Method, device and equipment for adjusting clamping force of magnetostrictive displacement sensor and storage medium
By combining a physical constraint-based long short-term memory network model with a piezoelectric ceramic actuator, precise and rapid dynamic adjustment of the clamping force of the magnetostrictive displacement sensor is achieved, solving the problems of low efficiency and large error in clamping force control in the prior art, and realizing high-precision control of the magnetostrictive displacement sensor.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- HEBEI UNIV OF SCI & TECH
- Filing Date
- 2025-12-31
- Publication Date
- 2026-05-12
AI Technical Summary
In the existing technology, the clamping force control of magnetostrictive displacement sensors mainly relies on manual adjustment or simple linear models, which leads to low efficiency and easy introduction of errors. It is difficult to cope with nonlinear changes such as different material parameters and environmental temperature fluctuations, and cannot meet the requirements of high-precision control.
By employing a physical constraint-based long short-term memory network model combined with a piezoelectric ceramic actuator, and through a control process of data acquisition, model prediction, data fusion, and closed-loop correction, precise and rapid dynamic adjustment of clamping force is achieved. This includes acquiring working parameters, predicting clamping force, calculating and fusing clamping force, and feedback adjustment.
It significantly improves the accuracy and stability of clamping force control, can adapt to the nonlinear effects of multiple factors such as temperature and material parameters, and enhances the industrial versatility and practical application value of the method.
Smart Images
Figure CN122018586A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of sensor control, and in particular to a method, apparatus, device, and storage medium for adjusting the clamping force of a magnetostrictive displacement sensor. Background Technology
[0002] Magnetostrictive displacement sensors are widely used in precision manufacturing, materials testing, petrochemicals, and other fields due to their wide measurement range, fast response speed, and strong anti-interference ability. Their measurement accuracy is directly related to the stability of the clamping force. Insufficient clamping force can easily cause the measured object to shift, while excessive force may cause deformation of the object or damage to the sensor, both of which will lead to measurement errors.
[0003] In existing technologies, clamping force control mainly relies on manual adjustment or simple linear models based on empirical formulas. Manual adjustment is inefficient and prone to human error; simple linear models cannot accurately handle nonlinear changes caused by factors such as different material parameters and fluctuations in ambient temperature. Existing solutions cannot meet the high-precision control requirements of magnetostrictive displacement sensors in industrial scenarios. Summary of the Invention
[0004] In view of this, embodiments of the present invention provide a method, apparatus, device and storage medium for adjusting the clamping force of a magnetostrictive displacement sensor, so as to improve the clamping force control accuracy of the magnetostrictive displacement sensor.
[0005] In a first aspect, embodiments of the present invention provide a method for adjusting the clamping force of a magnetostrictive displacement sensor, comprising: Obtain the operating parameters of the magnetostrictive displacement sensor; wherein, the operating parameters include the target clamping force, as well as the measured clamping force and output displacement of the magnetostrictive displacement sensor; Based on the operating parameters, the predicted clamping force of the magnetostrictive displacement sensor is determined using a long short-term memory network model based on physical constraints. Calculate the fusion clamping force based on the measured clamping force and the output displacement; The clamping force adjustment amount of the magnetostrictive displacement sensor is determined based on the target clamping force, the predicted clamping force, and the fused clamping force; based on the clamping force adjustment amount, the clamping force of the magnetostrictive displacement sensor is adjusted by feedback through a piezoelectric ceramic actuator.
[0006] In one possible implementation, the operating parameters also include the material parameters of the object being measured and the real-time temperature; The step of determining the predicted clamping force of the magnetostrictive displacement sensor based on the operating parameters and a long short-term memory network model with physical constraints includes: The output displacement, the real-time temperature, the measured clamping force, and the material parameters of the object under test are input into a pre-trained long short-term memory network model. The clamping force output by the long short-term memory network model is then corrected according to physical constraints to obtain the predicted clamping force.
[0007] In one possible implementation, the physical constraints include mechanical property constraints of the magnetostrictive displacement sensor, output force range constraints of the piezoelectric ceramic actuator, and material property constraints of the object being measured. The mechanical property constraint of the magnetostrictive displacement sensor is used to constrain the clamping force output by the long short-term memory network model to be within the allowable clamping force range corresponding to the mechanical property of the magnetostrictive displacement sensor. The piezoelectric ceramic actuator output force range constraint is used to constrain the clamping force output by the long short-term memory network model between the maximum and minimum values of the piezoelectric ceramic actuator output force; The material property constraint of the tested object is used to ensure that the deformation of the tested object is less than a preset threshold under the clamping force output by the long short-term memory network model.
[0008] In one possible implementation, calculating the fused clamping force based on the measured clamping force and the output displacement includes: Obtain the equivalent stiffness of the measurement system, and determine the equivalent clamping force based on the equivalent stiffness and the output displacement; The equivalent clamping force is temperature-compensated based on the real-time temperature to obtain the displacement clamping force. The weighted sum of the displacement clamping force and the measured clamping force is calculated to obtain the fused clamping force.
[0009] In one possible implementation, determining the clamping force adjustment amount of the magnetostrictive displacement sensor based on the target clamping force, the predicted clamping force, and the fused clamping force includes: Using the target clamping force as a reference, the measurement deviation between the target clamping force and the fused clamping force, and the prediction deviation between the target clamping force and the predicted clamping force are calculated respectively. The clamping force adjustment amount is obtained by calculating the weighted sum of the measured deviation and the predicted deviation based on the preset deviation adjustment weight.
[0010] In one possible implementation, the step of adjusting the clamping force of the magnetostrictive displacement sensor via a piezoelectric ceramic actuator based on the clamping force adjustment amount includes: The PID control signal of the piezoelectric ceramic actuator is determined based on the clamping force adjustment amount; The PID control signal of the piezoelectric ceramic actuator controls the clamping force of the magnetostrictive displacement sensor to be adjusted by feedback.
[0011] In one possible implementation, the method further includes: Every preset time interval, the running parameters within the preset time interval are used as new samples to incrementally train the long short-term memory network model and update the model's weight parameters.
[0012] Secondly, embodiments of the present invention provide a clamping force adjustment device for a magnetostrictive displacement sensor, comprising: The parameter acquisition module is used to acquire the operating parameters of the magnetostrictive displacement sensor; wherein, the operating parameters include the target clamping force, as well as the measured clamping force and output displacement of the magnetostrictive displacement sensor; The clamping force prediction module is used to determine the predicted clamping force of the magnetostrictive displacement sensor based on the operating parameters and a long short-term memory network model with physical constraints. The fusion calculation module is used to calculate the fusion clamping force based on the measured clamping force and the output displacement; The clamping force adjustment module is used to determine the clamping force adjustment amount of the magnetostrictive displacement sensor based on the target clamping force, the predicted clamping force, and the fused clamping force; and to perform feedback adjustment of the clamping force of the magnetostrictive displacement sensor through a piezoelectric ceramic actuator based on the clamping force adjustment amount.
[0013] Thirdly, embodiments of the present invention provide an electronic device, including a memory and a processor, wherein the memory stores a computer program and the processor executes the steps of the method as described in the first aspect or any implementation thereof.
[0014] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the method as described in the first aspect or any implementation thereof.
[0015] The beneficial effects of the embodiments of the present invention compared with the prior art are as follows: In this embodiment of the invention, based on the operating parameters and a long short-term memory network model with physical constraints, the predicted clamping force of the magnetostrictive displacement sensor is determined. This balances the model's ability to capture nonlinear relationships with the reasonable limitations of physical constraints, improving the accuracy and reliability of clamping force prediction. Based on the measured clamping force and output displacement, a fused clamping force is calculated, integrating displacement data monitored by the magnetostrictive displacement sensor and measured clamping force data. This avoids measurement errors from single force values and enhances the accuracy of clamping force feedback adjustment data. Based on the target clamping force, predicted clamping force, and fused clamping force, the clamping force adjustment amount of the magnetostrictive displacement sensor is determined. Based on this adjustment amount, a piezoelectric ceramic actuator is used to provide feedback adjustment of the clamping force of the magnetostrictive displacement sensor, achieving precise and rapid dynamic control of the clamping force, pushing the clamping force closer to the target value, and ensuring the accuracy of sensor displacement monitoring. This invention, through the construction of a complete control process—data acquisition, model prediction, preliminary control, data fusion, and closed-loop correction—forms a multi-dimensional collaborative clamping force adjustment system, significantly improving the accuracy and stability of clamping force control for magnetostrictive displacement sensors. Furthermore, it can adaptively respond to the nonlinear influences of multiple factors such as temperature and material parameters, thereby greatly reducing the deviation between the monitored displacement and the actual displacement of the measured object, effectively enhancing the industrial versatility and practical application value of the method. Attached Figure Description
[0016] Figure 1 This is a schematic diagram illustrating the implementation process of a magnetostrictive displacement sensor clamping force adjustment method according to an embodiment of the present invention; Figure 2 This is a flowchart illustrating the layered implementation of a method for adjusting the clamping force of a magnetostrictive displacement sensor according to an embodiment of the present invention. Figure 3 This is a flowchart of the decision layer of a magnetostrictive displacement sensor clamping force adjustment method provided in an embodiment of the present invention; Figure 4 This is an execution layer flowchart of a magnetostrictive displacement sensor clamping force adjustment method provided in an embodiment of the present invention; Figure 5 This is a schematic diagram of the structure of a magnetostrictive displacement sensor clamping force adjustment device provided in an embodiment of the present invention; Figure 6 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation
[0017] The present application will be described more clearly below with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the function of the present application, but do not limit the present application in any way. It should be noted that those skilled in the art can make several modifications and improvements without departing from the concept of the present application. These all fall within the protection scope of the present application.
[0018] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.
[0019] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.
[0020] In the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.
[0021] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.
[0022] Furthermore, the term "multiple" mentioned in the embodiments of this application should be interpreted as two or more.
[0023] Magnetostrictive displacement sensors generate reflected waves through magnetic field coupling and calculate displacement using the time of the reflected wave. Their detection accuracy depends on the stability of the components maintained by the clamping force. Insufficient clamping force can cause relative slippage or unstable contact between the magnet and the measured object, leading to oscillation or positional shift of the waveguide wire installed in the system. Changes in the relative position of the waveguide wire and the magnet or measured object alter the propagation path, time delay, and coupling amplitude of the reflected wave, resulting in positional shifts and errors in the sensor's displacement calculation. Conversely, excessive clamping force can cause stress changes or localized deformation in the waveguide wire or the measured object, altering the effective acoustic and magnetic propagation characteristics of the waveguide wire, which may also cause measurement errors or drift.
[0024] Existing technologies rely on manual or linear model adjustments. The former is inefficient and prone to large errors, while the latter cannot cope with complex and variable disturbances such as temperature and material changes. Therefore, developing a magnetostrictive displacement sensor clamping force adjustment method that can integrate real-time data from multiple influencing factors and adaptively control the clamping force has significant practical implications and application value.
[0025] See Figure 1 This invention provides a method for adjusting the clamping force of a magnetostrictive displacement sensor, detailed below: Step S101: Obtain the operating parameters of the magnetostrictive displacement sensor; wherein, the operating parameters include, but are not limited to, the target clamping force, as well as the measured clamping force and output displacement of the magnetostrictive displacement sensor.
[0026] In one possible implementation, a micro-force ring sensor is used to acquire the measured clamping force of the magnetostrictive displacement sensor in real time, and the micro-force ring sensor uses a non-continuous mode to periodically acquire data according to a set sampling period as the true value for periodic calibration.
[0027] A magnetostrictive displacement sensor clamps an object to be measured using a fixture, and measures the displacement changes of the object. The actual clamping force can be detected in real time by a micro-force loop sensor. This micro-force loop sensor uses a non-continuous participation mode, periodically collecting data according to a set sampling period. This ensures feedback effectiveness while reducing sensor energy consumption and wear, extending its service life. Real-time data acquisition also ensures timely subsequent processing and control, enabling rapid response to system changes. Temperature data is collected by a temperature sensor, providing a basis for temperature drift compensation and reducing the influence of ambient temperature on clamping force measurement and control. In this embodiment of the invention, multiple operating parameters of the magnetostrictive displacement sensor are acquired, providing comprehensive and real-time basic data support for clamping force prediction and adjustment, avoiding control deviations caused by missing or delayed parameters.
[0028] Step S102: Based on the operating parameters, determine the predicted clamping force of the magnetostrictive displacement sensor using a Long Short-Term Memory (LSTM) network model based on physical constraints.
[0029] To adjust the clamping force of the magnetostrictive displacement sensor, this embodiment of the invention requires real-time correction of the clamping force based on the changes in the actual clamping force and output displacement of the magnetostrictive displacement sensor. However, the measured clamping force of the magnetostrictive displacement sensor is obtained from a force sensor, and the measurement results may be affected by factors such as sensor system aging and ambient temperature drift, failing to accurately reflect the actual clamping force applied by the magnetostrictive displacement sensor to the clamped object. Therefore, this embodiment of the invention receives operating parameters from the sensor device and determines the predicted clamping force of the magnetostrictive displacement sensor based on a physically constrained LSTM model. This avoids the problem of inaccurate measured clamping force affecting the accuracy of clamping force adjustment, while also taking into account the reasonable limitations of physical constraints, thus improving the accuracy and reliability of clamping force prediction.
[0030] In one possible implementation, the operating parameters also include the material parameters and real-time temperature of the object being measured; based on the operating parameters, a long short-term memory network model based on physical constraints is used to determine the predicted clamping force of the magnetostrictive displacement sensor, including: The output displacement, real-time temperature, measured clamping force, and material parameters of the object under test are input into the pre-trained long short-term memory network model. The clamping force output by the long short-term memory network model is then corrected according to the physical constraints to obtain the predicted clamping force.
[0031] For example, physical constraints may include, but are not limited to, constraints on the mechanical properties of magnetostrictive displacement sensors, constraints on the output force range of piezoelectric ceramic actuators, and constraints on the material properties of the object being measured.
[0032] In one possible implementation, the mechanical properties of the magnetostrictive displacement sensor are constrained to ensure that the clamping force output by the long short-term memory network model is within the allowable clamping force range corresponding to the mechanical properties of the magnetostrictive displacement sensor. The piezoelectric ceramic actuator output force range constraint is used to constrain the clamping force output by the long short-term memory network model between the maximum and minimum values of the piezoelectric ceramic actuator output force; The material property constraint of the tested object is used to ensure that the deformation of the tested object is less than a preset threshold under the clamping force output by the long short-term memory network model.
[0033] In this embodiment of the invention, multi-dimensional working parameters are used to provide comprehensive input to the clamping force prediction model, which enhances the nonlinear prediction capability of the model. At the same time, the physical constraints of the clamping force are used to correct the output of the model, which not only ensures the accuracy of the predicted clamping force, but also avoids it from exceeding the bearing limit of the sensing device and the object being measured, thus ensuring that the prediction results are safe and feasible.
[0034] Step S103: Calculate the fusion clamping force based on the measured clamping force and output displacement.
[0035] Since the measured clamping force is easily affected by sensor measurement errors, this embodiment of the invention considers the inherent mechanical relationship between output displacement and clamping force. Therefore, calculating the fused clamping force integrates the effective information of output displacement and measured clamping force, comprehensively reflecting the clamping force state of the magnetostrictive displacement sensor. Compared to relying on only a single data point, the fused clamping force can overcome the limitations of using measured data alone and reduce error interference. In this embodiment of the invention, the fused clamping force is calculated based on the actual clamping force and output displacement, which can integrate the displacement data monitored by the magnetostrictive displacement sensor and the measured clamping force data, enhancing the accuracy of the clamping force feedback data.
[0036] Step S104: Determine the clamping force adjustment amount of the magnetostrictive displacement sensor based on the target clamping force, the predicted clamping force, and the fused clamping force; Based on the clamping force adjustment amount, perform feedback adjustment of the clamping force of the magnetostrictive displacement sensor through a piezoelectric ceramic actuator.
[0037] The clamping force adjustment amount integrates the target clamping force, the predicted clamping force, and the fused clamping force under actual working conditions. It represents the adjustment range required to bring the actual clamping force closer to the target value. By adjusting the clamping force, the correction requirements of the clamping force can be precisely quantified, providing a more comprehensive reflection of the current true state of the clamping force and improving the reliability of feedback adjustment. In this embodiment of the invention, the adjustment amount is determined and adjusted via feedback from a piezoelectric ceramic actuator to achieve precise and rapid dynamic control of the clamping force, ensuring the accuracy of sensor displacement monitoring.
[0038] In one possible implementation, the clamping force adjustment of the magnetostrictive displacement sensor is determined based on the target clamping force, the predicted clamping force, and the fused clamping force, including: Using the target clamping force as a benchmark, the measurement deviation between the target clamping force and the fusion clamping force, as well as the prediction deviation between the target clamping force and the predicted clamping force, are calculated respectively. Based on the preset deviation adjustment weights, the weighted sum of the measured deviation and the predicted deviation is calculated to obtain the clamping force adjustment amount.
[0039] In this embodiment of the invention, by integrating real-time measurement deviation and prediction deviation, and calculating the adjustment amount based on preset weights, the accuracy of the adjustment amount is improved. By integrating the weighted contributions of the two deviations with the target clamping force as the benchmark, the clamping force adjustment amount can reflect both the difference between the sensor's measured result and the target value under the current working condition, and also integrate the model's prediction result, providing a reliable basis for feedback adjustment.
[0040] In one possible implementation, the clamping force of the magnetostrictive displacement sensor is adjusted by a piezoelectric ceramic actuator based on the clamping force adjustment amount, including: The PID control signal for the piezoelectric ceramic actuator is determined based on the clamping force adjustment. The PID control signal of the piezoelectric ceramic actuator is used to control the clamping force of the magnetostrictive displacement sensor to be adjusted by feedback.
[0041] In this embodiment of the invention, a PID controller is used to control the piezoelectric ceramic actuator to achieve precise and rapid adjustment of the clamping force. The adjustment data is used to update the LSTM model, enabling the feedback adjustment to have adaptive optimization capabilities. This allows the system to adapt to long-term changes in the magnetostrictive displacement sensor, significantly improving the accuracy and robustness of the clamping force control.
[0042] This invention constructs a complete control process of data acquisition, model prediction, preliminary control, data fusion, and closed-loop correction, forming a multi-dimensional collaborative clamping force adjustment system. This significantly reduces the deviation between the monitored displacement and the actual displacement of the measured object, and significantly improves the accuracy and stability of the clamping force control of the magnetostrictive displacement sensor. Furthermore, it can adaptively cope with the nonlinear effects of multiple factors such as temperature and material parameters, effectively enhancing the industrial versatility and practical application value of the method.
[0043] In some embodiments, relying solely on measured clamping force results for clamping force adjustment is susceptible to limitations in the accuracy of the measured results. However, the output displacement of a magnetostrictive displacement sensor is correlated with the clamping force and can serve as a supplementary basis for clamping force determination. Considering that temperature can affect the material stiffness and alter the correlation between displacement and clamping force, this solution employs a multi-step processing method to accurately fuse measured data and displacement correlation data to improve the reliability of this supplementary basis. Therefore, based on the measured clamping force and output displacement, the fused clamping force is calculated, including: Obtain the equivalent stiffness of the measurement system, and determine the equivalent clamping force based on the equivalent stiffness and output displacement; The displacement clamping force is obtained by temperature compensation of the equivalent clamping force based on the real-time temperature. The weighted sum of the displacement clamping force and the measured clamping force is calculated to obtain the combined clamping force.
[0044] Specifically, the measurement system includes a fixture, the object being measured, a piezoelectric ceramic actuator, and the connection structure between them. The equivalent stiffness of the measurement system is determined by the forces acting on the system and the total deformation of the system according to Hooke's Law. It reflects the comprehensive elastic characteristics of the piezoelectric ceramic actuator, fixture structure, contact interface, and the object being measured under stress, and is a key parameter for converting the displacement output of the magnetostrictive displacement sensor into clamping force.
[0045] In this embodiment of the invention, by integrating the measured clamping force with the temperature-compensated displacement clamping force, the accuracy limitations of a single measured data point are effectively compensated, the reliability of the clamping force feedback data is improved, the displacement data fully plays its supplementary role, and the resulting integrated clamping force is more in line with the actual working conditions, reducing the impact of single data deviation on the adjustment accuracy.
[0046] For example, operating parameters include real-time temperature, target clamping force, output displacement, predicted clamping force, fused clamping force, and clamping force adjustment.
[0047] In some embodiments, the clamping force prediction model cannot adapt to the dynamic changes in factors such as temperature and material properties in real time, and the prediction accuracy is prone to decay after multiple feedback adjustments. Therefore, the model needs to be optimized based on actual adjustment data. Thus, the method provided in this embodiment may further include: At preset intervals, the running parameters within the preset time are used as new samples to incrementally train the Long Short-Term Memory network model and update the model's weight parameters.
[0048] In this embodiment of the invention, multi-dimensional operating parameters are used as new training samples, and the weights of the LSTM model are optimized through incremental training, so that the model can continuously adapt to changes in actual working conditions and improve the accuracy of subsequent clamping force prediction.
[0049] See Figure 2 The present invention provides a layered implementation flowchart of a method for adjusting the clamping force of a magnetostrictive displacement sensor, which is described in detail below: (1) System initialization The hardware connections for the sensor layer, execution layer, edge computing layer, and cloud layer were completed according to the design. The sensor layer includes a magnetostrictive displacement sensor (resolution 0.001mm), a micro-force ring sensor (range 0-50N, accuracy ±0.05N), and a temperature sensor (accuracy ±0.3℃); the execution layer is equipped with a piezoelectric ceramic actuator (piezoelectric constant...). =300pm / V); the edge computing layer includes a physically constrained LSTM prediction layer and a decision layer, using industrial-grade edge servers; the cloud layer is deployed on cloud servers.
[0050] Preset the sampling period of the micro-force ring sensor (e.g., 100ms), the initial parameters of the LSTM model based on physical constraints (e.g., randomly initialize the weight matrices W and U), and the initial parameters of the PID controller in the edge computing layer. =2.0, =0.1, =0.05); set parameters such as model optimization cycle and data storage path in the cloud layer.
[0051] Start the real-time data processing program at the edge computing layer and the data receiving and model optimization program at the cloud layer to ensure that the communication protocols at each layer (such as MQTT) are running normally and complete the system self-test.
[0052] (2) Sensor layer data acquisition and transmission The magnetostrictive displacement sensor continuously outputs the change in clamp displacement. The data is simultaneously transmitted to the edge computing layer; the micro-force ring sensor collects the measured clamping force according to a set cycle. Each data collection is immediately uploaded to the edge computing layer; the temperature sensor collects the ambient temperature in real time. It is sent along with the output displacement.
[0053] After receiving sensor data, the edge computing layer triggers a preprocessing process to prepare for prediction by the physically constrained LSTM model.
[0054] (3) Physically constrained LSTM model prediction layer operation Edge computing layer receives , , Data according to Standardization is carried out, among which, It is a moment The input data includes various raw data measured by the sensors; and The input data at time 10:00 and 11:00 respectively The mean and standard deviation; The standardized data is input into the LSTM model to calculate the input gate, forget gate, output gate, cell state, and hidden state, and then... Obtain the raw prediction results of the LSTM model.
[0055] According to the maximum clamping force in the physical constraint correction condition, (For example, the maximum output force of the piezoelectric ceramic actuator is 30N) and the minimum clamping force is (For example, the minimum allowable clamping force of a magnetostrictive displacement sensor is 0N), according to The prediction results are corrected, and the corrected predicted clamping force is obtained. Transmitted to the decision-making level.
[0056] (4) Generation of instructions from the decision-making level See Figure 3 The present invention provides a decision-level flowchart of a method for adjusting the clamping force of a magnetostrictive displacement sensor, which is described in detail below: The decision-making level receives the preset target clamping force Predicting clamping force and the fusion clamping force fed back from the execution layer ,according to Calculate the clamping force adjustment amount (deviation adjustment weight) =0.6, =0.4).
[0057] Adjust the clamping force Substitute into the PID control calculation formula Calculate the control signal increment, and then... ( =0.5V / N) is converted into the input voltage of the piezoelectric ceramic driver.
[0058] according to ( =0.8, =0.2) Generate a PID control signal and send it to the execution layer.
[0059] (5) Adjustment and feedback of clamping force of the execution layer See Figure 4 The present invention provides an execution layer flowchart of a method for adjusting the clamping force of a magnetostrictive displacement sensor, which is described in detail below: Based on the output displacement measured by the magnetostrictive displacement sensor ,according to Calculate the displacement clamping force, where the equivalent stiffness of the measurement system is... =500N / mm, temperature sensitivity coefficient =0.002 / ℃, calibration reference temperature =25℃; Micro-force ring sensor collects measured clamping force ,according to ( =0.7) Calculate the fusion and feed it back to the decision-making level.
[0060] Obtain the equivalent stiffness of the measurement system The method can be: An offline calibration method was used, in which a mechanical dial indicator with an accuracy of 1 μm was installed on the fixture side to record the deformation of the measurement system. Record the temperature during calibration. A series of known force values (e.g., 0–25 N) are applied sequentially using a calibrated loading device. After stabilizing at each loading point, the forces acting on the measurement system are recorded. Total deformation of the measurement system According to Hooke's Law, the total deformation of the measurement system satisfies The temperature was obtained by fitting using the least squares method. Below And remove abnormal residuals to improve accuracy, among which This represents the residual. Repeated offline calibration under different temperature conditions yields the temperature-dependent equivalent stiffness. And fit a linear temperature compensation model. When calculating the displacement clamping force, the ambient temperature should be taken into account. Fit the equivalent stiffness of the measurement system at this temperature.
[0061] The piezoelectric ceramic actuator receives the input voltage of the piezoelectric ceramic actuator sent by the decision layer. ,according to Displacement is generated and converted into clamping force through the actuator:
[0062] in, It is the piezoelectric constant of the piezoelectric ceramic material.
[0063] (6) Cloud layer model optimization and synchronization The cloud layer receives historical data uploaded weekly by the edge computing layer (including but not limited to) , , , Control commands are organized and stored according to timestamps.
[0064] After preprocessing historical data, incremental learning is used to update the weight parameters of the LSTM model (such as optimizing the weight matrix W and U through stochastic gradient descent) to adapt the model to changes such as equipment aging and temperature drift.
[0065] The optimized model parameters are packaged and sent to the edge computing layer. Upon receiving the parameters, the edge computing layer immediately updates its local LSTM model to ensure that the latest model is used for the next prediction.
[0066] (7) Circular feedback regulation and abnormal handling Repeat steps (2) to (6) to achieve continuous and precise control of clamping force, and the edge computing layer records the control error in real time.
[0067] If no sensor data is received for more than 500ms, the edge computing layer automatically replaces the predicted clamping force with the predicted value of the most recent valid data; if the error between the model prediction result and the measured clamping force continues to exceed the preset threshold, the cloud layer performs emergency model optimization to ensure stable operation.
[0068] Through the above process, the embodiments of the present invention can achieve high-precision control of clamping force, which is suitable for high-precision scenarios such as precision manufacturing and material testing.
[0069] In this embodiment of the invention, data is collected collaboratively by multiple sensors. This is followed by LSTM model prediction with physical constraint correction, multi-force fusion calculation, and precise PID control. Combined with the linear driving characteristics of the piezoelectric ceramic actuator, high-precision closed-loop adjustment of the clamping force is achieved, effectively reducing the impact of single data errors and nonlinear interference on control accuracy. Continuous optimization of model parameters through incremental learning at the cloud layer adapts to long-term changes such as equipment aging and temperature drift. Coupled with cyclic feedback and anomaly handling mechanisms, this ensures the stability of the magnetostrictive displacement sensor's measurement accuracy and enhances the system's long-term reliable operation in precision manufacturing and other scenarios.
[0070] See Figure 5 This invention provides a magnetostrictive displacement sensor clamping force adjustment device 5, comprising: The parameter acquisition module 51 is used to acquire the operating parameters of the magnetostrictive displacement sensor; the operating parameters include the target clamping force, as well as the measured clamping force and output displacement of the magnetostrictive displacement sensor. The clamping force prediction module 52 is used to determine the predicted clamping force of the magnetostrictive displacement sensor based on the working parameters and a long short-term memory network model based on physical constraints. The fusion calculation module 53 is used to calculate the fusion clamping force based on the measured clamping force and the output displacement; The clamping force adjustment module 54 is used to determine the clamping force adjustment amount of the magnetostrictive displacement sensor based on the target clamping force, the predicted clamping force, and the fused clamping force; and to provide feedback adjustment of the clamping force of the magnetostrictive displacement sensor through a piezoelectric ceramic actuator based on the clamping force adjustment amount.
[0071] In one possible implementation, the clamping force prediction module 52 is used to input the output displacement, real-time temperature, measured clamping force and material parameters of the object being measured into a pre-trained long short-term memory network model, and to correct the clamping force output by the long short-term memory network model according to the physical constraints to obtain the predicted clamping force.
[0072] In one possible implementation, the fusion calculation module 53 is used to obtain the equivalent stiffness of the measurement system and determine the equivalent clamping force based on the equivalent stiffness and the output displacement. The displacement clamping force is obtained by temperature compensation of the equivalent clamping force based on the real-time temperature. The weighted sum of the displacement clamping force and the measured clamping force is calculated to obtain the combined clamping force.
[0073] In one possible implementation, the clamping force adjustment module 54 is used to calculate, based on the target clamping force, the measurement deviation between the target clamping force and the fused clamping force, and the prediction deviation between the target clamping force and the predicted clamping force. Based on the preset deviation adjustment weights, the weighted sum of the measured deviation and the predicted deviation is calculated to obtain the clamping force adjustment amount.
[0074] In one possible implementation, the clamping force adjustment module 54 is also used to determine the PID control signal of the piezoelectric ceramic actuator based on the clamping force adjustment amount; The PID control signal of the piezoelectric ceramic actuator is used to control the clamping force of the magnetostrictive displacement sensor to be adjusted by feedback.
[0075] In this embodiment of the invention, the clamping force adjustment device for the magnetostrictive displacement sensor, through multi-module collaboration, uses multi-dimensional working parameters as a basis and combines physical constraint LSTM model prediction data, measured data, and displacement data for fusion calculation to accurately determine the clamping force adjustment amount; it utilizes PID control and piezoelectric ceramic actuator to achieve rapid response adjustment of the clamping force, and continuously updates the model through incremental training, which not only ensures the accuracy and stability of clamping force control, but also adapts to changes in working conditions, effectively improving the measurement reliability of the magnetostrictive displacement sensor.
[0076] See Figure 6 The diagram shows a schematic of the electronic device 6 provided in an embodiment of the present invention, which is described in detail below: like Figure 6 As shown, the electronic device 6 of this embodiment includes a processor 60 and a memory 61. The memory 61 stores a computer program 62. When the processor 60 executes the computer program 62, it implements the steps in the various method embodiments described above. Alternatively, when the processor 60 executes the computer program 62, it implements the functions of each module in the various device embodiments described above.
[0077] For example, computer program 62 may be divided into one or more modules / units, which are stored in memory 61 and executed by processor 60 to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of computer program 62 in electronic device 6.
[0078] Electronic device 6 may include, but is not limited to, processor 60 and memory 61. Those skilled in the art will understand that... Figure 6 This is merely an example of electronic device 6 and does not constitute a limitation on electronic device 6. It may include more or fewer components than shown, or combine certain components, or different components. For example, electronic device 6 may also include input / output devices, network access devices, buses, etc.
[0079] The processor 60 can be a central processing unit (CPU), or 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 can be a microprocessor or any conventional processor.
[0080] The memory 61 can be an internal storage unit of the electronic device 6, such as a hard disk or RAM. The memory 61 can also be an external storage device of the electronic device 6, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card. Furthermore, the memory 61 can include both internal and external storage units of the electronic device 6. The memory 61 is used to store the computer program 62 and other programs and data required by the electronic device 6. The memory 61 can also be used to temporarily store data that has been output or will be output.
[0081] For the sake of simplicity and clarity, only the above-described functional modules / units are used as examples. In practical applications, the functions described above can be assigned to different functional modules / units as needed. These modules / units can be implemented in hardware, software, or a combination of both.
[0082] This invention also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the methods described in the above-described method embodiments.
[0083] This invention also provides a computer program product, including a computer program. When the computer program is executed by a processor, it implements the methods described in the above-described method embodiments.
[0084] Computer programs include computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. Computer-readable media can include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.
[0085] In the above embodiments, the descriptions of each embodiment have their own emphasis. Parts not detailed or described in a particular embodiment can be referred to in the relevant descriptions of other embodiments. Unless otherwise specified or in conflict with logic, the terminology and / or descriptions between different embodiments are consistent and can be referenced interchangeably. Technical features in different embodiments can be combined to form new embodiments based on their inherent logical relationships.
[0086] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.
Claims
1. A method for adjusting the clamping force of a magnetostrictive displacement sensor, characterized in that, include: Obtain the operating parameters of the magnetostrictive displacement sensor; wherein, the operating parameters include the target clamping force, as well as the measured clamping force and output displacement of the magnetostrictive displacement sensor; Based on the operating parameters, the predicted clamping force of the magnetostrictive displacement sensor is determined using a long short-term memory network model based on physical constraints. Calculate the fusion clamping force based on the measured clamping force and the output displacement; The clamping force adjustment amount of the magnetostrictive displacement sensor is determined based on the target clamping force, the predicted clamping force, and the fused clamping force; based on the clamping force adjustment amount, the clamping force of the magnetostrictive displacement sensor is adjusted by feedback through a piezoelectric ceramic actuator.
2. The method for adjusting the clamping force of a magnetostrictive displacement sensor according to claim 1, characterized in that, The operating parameters also include the material parameters and real-time temperature of the object being measured; The step of determining the predicted clamping force of the magnetostrictive displacement sensor based on the operating parameters and a long short-term memory network model with physical constraints includes: The output displacement, the real-time temperature, the measured clamping force, and the material parameters of the object under test are input into a pre-trained long short-term memory network model. The clamping force output by the long short-term memory network model is then corrected according to physical constraints to obtain the predicted clamping force.
3. The method for adjusting the clamping force of a magnetostrictive displacement sensor according to claim 2, characterized in that, The physical constraints include mechanical property constraints of the magnetostrictive displacement sensor, output force range constraints of the piezoelectric ceramic actuator, and material property constraints of the object being measured. The mechanical property constraint of the magnetostrictive displacement sensor is used to constrain the clamping force output by the long short-term memory network model to be within the allowable clamping force range corresponding to the mechanical property of the magnetostrictive displacement sensor. The piezoelectric ceramic actuator output force range constraint is used to constrain the clamping force output by the long short-term memory network model between the maximum and minimum values of the piezoelectric ceramic actuator output force; The material property constraint of the tested object is used to ensure that the deformation of the tested object is less than a preset threshold under the clamping force output by the long short-term memory network model.
4. The method for adjusting the clamping force of a magnetostrictive displacement sensor according to claim 1, characterized in that, The calculation of the fused clamping force based on the measured clamping force and the output displacement includes: Obtain the equivalent stiffness of the measurement system, and determine the equivalent clamping force based on the equivalent stiffness and the output displacement; The equivalent clamping force is temperature-compensated based on the real-time temperature to obtain the displacement clamping force. The weighted sum of the displacement clamping force and the measured clamping force is calculated to obtain the fused clamping force.
5. The method for adjusting the clamping force of a magnetostrictive displacement sensor according to claim 1, characterized in that, The step of determining the clamping force adjustment amount of the magnetostrictive displacement sensor based on the target clamping force, the predicted clamping force, and the fused clamping force includes: Using the target clamping force as a reference, the measurement deviation between the target clamping force and the fused clamping force, and the prediction deviation between the target clamping force and the predicted clamping force are calculated respectively. The clamping force adjustment amount is obtained by calculating the weighted sum of the measured deviation and the predicted deviation based on the preset deviation adjustment weight.
6. The method for adjusting the clamping force of a magnetostrictive displacement sensor according to any one of claims 1 to 5, characterized in that, The step of adjusting the clamping force of the magnetostrictive displacement sensor based on the clamping force adjustment amount via a piezoelectric ceramic actuator includes: The PID control signal of the piezoelectric ceramic actuator is determined based on the clamping force adjustment amount; The PID control signal of the piezoelectric ceramic actuator controls the clamping force of the magnetostrictive displacement sensor to be adjusted by feedback.
7. The method for adjusting the clamping force of a magnetostrictive displacement sensor according to any one of claims 1 to 5, characterized in that, The method further includes: Every preset time interval, the running parameters within the preset time interval are used as new samples to incrementally train the long short-term memory network model and update the model's weight parameters.
8. A clamping force adjustment device for a magnetostrictive displacement sensor, characterized in that, include: The parameter acquisition module is used to acquire the operating parameters of the magnetostrictive displacement sensor; wherein, the operating parameters include the target clamping force, as well as the measured clamping force and output displacement of the magnetostrictive displacement sensor; The clamping force prediction module is used to determine the predicted clamping force of the magnetostrictive displacement sensor based on the operating parameters and a long short-term memory network model with physical constraints. The fusion calculation module is used to calculate the fusion clamping force based on the measured clamping force and the output displacement; The clamping force adjustment module is used to determine the clamping force adjustment amount of the magnetostrictive displacement sensor based on the target clamping force, the predicted clamping force, and the fused clamping force; and to perform feedback adjustment of the clamping force of the magnetostrictive displacement sensor through a piezoelectric ceramic actuator based on the clamping force adjustment amount.
9. An electronic device, characterized in that, It includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method as described in any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method as described in any one of claims 1 to 7.