipment at preset time intervals;

[0056] Correct the calibration value of the preset parameters of the mechanical equipment according to the historical operation data.

[0057] The third aspect of the embodiments of the present invention provides an automatic calibration system for an inductive linear displacement sensor, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the method in the first aspect or any implementation manner of the first aspect are implemented.

[0058] The fourth aspect of the embodiments of the present invention provides a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the steps of the method in the first aspect or any implementation manner of the first aspect are implemented.

[0059] The beneficial effects of the embodiments of the present invention compared with the prior art are as follows:

[0060] In the embodiments of the present invention, when the mechanical equipment works under standard conditions, a static first mechanical clearance value is measured; according to the real-time operation data and environmental data of the mechanical equipment, through a pre-trained prediction model, a dynamic second mechanical clearance value of the mechanical equipment is predicted; by constructing a mechanical parameter calculation function, the first mechanical clearance value is adjusted to obtain a third mechanical clearance value; the final mechanical clearance value is obtained by comprehensively considering the first mechanical clearance value, the second mechanical clearance value, and the third mechanical clearance value. This solution not only realizes the real-time prediction of mechanical clearance, but also comprehensively considers the influence of factors such as environment, load, and mechanical vibration on the change of mechanical clearance through the fusion of physical models and sensor data (operation data and environmental data), improving the accuracy of prediction. Description of the Drawings

[0061] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for use in the embodiments or the description of the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0062] Figure 1 It is a schematic structural diagram of a mechanical equipment provided by an embodiment of the present invention;

[0063] Figure 2 It is a schematic implementation flow diagram of a mechanical clearance prediction method provided by an embodiment of the present invention Figure 1 ;

[0064] Figure 3 It is a schematic implementation flow diagram of a mechanical clearance prediction method provided by an embodiment of the present invention Figure 2 ;

[0065] Figure 4It is a schematic diagram of the mechanical clearance prediction device provided by an embodiment of the present invention;

[0066] Figure 5 It is a schematic diagram of the automatic calibration system of an inductive linear displacement sensor provided by an embodiment of the present invention. Detailed implementation manners

[0067] In the following description, specific details such as specific system structures and technologies are presented for the purpose of illustration rather than limitation, so as to thoroughly understand the embodiments of the present invention. However, those skilled in the art should clearly understand that the present invention can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from interfering with the description of the present invention.

[0068] In order to illustrate the technical solutions described in the present invention, specific embodiments are used for illustration below.

[0069] The automatic calibration system of the resistive displacement sensor is a device for automatically calibrating the resistive displacement sensor, mainly composed of a control module, an execution module, a host computer, and other parts. This system uses automatic control technology to control the linear movement module in the execution module, so that the movable fixture approaches or moves away from the fixed fixture, thereby performing loading and unloading operations on the resistive displacement sensor. During the loading process, the dial indicator measures the displacement of the linear movement module in real time, and the dynamic resistance strain gauge measures the change value of the resistive displacement sensor jointly clamped by the movable fixture and the fixed fixture. The control module automatically completes operations such as loading and unloading, displacement monitoring, and data recording according to the displacement data of the high-resolution dial indicator and the strain gauge value of the dynamic resistance strain gauge, thereby realizing the automatic calibration of the resistive displacement sensor.

[0070] Figure 1 It is a schematic diagram of the mechanical equipment in the automatic calibration system of the inductive linear displacement sensor provided by an embodiment of the present invention, and the explanations of the serial numbers are as follows: 1 is the support plate, 2 is the first middle sliding block, 3 is the second middle sliding block, 4 is the fixed vertical plate, 5 is the high-precision lead screw, 6 is the large optical axis, 7 is the small optical axis, 8 is the clamping block, 9 is the front bottom plate, 10 is the rear bottom plate, 11 is the large gear, 12 is the small gear, 13 is the support block 1 and support block 2, 14 is the servo motor, 15 is the motor reducer, 16 is the motor bottom plate, 17 is the bracket, 18 is the inductive linear displacement sensor, and 19 is the dial indicator.

[0071] In the automatic calibration system of the inductive linear displacement sensor, mechanical clearance will have various effects on the calibration process and results, including:

[0072] (1) Mechanical clearances in the system can cause uncertainties in displacement transmission. For example, in a linear motion module, when a stepper motor drives a moving fixture, if there is a clearance, at the initial stage when the motor starts to act, it will first fill the clearance, while the moving fixture does not actually start to move at this time. This results in a deviation between the displacement measured by the micrometer and the actual displacement acting on the strain gauge, making the displacement measurement inaccurate and thus affecting the calibration results based on displacement measurement.

[0073] (2) During multiple loading and unloading cycles, the influence of mechanical clearances will gradually accumulate. The changes and compensation conditions of the clearance may be different each time. As the number of cycles increases, the cumulative error will continuously increase, making the error of displacement measurement more and more obvious, and ultimately seriously affecting the reliability of strain gauge calibration.

[0074] (3) Mechanical clearances can cause additional vibrations and noises during system operation. When a stepper motor drives a moving part, collisions and impacts will occur at the clearance, causing vibrations in the system. These vibrations not only interfere with the normal measurement of the micrometer and strain gauge, but also may cause increased wear of system components and shorten the service life of the system.

[0075] (4) The existence of clearances will cause a lag phenomenon in the system response. When it is necessary to adjust the loading force or displacement, due to the influence of the clearance, the system cannot respond accurately in a timely manner, resulting in a decrease in the control accuracy of the calibration process and even possible instability in the calibration process.

[0076] (5) During the calibration process of inductive linear displacement sensors in multiple batches, the influence of mechanical clearances may lead to significant differences between the calibration results of different batches. Even for strain gauges of the same specification, due to the different performances of the clearances at different times and under different loading conditions, the calibration results lack consistency, causing difficulties in the quality assessment and use of strain gauges.

[0077] Therefore, it is very necessary to accurately predict the mechanical clearances of mechanical equipment in real time. Traditional mechanical clearance measurement methods usually rely on manual measuring devices, which can only provide static initial measurement values and cannot dynamically track the changes of mechanical clearances. Moreover, traditional methods ignore the real-time changes of external factors and have poor generalization ability, and cannot provide stable and accurate predictions in a changing working environment.

[0078] To this end, this embodiment proposes a mechanical clearance prediction method based on dynamic weighted fusion of a prediction model and calculated mechanical parameters, comprehensively considering the influence of factors such as environment, load, and mechanical vibration on the change of mechanical clearance, so as to provide more accurate prediction results. In this embodiment, the operating data and environmental data of the mechanical equipment during operation can be obtained through sensors, and these data are transmitted to the controller in the inductive linear displacement sensor automatic calibration system. The controller executes the mechanical clearance prediction method according to these data to achieve the prediction of mechanical clearance. It should be noted that this embodiment does not limit the device for executing the mechanical clearance prediction method. For example, it can also be a remote server, etc.

[0079] Figure 2 is a schematic diagram of the implementation process of the mechanical clearance prediction method provided by the embodiment of the present invention. The following will be based on Figure 2 , and this method will be described in detail:

[0080] Step S201, obtain the first mechanical clearance value of the mechanical equipment, and obtain the operating data and environmental data of the mechanical equipment during operation.

[0081] In this embodiment, the first mechanical clearance value can be the mechanical clearance value measured by a measuring tool under certain conditions when the mechanical equipment runs for the first time.

[0082] Taking Figure 1 as an example of mechanical equipment, when the equipment runs, the servo motor will run a set number of pulses. During this process, due to a series of operation cooperations, the first sliding block will also move accordingly, so the value of the dial indicator will change, and the mechanical clearance value can be measured. The specific steps are as follows:

[0083] (1) Set standard conditions, including standard environmental temperature and humidity (for example, temperature T = 25°C, humidity H = 50%, which can be changed according to the actual situation).

[0084] (2) Check the status of the mechanical equipment: Ensure that components such as the servo motor, mechanical device, and dial indicator are correctly installed and operating normally.

[0085] (3) Equipment initialization: Calibrate the dial indicator, reset the servo motor to the initial position, and zero the value of the dial indicator or record the initial reading;

[0086] (4) Set the operating parameters of the servo motor: Set the number of pulses for the servo motor to run, ensure that the first sliding block can move a certain distance without exceeding the range of the dial indicator, select the servo motor to move in the low-speed mode, and observe the change of the dial indicator;

[0087] (5) Measure the basic clearance during forward and reverse operation: Start the servo motor, run the set number of pulses, let the motor move forward and backward, observe the movement of the first sliding block. When the dial indicator just shows a reading, record the pulse value m of the operation at this time. Calculate the clearance distance Δx1 according to the following formula:

[0088]

[0089] Among them, m1 is the pulse value required for one revolution; M is the pitch, that is, the distance that the lead screw theoretically advances in one revolution.

[0090] (6) Take the average value of multiple measurements to improve the measurement accuracy.

[0091] In this embodiment, the operation data of the mechanical equipment may include but are not limited to: the mechanical transmission radius, mechanical stiffness coefficient, reverse acceleration, load torque of the mechanical equipment, the direction, pulse, and speed of the servo motor, etc. The environmental data may include but are not limited to: temperature and humidity. These data can be obtained through control signals, sensors, etc., and will not be introduced in detail here.

[0092] Step S202, according to the operation data and environmental data, predict the second mechanical clearance value of the mechanical equipment through a pre-trained prediction model.

[0093] Here, the prediction model can be any neural network model. For example, it is a deep learning model. Through the deep learning model, it is possible to comprehensively consider the environmental data and monitor and dynamically adjust the prediction result of the mechanical clearance in real time during the operation of the mechanical equipment.

[0094] The deep learning model includes an input layer, an intermediate layer, and an output layer. The intermediate layer contains a multi-scale convolution module, a deep temporal memory module, and a multi-head attention module. The output layer includes a fully connected layer and a ReLU activation function connected in sequence.

[0095] The input layer inputs the operation data and environmental data into the intermediate layer after normalization processing.

[0096] The multi-scale convolution module includes three one-dimensional convolutional (Conv1D) layers, a fully connected layer, and a normalization layer connected in sequence. The input data first passes through the three Conv1D layers, and these convolutional layers are responsible for extracting local features in the input time series signal. Each layer of convolutional operation continuously extracts more abstract features, enhancing the model's perception ability of signal changes. Then, the data enters the fully connected layer, which maps the features extracted from the convolutional layer to a higher-dimensional space to prepare for subsequent deep temporal processing. The normalization layer follows immediately to normalize the output of the fully connected layer to accelerate model convergence and stabilize the training process.

[0097] The deep temporal memory module includes a bidirectional LSTM layer, an LSTM layer, a residual connection layer, and a normalization layer connected in sequence. The data then enters the bidirectional LSTM layer. The LSTM (Long Short-Term Memory network) can capture long-term dependencies in the input data. In the bidirectional LSTM, the network processes the temporal data in two directions (forward and backward), which helps to better understand the time series characteristics of the input signal. Then, the LSTM layer further captures the complex relationships in the time series. The residual connection layer connects the LSTM layer to the subsequent layer. By directly adding the output of the LSTM layer to the input of the current layer, it helps to alleviate the vanishing gradient problem and promotes information transmission. The normalization layer ensures the stability of the output of the LSTM layer and helps to accelerate the training process of the model.

[0098] The multi-head attention module includes a multi-head attention layer, a temporal feature flattening layer, and a Dropout layer connected in sequence. The data passes through the multi-head attention layer. This step mainly automatically learns which features (such as environmental factors like temperature T, humidity H, etc.) are most important for gap prediction by calculating the relationships between the input features. The multi-head attention layer assigns a weight to each input feature, enabling the model to focus on the important input information. Then, the data enters the temporal feature flattening layer to flatten the features output by the multi-head attention layer for subsequent fully connected layer processing. Finally, the Dropout layer randomly discards a part of the neurons to prevent the model from overfitting and improve the generalization ability of the model.

[0099] Finally, the data enters the output layer. The output layer includes a fully connected layer and a ReLU activation function. Among them, the fully connected layer is used to map the extracted features to the target space, and the ReLU activation function increases the non-linear ability of the model, which helps to improve the fitting effect of the model.

[0100] In step S203, a mechanical parameter calculation function is constructed based on the operating data, and the first mechanical gap value is adjusted according to the mechanical parameter calculation function to obtain the third mechanical gap value.

[0101] In this embodiment, the mechanical parameters of the device can also be extracted from the operating data. The mechanical parameter calculation function is constructed through the principles of mechanical dynamics. The first mechanical gap value is adjusted according to the mechanical parameter calculation function to obtain the third mechanical gap value. The physical model can accurately quantify the impact of mechanical transmission, provide a reliable compensation benchmark for mechanical gap calculation, and enhance the operability and practicality.

[0102] In step S204, the first mechanical gap value, the second mechanical gap value, and the third mechanical gap value are weighted and fused to obtain the mechanical gap value of the mechanical device.

[0103] Here, the weights of the first mechanical clearance value, the second mechanical clearance value, and the third mechanical clearance value can be preset fixed values, which are not limited in this embodiment. By combining the basic first mechanical clearance value, the second mechanical clearance value predicted by the deep learning model, and the third mechanical clearance value calculated by the mechanical parameter calculation function, the influences of mechanical factors and external environmental factors can be comprehensively considered, and the prediction accuracy can be improved.

[0104] In the embodiment of the present invention, a static first mechanical clearance value is measured when the mechanical equipment works under standard conditions; according to the real-time operation data and environmental data of the mechanical equipment, a dynamic second mechanical clearance value of the mechanical equipment is predicted through a pre-trained prediction model; by constructing a mechanical parameter calculation function, the first mechanical clearance value is adjusted to obtain a third mechanical clearance value; by integrating the first mechanical clearance value, the second mechanical clearance value, and the third mechanical clearance value, a final mechanical clearance value is obtained. This solution not only realizes the real-time prediction of the mechanical clearance, but also comprehensively considers the influences of factors such as the environment, load, and mechanical vibration on the change of the mechanical clearance through the fusion of the physical model and sensor data (operation data and environmental data), and improves the accuracy of the prediction.

[0105] For the convenience of understanding this solution, Figure 3 a more detailed implementation process schematic diagram of the mechanical clearance prediction method is provided, including:

[0106] Step S301, when the mechanical equipment runs under standard conditions for the first time, measure the first mechanical clearance value with a micrometer. For the specific implementation method, refer to Figure 2 the description in the embodiment.

[0107] Step S302, obtain the operation data and environmental data when the mechanical equipment works.

[0108] Step S303, extract the direction, pulse, and speed of the servo motor from the operation data, and extract the temperature and humidity from the environmental data, and input these data into the deep learning model together. The deep learning model outputs the second mechanical clearance value of the mechanical equipment.

[0109] Step S304, extract mechanical parameters from the operation data: mechanical transmission radius, mechanical stiffness coefficient, reverse acceleration, load torque, and speed of the servo motor. By using the principle of mechanical dynamics to construct a mechanical parameter calculation function, the first mechanical clearance is adjusted to obtain a third mechanical clearance.

[0110] The explanations and acquisition methods of each parameter are as follows:

[0111] ① Load torque (T l ): Obtained through the installed servo motor model.

[0112] ② Mechanical transmission radius (r): It is the transmission radius of the high-precision lead screw, an inherent parameter of the equipment, determined by the mechanical design drawing, with the unit of mm.

[0113] ③ Mechanical stiffness coefficient (ks): It is an inherent parameter of the equipment, characterizing the anti-deformation ability of the mechanical structure, calibrated through material mechanics experiments, with the unit of N / m.

[0114] ④ Reverse acceleration (a): It is calculated by differentiating the rotational speed signal collected by the encoder, with the unit of N·m, and the formula is where ω is the angular velocity, with the unit of rad / s 2 .

[0115] ⑤ Mechanical lag time (t lag ): It characterizes the delay of the mechanical system in responding to commands, and the calculation formula is:

[0116]

[0117] where, Δx gap is the first mechanical clearance value calibrated through basic experiments (such as measured with a micrometer during the first installation), with the unit of m; v motor is the real-time rotational speed of the servo motor, collected by the encoder and converted into linear velocity (v motor =ω·r), with the unit of m / s.

[0118] The mechanical parameter calculation function is:

[0119]

[0120] where: The first term Based on the elastic deformation theory, the load torque T l is converted into a force through the transmission radius r and the elastic deformation amount generated under the mechanical stiffness coefficient k s is Combined to get The second term Based on the kinematic acceleration cumulative effect, the displacement deviation caused by the reverse acceleration a within the lag time t lag is approximately the uniform acceleration motion formula

[0121] This mechanical parameter calculation function adjusts the first mechanical clearance through physical formulas to obtain the third mechanical clearance related to the mechanical parameters, quantifying the influence of mechanical transmission characteristics on the mechanical clearance.

[0122] Step S305, according to the environmental data, determine the weights of the first mechanical clearance value, the second mechanical clearance value, and the third mechanical clearance value.

[0123] Here, the weights of the first mechanical clearance value and the third mechanical clearance value are negatively correlated with temperature, and the weight of the second mechanical clearance value is positively correlated with temperature; the weights of the second mechanical clearance value and the third mechanical clearance value are negatively correlated with humidity, and the weight of the first mechanical clearance value is positively correlated with humidity; the sum of the weights of the first mechanical clearance value, the second mechanical clearance value, and the third mechanical clearance value is equal to 1.

[0124] The purpose of this step is to implement a dynamic weight allocation mechanism that dynamically adjusts the weighting coefficients according to environmental factors such as temperature and humidity, enabling the weighting coefficients to adaptively adjust according to real-time environmental changes.

[0125] The rules for temperature's effect on weight allocation are as follows:

[0126] When humidity remains constant and temperature increases, the mechanical clearance may decrease due to the thermal expansion effect. In this case, the weight (w3) of the third mechanical parameter value should be reduced because after the temperature rises, the gap change calculated by the physical model is restricted and cannot fully predict the details of the change. The weight (w2) of the second mechanical parameter value should be increased because the deep learning model can more flexibly adapt to the complex non-linear effects brought about by temperature changes. Especially when the temperature changes significantly, the deep learning model usually performs better. The weight (w1) of the first mechanical parameter value should be decreased because the first mechanical parameter value is static and may not be able to well reflect the dynamic impact of temperature on the mechanical clearance. Conversely, a decrease in temperature usually causes the material to contract, resulting in an increase in the mechanical clearance. In this case, the weight (w3) of the third mechanical parameter value should be increased because at low temperatures, the mechanical model's prediction of contraction is usually more accurate. The weight (w2) of the second mechanical parameter value should be reduced because the mechanical changes at low temperatures are relatively simple and the physical model may perform better.

[0127] The rules for humidity's effect on weight allocation are as follows:

[0128] When the temperature remains constant and the humidity increases, it usually leads to a deterioration of the lubrication state, an increase in friction, and thus a contraction of the mechanical clearance. The weight (w3) of the third mechanical parameter value should be reduced because the physical model is less sensitive to changes in lubrication and friction when the humidity increases. The weight (w2) of the second mechanical parameter value should be reduced because the deep learning model may rely too much on historical data when the humidity changes and thus fails to accurately capture the immediate impact of humidity changes. The weight (w1) of the first mechanical parameter value should be increased because the first mechanical parameter value is relatively more robust to humidity changes. Especially when the humidity increases, the first mechanical parameter value usually does not change significantly. Conversely, when the humidity decreases, it usually improves the lubrication state, reduces friction, and thus increases the mechanical clearance. The weight of the third mechanical parameter value (w3) should be increased because humidity changes affect the friction characteristics of lubrication, and the physical model can usually better predict the change of clearance under humidity changes.

[0129] Exemplarily, a temperature correction factor α T and a humidity correction factor α H can be set as:

[0130] α T = 1 + K T (T - T0)

[0131] α H = 1 + K H (H - H0)

[0132] Where:

[0133] T0 is the standard temperature (usually taken as 25 °C), and H0 is the standard humidity (usually taken as 50% RH);

[0134] K T is the temperature sensitivity coefficient, determined by a thermal expansion experiment, and the typical value is KT = 0.002 / °C;

[0135] K H is the humidity sensitivity coefficient, determined by a friction coefficient calibration experiment, and the typical value is KH = 0.001.

[0136] The weight of the first mechanical clearance value is:

[0137] The weight of the second mechanical clearance value is:

[0138] The weight of the third mechanical clearance value is:

[0139] Satisfying w1 + w2 + w3 = 1.

[0140] Step S306 , performing data cleaning on the first mechanical clearance value, the second mechanical clearance value, and the third mechanical clearance value through a Kalman filter to filter out abnormal data values.

[0141] Specific processes may include:

[0142] ① Establish state equation and observation equation, taking the mechanical clearance change rate as the state variable;

[0143] ② Dynamically adjust the confidence of the predicted value and the observed value through the Kalman gain to suppress pulse interference.

[0144] Step S307: weighted fusion is performed on the first mechanical clearance value, the second mechanical clearance value and the third mechanical clearance value to obtain a mechanical clearance value of the mechanical equipment.

[0145] The calculation formula is:

[0146] Δx final =w1·Δx base +w2·Δx DL +w3·Δx mach

[0147] Δx base : The first mechanical clearance value (recorded during the first operation);

[0148] Δx DL : The second mechanical clearance value;

[0149] Δx mach : The third mechanical clearance value;

[0150] w1, w2, w3: Dynamic adjustment function, i.e. weights.

[0151] This embodiment realizes real-time monitoring of the mechanical gap during the operation of the equipment, solving the problem that the traditional method cannot respond to the change of the mechanical gap in real time. The present invention does not rely on a single method alone, but uses a deep learning model, while integrating physical model calculations and sensor data, to take into account various factors that may affect the gap change, such as temperature and humidity, load, and mechanical vibration. Specifically, by integrating the first mechanical gap value, the second mechanical gap value, and the third mechanical gap value, and dynamically adjusting the weighting coefficient based on environmental factors such as temperature and humidity, the weighting coefficient can be adapted to the real-time environmental changes, thereby significantly improving the accuracy of the prediction.

[0152] In some embodiments, the mechanical clearance prediction method further includes a process of pre-training a deep learning model:

[0153] When the mechanical equipment has reverse movement, record the input direction signal, current pulse, rotation speed, and environmental data (temperature and humidity) at this time through the data acquisition module. After collecting the conditional data, measure the mechanical clearance value at this time according to the displacement of the dial indicator. At this time, the acquisition of a dataset is completed; by changing parameter values such as the direction signal, previous pulse, rotation speed, and environmental data (temperature and humidity), the mechanical clearance values under different working conditions are collected to construct a high-quality dataset.

[0154] Preprocess the dataset, including filtering to remove abnormal data.

[0155] Based on the dataset, train and construct a prediction model of a deep learning model, and use the Adam optimizer to perform backpropagation supervised training on the prediction model to determine the final prediction model.

[0156] In some embodiments, the mechanical clearance prediction method further includes: obtaining the historical operation data of the mechanical equipment at preset intervals; according to the historical operation data, correcting the calibration value of the preset parameters of the mechanical equipment. For example, the calibration values of the mechanical stiffness coefficient and the mechanical hysteresis time can be corrected through historical data feedback to adapt to equipment aging.

[0157] It should be understood that the magnitudes of the sequence numbers of the steps in the above embodiments do not mean the order of execution. The order of execution of each process should be determined according to its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

[0158] Figure 4 The structural schematic diagram of the mechanical clearance prediction device provided by the embodiment of the present invention is shown. For the sake of convenience of description, only the parts related to the embodiment of the present invention are shown and are described in detail as follows:

[0159] As Figure 4 shown, the mechanical clearance prediction device 40 includes:

[0160] An acquisition module 41, configured to acquire a first mechanical clearance value of the mechanical equipment, and acquire operation data and environmental data when the mechanical equipment is working; wherein, the first mechanical clearance value is obtained by measuring when the mechanical equipment works under standard conditions.

[0161] A prediction module 42, configured to predict a second mechanical clearance value of the mechanical equipment according to the operation data and the environmental data through a pre-trained prediction model.

[0162] A calculation module 43, configured to construct a mechanical parameter calculation function according to the operation data, and adjust the first mechanical clearance value according to the mechanical parameter calculation function to obtain a third mechanical clearance value.

[0163] The fusion module 44 is used to perform weighted fusion on the first mechanical clearance value, the second mechanical clearance value, and the third mechanical clearance value to obtain the mechanical clearance value of the mechanical equipment.

[0164] In a possible implementation manner, the operating data includes the direction, pulses, and rotational speed of the servo motor of the mechanical equipment, the environmental data includes temperature and humidity, and the prediction model is a deep learning model;

[0165] The prediction module 42 is used for:

[0166] Input the direction, pulses, rotational speed, temperature, and humidity into the deep learning model to obtain the second mechanical clearance value of the mechanical equipment.

[0167] In a possible implementation manner, the operating data includes the mechanical transmission radius, mechanical stiffness coefficient, reverse acceleration, load torque, and rotational speed of the servo motor of the mechanical equipment;

[0168] The calculation module 43 is used for:

[0169] Construct a mechanical parameter calculation function:

[0170] where, Δx mach is the third mechanical clearance value; T l is the load torque; r is the mechanical transmission radius; k s is the mechanical stiffness coefficient; a is the reverse acceleration; t lag is the mechanical lag time, Δx gap is the first mechanical clearance value, v motor is the rotational speed of the servo motor.

[0171] In a possible implementation manner, the environmental data includes temperature and humidity;

[0172] Before performing weighted fusion on the first mechanical clearance value, the second mechanical clearance value, and the third mechanical clearance value, the fusion module 44 is further used for:

[0173] Determine the weights of the first mechanical clearance value, the second mechanical clearance value, and the third mechanical clearance value according to the environmental data;

[0174] where, the weights of the first mechanical clearance value and the third mechanical clearance value are negatively correlated with temperature, and the weight of the second mechanical clearance value is positively correlated with temperature; the weights of the second mechanical clearance value and the third mechanical clearance value are negatively correlated with humidity, and the weight of the first mechanical clearance value is positively correlated with humidity; the sum of the weights of the first mechanical clearance value, the second mechanical clearance value, and the third mechanical clearance value is equal to 1.

[0175] In a possible implementation manner, the fusion module 44 is specifically used for:

[0176] According to Determine the weight of the first mechanical clearance value;

[0177] According to Determine the weight of the second mechanical clearance value;

[0178] According to Determine the weight of the third mechanical clearance value;

[0179] Among them, The temperature correction factor α T = 1 + K T (T - T0), where T0 is the standard temperature and K T Is the temperature sensitivity coefficient, and T is the real-time temperature; the humidity correction factor α H = 1 + K H (H - H0), where H0 is the standard humidity and K H Is the humidity sensitivity coefficient, and H is the real-time humidity.

[0180] In a possible implementation, before weighted fusion of the first mechanical clearance value, the second mechanical clearance value, and the third mechanical clearance value, the fusion module 44 is further configured to:

[0181] Perform data cleaning on the first mechanical clearance value, the second mechanical clearance value, and the third mechanical clearance value through a Kalman filter to filter out abnormal data values.

[0182] In a possible implementation, the acquisition module 41 is further configured to:

[0183] Acquire the historical operation data of the mechanical equipment at preset intervals;

[0184] According to the historical operation data, correct the calibration value of the preset parameters of the mechanical equipment.

[0185] In the embodiment of the present invention, a static first mechanical clearance value is measured when the mechanical equipment works under standard conditions; according to the real-time operation data and environmental data of the mechanical equipment, a dynamic second mechanical clearance value of the mechanical equipment is predicted through a pre-trained prediction model; by constructing a mechanical parameter calculation function, the first mechanical clearance value is adjusted to obtain a third mechanical clearance value; the first mechanical clearance value, the second mechanical clearance value, and the third mechanical clearance value are integrated to obtain the final mechanical clearance value. This solution not only realizes the real-time prediction of the mechanical clearance, but also comprehensively considers the influence of factors such as the environment, load, and mechanical vibration on the change of the mechanical clearance through the fusion of the physical model and sensor data (operation data and environmental data), improving the accuracy of the prediction.

[0186] Figure 5 It is a schematic diagram of an automatic calibration system 50 for an inductive linear displacement sensor provided by an embodiment of the present invention. AsFigure 5 As shown, the automatic calibration system 50 of the inductive linear displacement sensor in this embodiment includes: a processor 51, a memory 52, and a computer program 53 stored in the memory 52 and executable on the processor 51, such as a mechanical clearance prediction program. When the processor 51 executes the computer program 53, it implements the steps in the above-mentioned various embodiments of the mechanical clearance prediction method, such as Figure 2 the steps S201 to S204 shown. Alternatively, when the processor 51 executes the computer program 53, it implements the functions of each module / unit in the above-mentioned device embodiments, such as Figure 4 the functions of the modules 41 to 44 shown.

[0187] Exemplarily, the computer program 53 can be divided into one or more modules / units. The one or more modules / units are stored in the memory 52 and executed by the processor 51 to complete the present invention. The one or more modules / units can be a series of computer program instruction segments capable of performing specific functions, and these instruction segments are used to describe the execution process of the computer program 53 in the automatic calibration system 50 of the inductive linear displacement sensor.

[0188] The automatic calibration system 50 of the inductive linear displacement sensor may include, but is not limited to, a processor 51 and a memory 52. Those skilled in the art can understand that Figure 5 this is only an example of the automatic calibration system 50 of the inductive linear displacement sensor, and does not constitute a limitation on the automatic calibration system 50 of the inductive linear displacement sensor. It may include more or fewer components than shown in the figure, or combine some components, or different components. For example, the automatic calibration system 50 of the inductive linear displacement sensor may further include input / output devices, network access devices, buses, etc.

[0189] The so-called processor 51 may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0190] The memory 52 may be an internal storage unit of the inductive linear displacement sensor automatic calibration system 50, such as the hard disk or memory of the inductive linear displacement sensor automatic calibration system 50. The memory 52 may also be an external storage device of the inductive linear displacement sensor automatic calibration system 50, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc. equipped on the inductive linear displacement sensor automatic calibration system 50. Further, the memory 52 may also include both an internal storage unit of the inductive linear displacement sensor automatic calibration system 50 and an external storage device. The memory 52 is used to store the computer program and other programs and data required by the inductive linear displacement sensor automatic calibration system 50. The memory 52 may also be used to temporarily store the data that has been output or will be output.

[0191] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the above-mentioned division of each functional unit and module is used as an example. In actual applications, the above functions can be allocated to different functional units and modules according to needs, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiment can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above integrated units can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of each functional unit and module are only for the convenience of mutual distinction and do not limit the protection scope of the present application. The specific working processes of the units and modules in the above system can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0192] In the above embodiments, the descriptions of the respective embodiments have their own emphases. For the parts not detailed or recorded in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0193] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or by a combination of computer software and electronic hardware. Whether these functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0194] In the embodiments provided by the present invention, it should be understood that the disclosed device / system and method can be implemented in other ways. For example, the device / inductive linear displacement sensor automatic calibration system embodiments described above are merely illustrative. For example, the division of the modules or units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections between each other can be through some interfaces. The indirect couplings or communication connections of the devices or units can be in electrical, mechanical or other forms.

[0195] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0196] In addition, in each embodiment of the present invention, the functional units can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

[0197] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, to implement all or part of the processes in the above method embodiments of the present invention, it can also be completed by a computer program instructing relevant hardware. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-mentioned method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file or some intermediate form, etc. The computer-readable medium can include: any entity or device that can carry the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal, and software distribution medium, etc.

[0198] The above-described embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; 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 on 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 the various embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A mechanical clearance prediction method, characterized in that, Including: Obtaining a first mechanical clearance value of a mechanical device, and obtaining operation data and environmental data when the mechanical device is operating; wherein, the first mechanical clearance value is measured when the mechanical device operates under standard conditions; Predicting a second mechanical clearance value of the mechanical device through a pre-trained prediction model according to the operation data and the environmental data; Constructing a mechanical parameter calculation function according to the operation data, and adjusting the first mechanical clearance value according to the mechanical parameter calculation function to obtain a third mechanical clearance value; Performing weighted fusion on the first mechanical clearance value, the second mechanical clearance value, and the third mechanical clearance value to obtain a mechanical clearance value of the mechanical device.

2. The mechanical clearance prediction method according to claim 1, wherein The operation data includes the direction, pulse, and rotation speed of the servo motor of the mechanical device, and the environmental data includes temperature and humidity, and the prediction model is a deep learning model; The predicting the second mechanical clearance value of the mechanical device through a pre-trained prediction model according to the operation data and the environmental data includes: Inputting the direction, the pulse, the rotation speed, the temperature, and the humidity into the deep learning model to obtain the second mechanical clearance value of the mechanical device.

3. The mechanical clearance prediction method according to claim 1, wherein The operation data includes the mechanical transmission radius, mechanical stiffness coefficient, reverse acceleration, load torque, and rotation speed of the servo motor of the mechanical device; The constructing a mechanical parameter calculation function according to the operation data includes: Construct a mechanical parameter calculation function: Among them, Δx mach is the third mechanical clearance value; T l is the load torque; r is the mechanical transmission radius; k s is the mechanical stiffness coefficient; a is the reverse acceleration; t lag is the mechanical hysteresis time, Δx gap is the first mechanical clearance value, v motor is the rotational speed of the servo motor.

4. The mechanical clearance prediction method according to claim 1, characterized in that The environmental data includes temperature and humidity; Before performing weighted fusion on the first mechanical clearance value, the second mechanical clearance value, and the third mechanical clearance value, it further includes: Determining weights of the first mechanical clearance value, the second mechanical clearance value, and the third mechanical clearance value according to the environmental data; Wherein, the weights of the first mechanical clearance value and the third mechanical clearance value are negatively correlated with the temperature, and the weight of the second mechanical clearance value is positively correlated with the temperature; the weights of the second mechanical clearance value and the third mechanical clearance value are negatively correlated with the humidity, and the weight of the first mechanical clearance value is positively correlated with the humidity; the sum of the weights of the first mechanical clearance value, the second mechanical clearance value, and the third mechanical clearance value is equal to 1.

5. The mechanical clearance prediction method according to claim 4, characterized in that, The determining the weights of the first mechanical clearance value, the second mechanical clearance value, and the third mechanical clearance value according to the environmental data includes: According to Determine the weight of the first mechanical clearance value; According to determine the weight of the second mechanical clearance value; According to Determine the weight of the third mechanical clearance value; Among them, temperature correction factor α T = 1 + K T (T - T0), where T0 is the standard temperature, and K T is the temperature sensitivity coefficient, and T is the real-time temperature; humidity correction factor α H = 1 + K H (H - H0), where H0 is the standard humidity, and K H is the humidity sensitivity coefficient, and H is the real-time humidity.

6. The mechanical clearance prediction method according to any one of claims 1 to 5, characterized in that Before performing weighted fusion on the first mechanical clearance value, the second mechanical clearance value, and the third mechanical clearance value, it further includes: Performing data cleaning on the first mechanical clearance value, the second mechanical clearance value, and the third mechanical clearance value through a Kalman filter to filter out abnormal data values.

7. The mechanical clearance prediction method according to any one of claims 1 to 5, characterized in that, The method further includes: Obtaining historical operation data of the mechanical device at preset time intervals; Correcting a calibration value of a preset parameter of the mechanical device according to the historical operation data.

8. A mechanical clearance prediction device, characterized in that, Including: An acquisition module, configured to acquire a first mechanical clearance value of a mechanical device, as well as operating data and environmental data when the mechanical device is operating; wherein, the first mechanical clearance value is obtained by measuring when the mechanical device operates under standard conditions; A prediction module, configured to predict a second mechanical clearance value of the mechanical device according to the operating data and the environmental data through a pre-trained prediction model; A calculation module, configured to construct a mechanical parameter calculation function according to the operating data, and adjust the first mechanical clearance value according to the mechanical parameter calculation function to obtain a third mechanical clearance value; A fusion module, configured to perform weighted fusion on the first mechanical clearance value, the second mechanical clearance value and the third mechanical clearance value to obtain the mechanical clearance value of the mechanical device.

9. An automatic calibration system for an inductive linear displacement sensor, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images