A multi-degree-of-freedom probe testing system with temperature and magnetic field synergic regulation

Abstract

The application provides a multi-degree-of-freedom probe test system with temperature and magnetic field synergic regulation, and relates to the technical field of probe test control.The implementation process of the application is as follows: probe test temperature adaptability inspection is performed, sample directional cold quantity adjustment and constant temperature maintenance mode are executed based on the inspection result, after the end of the probe test temperature adaptability inspection, probe temperature and magnetic coupling interference analysis is performed, and temperature and magnetic coupling interference synergic compensation measures are executed based on the analysis result, after the end of the probe temperature and magnetic coupling interference analysis, probe space drift online compensation is performed, after the end of the probe space drift online compensation, probe driving nonlinearity self-calibration is performed, and the technical problems of probe positioning deviation, contact pressure out of control and insufficient test accuracy can be solved.

Description

Technical Field

[0001] This invention relates to the field of probe testing and control technology, and in particular to a multi-degree-of-freedom probe testing system with coordinated temperature and magnetic field control. Background Technology

[0002] As the feature sizes of semiconductor chips, micro- and nanoelectronic devices, and advanced energy storage materials continue to evolve towards the nanoscale and atomic scale, traditional fixed probes or low-precision positioning methods can no longer meet the testing accuracy and reliability requirements of nanoscale, complex three-dimensional structures, heterojunction interfaces, and multi-physics field superposition environments. Probe testing technology has emerged to address this need. Existing probe testing systems mainly include the following core modules: a vacuum testing chamber, which provides a high-vacuum sealed environment to isolate external interference and ensure the stability of low-temperature and magnetic field control; a probe stage, which serves as the core platform for sample carrying and probe positioning, integrating temperature control modules such as heating rods and closed-loop cooling circuits, and magnetic field generating modules such as electromagnets and superconducting magnets, used to fix the sample to be tested and provide a basis for coordinated temperature and magnetic field control; a probe stage assembly, consisting of a stepper motor or piezoelectric ceramic drive mechanism and the probe body, completing the precise positioning of the probe and contact with the sample; and an external testing interface, used to achieve low-noise signal communication between the probe inside the chamber and external testing instruments such as source meters, network analyzers, and magnetometers.

[0003] The main implementation process of existing multi-degree-of-freedom probe testing methods is as follows: First, samples such as semiconductor wafers, magnetic thin films, and superconducting materials are fixed on the sample stage inside a vacuum testing chamber. Next, the probe stage is driven by a stepper motor or piezoelectric ceramic to perform coarse and fine adjustments to the probe's positioning. With the assistance of a high-magnification coaxial illumination microscope such as an optical microscope, the probe is initially aligned with the test points on the sample. Subsequently, the probe is finely adjusted based on control algorithms such as PID (Proportional-Integral-Derivative). Finally, external testing instruments, such as source meters, network analyzers, and magnetometers, are activated. Under preset temperature and magnetic field co-regulation conditions, the electrical, magnetic, and thermal multi-parameter signals of the sample under test are acquired and analyzed, and the performance evaluation results of the sample under test are output, such as the relationship between hysteresis loop and resistivity as a function of temperature / magnetic field.

[0004] The aforementioned existing technologies also have the following technical problems: Existing probe testing systems employ large-volume cooling devices such as closed-loop refrigerators or liquid helium Dewars and superconducting magnets in practical applications. This results in a massive overall heat sink, typically fixed in a laboratory, failing to meet the flexible requirements of on-site testing, rapid deployment, or in-situ characterization. While portability can be achieved by compressing the volume of the cooling device and testing chamber, the limited size of the probe testing system may lead to fluctuations in temperature and other conditions when using it to test samples such as superconducting materials and two-dimensional magnetic materials in extremely low-temperature environments. The problem of insufficient control precision is that due to the fluctuation of the extremely low temperature environment and the insufficient temperature control precision, the resistivity of the probe leads, contact electrodes and signal transmission paths may further cause time-varying drift. At the same time, when the temperature fluctuation is superimposed on the strong magnetic field, the strong magnetic field will induce parasitic Hall voltage, eddy current interference and electromagnetic coupling noise in the signal transmission circuit. Existing technologies usually use differential signal measurement, low-frequency AC phase-locked amplification and shielded twisted-pair transmission to suppress the superimposed electrical interference of the extremely low temperature and strong magnetic environment. This may cause residual noise related to temperature and magnetic changes to be mixed into the feedback signals of force sensors, piezoelectric ceramic displacement feedback and other sensors. Furthermore, during the initial alignment of the probe with the target point of the test sample based on the feedback signal containing residual noise, the residual noise in the feedback signal may obscure the true contact characteristics between the probe tip and the surface of the test sample, such as contact resistance abrupt changes and force feedback reference values. This can cause time-varying and nonlinear drift in the three-dimensional spatial position of the probe tip relative to the target point of the test sample. Existing technologies typically address this time-varying and nonlinear drift by using room-temperature calibrated static position coordinates combined with manual alignment using an optical microscope. This may further lead to cross-drift and hysteresis errors in temperature and magnetic regulation, resulting in errors in the mapping relationship between probe position feedback signals, such as displacement voltage signals from piezoelectric ceramic actuators and contact force signals detected by force sensors, and the actual spatial displacement. For example, the voltage-displacement curve of piezoelectric ceramics may be distorted due to magnetostriction, causing the PID controller to output incorrect drive voltage based on the distorted feedback signal. Consequently, when implementing multi-degree-of-freedom probe fine-tuning based on PID and other control algorithms, this can lead to problems such as probe positioning deviation, contact pressure runaway, and contact resistance fluctuations, ultimately resulting in low control and testing accuracy of the test sample.

[0005] In summary, existing probe testing systems cannot meet the flexible requirements of on-site testing and rapid deployment due to their large-volume cooling devices and superconducting magnets. While compressing the volume to achieve portability, they face problems such as fluctuations in extremely low-temperature environments, insufficient temperature control accuracy, and coupling interference issues caused by time-varying resistivity drift and electromagnetic noise induced by strong magnetic fields. Existing methods cannot completely suppress residual noise mixed into the feedback signal, leading to distortion of the piezoelectric ceramic voltage-displacement mapping and distorted drive voltage output of the PID controller. Ultimately, this results in probe positioning deviation, uncontrolled contact pressure, and low testing accuracy. Summary of the Invention

[0006] In view of this, embodiments of the present invention provide a multi-degree-of-freedom probe testing system with coordinated temperature and magnetic field control, comprising: a temperature adaptation and control module, a temperature and magnetic interference coordinated compensation module, and a probe control module. The temperature adaptation and control module is used to: perform a probe testing temperature adaptability test to evaluate the stability of the portable probe testing system in extremely low-temperature environments under volume-constrained conditions, and execute sample-oriented cooling adjustment and isothermal maintenance modes based on the test results. The temperature and magnetic interference coordinated compensation module is used to: perform quantitative analysis of probe temperature and magnetic coupling interference after the probe testing temperature adaptability test, and execute temperature and magnetic coupling interference coordinated compensation measures based on the analysis results. The probe control module is used to: perform online probe spatial drift compensation after the probe temperature and magnetic coupling interference quantitative analysis, and perform probe-driven nonlinear self-calibration after the online probe spatial drift compensation.

[0007] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following: 1. By conducting probe temperature compatibility testing and implementing sample-oriented cooling capacity adjustment and isothermal maintenance modes based on the test results, it is helpful to prioritize the temperature stability of the test area under the conditions of limited portable probe system size and insufficient cooling power. This enables on-demand distribution of cooling capacity and advanced temperature compensation, suppressing fluctuations in the extremely low temperature environment, thus providing a long-term stable foundation for subsequent probe testing in extremely low temperatures. After the probe temperature compatibility testing is completed, quantitative analysis of probe temperature-magnetic coupling interference is performed, and temperature-magnetic coupling interference collaborative compensation measures are implemented based on the analysis results. This helps to accurately distinguish between resistivity time-varying drift and strong... Magnetic field-induced electromagnetic noise and other types of interference are identified. Residual noise is reduced to contaminate subsequent contact feature identification and spatial positioning, thereby ensuring the accuracy and reliability of the feedback signal. After the quantitative analysis of probe temperature-magnetic coupling interference, online compensation for probe spatial drift is performed. After the online compensation for probe spatial drift, probe drive nonlinearity self-calibration is performed. This helps to dynamically correct spatial displacement deviations caused by thermal drift and noise masking, so that the probe positioning and drive mapping relationship can adapt to environmental changes in real time. This fundamentally solves the problems of positioning deviation, pressure runaway and low test accuracy caused by residual noise, spatial drift and drive nonlinearity.

[0008] 2. Targeted acquisition of resistivity time-varying drift characteristic indicators helps overcome the shortcomings of existing technologies that cannot distinguish between resistivity drift and electromagnetic noise using only differential measurement and lock-in amplification. The resistivity time-varying drift characteristic indicators can quantify the linear correlation between resistivity and temperature fluctuations, extracting amplitude distortion that was originally mixed in the feedback signal. This reduces unnecessary compensation calculations when there is no significant drift. The resistivity time-varying drift characteristic indicators are compared with preset drift tolerance reference values, and the temperature-magnetic interference characteristic indicators are compared with preset temperature-magnetic interference reference values. If both the resistivity time-varying drift characteristic indicators and the temperature-magnetic interference characteristic indicators are less than the corresponding interference comparison reference values, online probe space drift compensation is performed; otherwise, temperature-magnetic coupling interference collaborative compensation measures are adopted. This helps solve the problem of insufficient or over-compensation caused by the use of the same fixed filtering scheme for all interferences in existing technologies, reducing the introduction of additional filtering delays or amplitude distortion, and achieving a balance between interference suppression accuracy and real-time performance.

[0009] 3. When the magnetic field strength changes rapidly over time, such as when the magnetic field scanning rate exceeds the preset rate threshold during a field sweep test, the voltage-displacement mapping of the piezoelectric ceramic exhibits time-varying nonlinear characteristics due to the coupling effect of magnetostriction and dynamic eddy current damping. The recursive least squares linear model is unable to track its dynamic changes in real time, which may lead to lag in feedforward compensation. Therefore, an alternative scheme of probe-driven nonlinear self-calibration is required. The preset expected displacement and magnetic field strength values ​​are input into a preset single-hidden-layer feedforward neural network, which outputs the corresponding open-loop driving voltage. This helps to overcome the shortcomings of existing technologies in tracking the time-varying nonlinearity of voltage and displacement in real time under strong magnetic field dynamic changes, reduces the compensation delay caused by model lag, and improves the real-time performance and accuracy of feedforward compensation during rapid field sweep. The open-loop driving voltage is superimposed on the output of the PID controller to compensate for the dynamic distortion under rapid field sweep. This helps to solve the problem that traditional PID controllers rely solely on feedback adjustment under strong magnetic field dynamic changes, resulting in slow response speed and easy oscillation. By reducing predictable nonlinear distortion in advance through feedforward compensation, the dynamic response speed and tracking accuracy of probe positioning can be improved while ensuring system stability. Attached Figure Description

[0010] Figure 1 This is a schematic diagram of the structure of a multi-degree-of-freedom probe testing system with coordinated temperature and magnetic field control provided in an embodiment of the present invention; Figure 2 This is a flowchart outlining the overall process of a multi-degree-of-freedom probe testing system with coordinated temperature and magnetic field control, as provided in an embodiment of the present invention. Figure 3 This is a logic diagram of a temperature and magnetic field coordinated control multi-degree-of-freedom probe testing system provided in an embodiment of the present invention, illustrating the coordinated compensation measures for temperature and magnetic field coupling interference. Figure 4 This is a diagram of a portable probe testing system device provided by a multi-degree-of-freedom probe testing system with coordinated temperature and magnetic field control according to an embodiment of the present invention. Figure 5 This is a schematic diagram of the overall structure of a portable probe testing system device provided by a multi-degree-of-freedom probe testing system with coordinated temperature and magnetic field control according to an embodiment of the present invention. Figure 6 This is a flowchart of the sample loading operation of a multi-degree-of-freedom probe testing system with coordinated temperature and magnetic field control provided in an embodiment of the present invention. Figure 7 This is a line graph of the temperature rise test of a multi-degree-of-freedom probe testing system with coordinated temperature and magnetic field control provided in an embodiment of the present invention. In the diagram: 1. Probe station; 2. Vacuum valve; 3. External test interface; 4. Sample to be tested; 5. Sealing cover; 6. Glove box. Detailed Implementation

[0011] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0012] It should be understood that the multi-degree-of-freedom probe testing system for coordinated temperature and magnetic field control provided in this application relies on a pre-constructed ultra-low temperature and high magnetic field testing parameter library. This library includes core judgment parameters verified by testing experts, such as preset ultra-low temperature values, preset temperature deviation reference values, preset fluctuation reference values, and preset stability reference values. Simultaneously, the library also includes historical test data of various quantum materials and superconducting devices that have undergone standardized preprocessing. This data covers temperature fluctuation time-series data, feedback signal waveforms, interference characteristic indicators, and corresponding compensation measure execution records and model iteration optimization data under different ultra-low temperature ranges, different magnetic field intensities, and different probe contact modes, such as voltage-displacement mapping parameters identified by recursive least squares. The library includes components such as the separation matrix for independent component analysis and the state-space model of the model predictive controller. It employs a domain-specific heterogeneous storage architecture, utilizing a time-series database to store continuously sampled sensor data such as temperature, magnetic field, resistance, and force feedback. Simultaneously, a key-value pair database stores discrete judgment thresholds and model parameters. Based on the probe positioning accuracy and signal quality feedback during real-time testing, personnel can periodically verify and dynamically adjust the various preset parameters stored in the library. This ensures that the library can continuously adapt to the differences in temperature and magnetic properties of different test samples and changes in testing conditions, such as precise temperature scanning tests near the critical temperature of superconducting materials and hysteresis loop scanning of two-dimensional magnetic materials under strong magnetic fields, meeting practical application needs.

[0013] It should also be understood that the specific preset values ​​in the multi-degree-of-freedom probe testing system with coordinated temperature and magnetic field control provided in this application embodiment, such as preset ultra-low temperature target values, preset maximum cooling power, preset maximum compensation power, preset fixed heating ratio, etc., are constructed and obtained through a unified offline calibration and online verification process: First, based on the physical characteristics of the sample to be tested, such as the superconducting critical temperature, and the hardware specifications of the probe testing system, such as the rated power of the cooling unit and the maximum output of the heating device, the initial suggested range of the preset values ​​is initially set by the testing experts; then, under standard environmental conditions, i.e., a vacuum degree better than 1e-5 mbar, no external magnetic field, and an ambient temperature of 300 K, an unloaded test is conducted, and the relevant parameters are gradually adjusted. The temperature response curve and stability of the sample stage are monitored. By analyzing indicators such as temperature overshoot, steady-state error, and adjustment time, the initially set preset values ​​are iteratively optimized using the golden section search method or gradient descent method until the system reaches the preset performance indicators, such as temperature fluctuation less than ±0.1. K and the adjustment time is less than 5 minutes, etc.; the optimized parameter values ​​are stored in the ultra-low temperature strong magnetic field test parameter library, and the corresponding test conditions are recorded, such as the thermal conductivity and heat capacity of the sample; thereafter, the above calibration process is repeated every time the sample type or system hardware is changed, and the preset values ​​are fine-tuned and adaptively updated through online monitoring data in actual tests to ensure that the preset values ​​are strictly matched with the current conditions.

[0014] Meanwhile, the specific preset ranges provided in the temperature and magnetic field coordinated control multi-degree-of-freedom probe testing system of this application embodiment, such as preset contact range, preset steady-state power range, preset ultra-low temperature range, etc., are constructed and obtained through a general process based on statistical analysis and boundary testing: First, the physical lower and upper limits of the range to be set are determined according to theoretical models or prior knowledge. For example, the lower limit of the preset contact range is determined by the radius of curvature of the probe tip, and the upper limit is determined by the signal attenuation requirements. Then, under test conditions, the range parameters are sampled at equal or logarithmic intervals, and multiple repeated tests are performed at each sampling point to record the system response, such as contact resistance stability, temperature fluctuation amplitude, temperature control success rate, etc. The mean and standard deviation of each test index are calculated, and outliers are identified and eliminated using the box plot method. The system response is required to meet preset qualification conditions, such as contact resistance relative standard deviation ≤5% and temperature fluctuation ≤±0.1. The parameter intervals corresponding to all sampling points of K are merged, and the convex hull or the smallest continuous coverage interval of the merged interval is taken as the preset range. The upper and lower boundaries of the range are stored in the ultra-low temperature strong magnetic field test parameter library, and a confidence level, such as a 95% confidence interval, is added. In practical applications, the system can automatically determine whether the current working condition falls within the preset range based on real-time feedback. If it exceeds the range, adaptive adjustment is triggered or a prompt is issued to ensure the reliability and repeatability of the test.

[0015] Example 1: This embodiment of the invention provides a multi-degree-of-freedom probe testing system with coordinated temperature and magnetic field control. For example... Figure 1 The schematic diagram of a multi-degree-of-freedom probe testing system with coordinated temperature and magnetic field control is shown. It includes a temperature adaptation and control module, a temperature and magnetic interference compensation module, and a probe control module. The temperature adaptation and control module is used to: during probe testing, perform a probe testing temperature adaptability test to evaluate the stability of the portable probe testing system in extremely low-temperature environments under volume-constrained conditions; and based on the test results, execute sample-oriented cooling capacity allocation to achieve directional cooling capacity distribution and advanced temperature compensation in the test area of ​​the sample under test, and a constant-temperature maintenance mode to maintain the long-term stability of the target temperature of the test environment. Through the temperature adaptation and control module, under conditions of limited volume and insufficient cooling power of the portable probe testing system, the fluctuation amplitude of the extremely low-temperature environment can be effectively suppressed, temperature control accuracy can be improved, and a long-term stable extremely low-temperature environment foundation can be provided for subsequent probe testing.

[0016] The temperature-magnetic interference co-compensation module is used to: after the probe test temperature adaptability verification is completed, perform probe temperature-magnetic coupling interference quantitative analysis to quantify the degree of interference from resistivity time-varying drift and electromagnetic noise induced by strong magnetic fields, and implement temperature-magnetic coupling interference co-compensation measures based on the analysis results to reduce resistivity drift signal amplitude distortion and electromagnetic coupling frequency band noise; through the temperature-magnetic interference co-compensation module, it can accurately distinguish between two different types of interference, resistivity time-varying drift and electromagnetic noise induced by strong magnetic fields, and selectively reduce amplitude distortion and frequency band noise in sequence, reducing residual noise pollution in subsequent contact feature identification and spatial positioning, thereby ensuring the accuracy and reliability of the feedback signal.

[0017] The probe control module is used for: after the quantitative analysis of probe temperature-magnetic coupling interference, to perform online compensation for probe spatial drift caused by residual noise masking probe contact characteristics and to dynamically correct the probe zero point; after the online compensation for probe spatial drift, to perform probe drive nonlinear self-calibration to correct voltage and displacement nonlinear distortion under strong magnetic fields; through the probe control module, precise positioning of the probe and closed-loop control of contact force can be achieved, fundamentally solving the problems of positioning deviation, pressure runaway and low test accuracy caused by residual noise, spatial drift and drive nonlinearity.

[0018] In this embodiment, a complete closed loop from environmental stabilization to signal purification and then to probe control can be formed through the temperature adaptation and control module, the temperature-magnetic interference co-compensation module, and the probe control module. Each module is progressive and interdependent. Specifically, the temperature adaptation and control module provides a temperature-stable test environment for subsequent signal processing, reducing the interference of temperature drift on the feedback signal. The temperature-magnetic interference co-compensation module further reduces resistivity drift and electromagnetic noise in the signal link on the basis of a stable environment, ensuring the authenticity of the feedback signal. The probe control module uses high-accuracy signals to achieve spatial drift compensation and nonlinear self-calibration, ultimately completing high-precision probe positioning and contact force control.

[0019] like Figure 2 The flowchart shown is an overview of a multi-degree-of-freedom probe testing system with coordinated temperature and magnetic field control. Figure 2 It is known that: the probe temperature adaptability test is performed, and the temperature control adaptability prediction result is obtained. It is determined whether the temperature control adaptability prediction result meets the low temperature stability discrimination condition. If it does, the isothermal maintenance mode is triggered, and the probe temperature-magnetic coupling interference is quantitatively analyzed. Otherwise, the sample directional cooling capacity adjustment is triggered. After the sample directional cooling capacity adjustment is completed, it is determined whether the probe temperature adaptability test is qualified. If not, an ultra-low temperature control failure prompt is sent. Otherwise, the isothermal maintenance mode is triggered, and the probe temperature-magnetic coupling interference is quantitatively analyzed. The resistivity time-varying drift characteristic index and the temperature-magnetic interference characteristic index are obtained. It is determined whether the resistivity time-varying drift characteristic index is less than the preset drift tolerance reference value, and whether the temperature-magnetic interference characteristic index is less than the preset temperature-magnetic interference reference value. If not, temperature-magnetic coupling interference collaborative compensation measures are taken. Otherwise, probe spatial drift online compensation is performed. After the probe spatial drift online compensation is completed, probe-driven nonlinear self-calibration is performed.

[0020] Furthermore, the specific process for probe temperature compatibility testing is as follows: Acquire the temperature test data of the test area of ​​the sample to be tested. The test area refers to a circular or square local area with a diameter not exceeding a preset contact range (e.g., 2mm) around the actual contact point between the probe tip and the sample surface during probe testing. This test area is a key area for acquiring electrical, magnetic, or thermal signals, and the preset contact range is set in advance by designated personnel. The temperature test data includes temperature distribution data and cooling device operating status data. The temperature distribution data includes ambient temperature, regional temperature gradient, and temperature fluctuation amplitude. The cooling device operating status data... This includes the output power of the cooling device and the response rate of heat exchange; the ambient temperature represents the average temperature value of various monitoring points within the test chamber of the portable probe testing system, used as a benchmark reference for temperature distribution data, reflecting the basic temperature field state of the test environment; the regional temperature gradient is represented by calculating the absolute value of the difference between the instantaneous temperature value at each point and the instantaneous temperature value at adjacent points, then calculating the ratio of the result of the absolute value of the difference calculation to the distance between the corresponding adjacent points, and finally averaging all the ratio calculation results; the temperature fluctuation amplitude is represented by calculating the difference between the highest and lowest temperature values ​​within the test chamber of the portable probe testing system during the temperature sampling period. The thermal exchange response rate is represented by calculating the difference between the current temperature and the initial temperature within the test chamber of the portable probe testing system, and then dividing the result by the adjustment time. The adjustment time represents the total time required for the cooling device to reach the current temperature in the sample test area after receiving a temperature control command and initiating cooling operation, reflecting the response efficiency and timeliness of temperature control of the cooling device. The temperature test data of the sample under test undergoes noise reduction and normalization preprocessing to obtain preprocessed standardized sample temperature field adaptation data. Noise reduction and normalization preprocessing involves removing abnormal jump values ​​from the temperature test data of the sample under test based on the 3σ criterion, and standardizing the data based on the Z-score normalization method. Sample temperature test data is normalized to eliminate differences in the dimensions of different parameters. Standardized sample temperature field adaptation data is input into preset temperature control correlation models such as random forest regression models or long short-term memory time-series prediction models. The models output predicted temperature deviation values ​​for a future period. Based on these predicted temperature deviation values, temperature control adaptation prediction results are obtained, including the temperature deviation coefficient of the test area of ​​the sample under test, the predicted temperature fluctuation trend, and the ultra-low temperature stability evaluation index. The temperature deviation coefficient is represented by dividing the predicted temperature deviation value by the preset ultra-low temperature target value. The predicted temperature fluctuation trend is represented by averaging the absolute values ​​of the predicted temperature deviation values ​​over a future period.The ultra-low temperature stability assessment index uses the root mean square error (RMSE) of the predicted temperature deviation over a future period compared to the preset temperature deviation. The temperature deviation coefficient quantifies the deviation between the real-time temperature of the sample test area and the preset ultra-low temperature value, reflecting the deviation in temperature control accuracy. The predicted temperature fluctuation trend value predicts the temperature change trend and fluctuation amplitude of the sample test area within the predicted time period. The ultra-low temperature stability assessment index evaluates the stability of the ultra-low temperature environment in the sample test area; a higher value indicates a more stable ultra-low temperature environment.

[0021] The temperature control adaptation status is determined based on the temperature control adaptation prediction results. The specific process is as follows: It is determined whether the temperature control adaptation prediction results meet the low temperature stability discrimination conditions. If so, the constant temperature maintenance mode is triggered, and probe temperature-magnetic coupling interference quantification analysis is performed. Otherwise, the sample directional cooling capacity adjustment is triggered. The low temperature stability discrimination conditions indicate that the temperature deviation coefficient is less than the preset temperature deviation reference value, the temperature fluctuation trend prediction value is less than the preset fluctuation reference value, and the ultra-low temperature stability evaluation index is greater than the preset stability reference value. Among them, the preset temperature deviation reference value is represented by the average value of the temperature deviation coefficient over a historical period, the preset fluctuation reference value is represented by the average value of the temperature fluctuation trend prediction value over a historical period, and the preset stability reference value is represented by the average value of the ultra-low temperature stability evaluation index over a historical period.

[0022] It should be noted that the preset temperature control correlation model used in this embodiment is specifically a random forest regression model. The specific training process is as follows: First, the standardized sample temperature field adaptation data is divided into a training set, a validation set, and a test set according to time order. The feature vector of each sampling point includes the current ambient temperature, regional temperature gradient, temperature fluctuation amplitude, cooling device output power, and heat exchange response rate, totaling five feature dimensions. Next, the standardized sample temperature field adaptation data within a preset time window in the training set is used as input, and the predicted temperature deviation value for a future period is used as the output label. The random forest regression algorithm is used for training. A preset number of subsets are generated with replacement using the Bootstrap method. The size of each subset is the same as the merged set. A decision tree is constructed for each subset. When splitting at each node of each tree, ... From the five features mentioned above, a preset number of features are randomly selected. The feature that minimizes the sum of the mean squared errors of the left and right child nodes after splitting, along with the splitting threshold, is chosen. The model is then recursively grown until the maximum depth or the number of leaf node samples is reached. The arithmetic mean of the temperature deviation predictions of all decision trees is used as the output of the random forest model. Then, the test error is obtained. If the test error is less than the preset accuracy threshold, the model is serialized and saved to the database. If the test error is not less than the preset accuracy threshold, the preset time window length or preset prediction steps are adjusted, and the above training process is re-executed until the accuracy requirements are met. The test error is represented by the result of the difference between the temperature deviation prediction and the true value. Finally, the trained model is used for real-time testing. The preset time window and preset number are set in advance by preset personnel, and the preset accuracy threshold is represented by the average error over a historical time period.

[0023] In this embodiment, the probe test temperature adaptability test helps to evaluate the stability of the portable probe test system in extremely low temperature environment under volume-constrained conditions in real time, reduce the risk of temperature overshoot caused by insufficient cooling power or heat leakage fluctuations, realize the transformation from passive constant temperature to active prediction temperature control mode, and thus control the temperature fluctuation range of the sample test area within a reasonable range under limited cooling capacity.

[0024] Furthermore, the specific process of sample-oriented cooling capacity adjustment is as follows: The temperature control adaptation prediction result is input into a preset temperature adjustment mapping table for querying, obtaining the cooling power adjustment step size and the heating compensation power adjustment step size. The cooling power adjustment step size represents the adjustment range of the cooling device's output power within a single control cycle, used to achieve step-by-step adjustment of the cooling power and reduce drastic temperature fluctuations caused by sudden power changes. The heating compensation power adjustment step size represents the adjustment range of the heating device's output compensation power within a single control cycle, used to achieve fine-grained adjustment of heating compensation and reduce deviations caused by temperature fluctuations. The cooling device's output power is adjusted step-by-step in the direction of decreasing temperature deviation coefficient using the cooling power adjustment step size, while the heating compensation power adjustment step size is adjusted step-by-step in the direction of decreasing predicted temperature fluctuation trend, achieving proactive compensation for temperature fluctuations. The temperature control adaptation prediction result is continuously monitored; if the temperature control adaptation prediction result meets the low-temperature stability criteria... If other conditions are not met, the current cooling power and heating compensation power will be maintained for temperature adjustment. When the temperature of the sample test area is within the preset ultra-low temperature range within a preset number of control cycles, and the temperature control adaptation prediction result meets the low temperature stability discrimination condition, the constant temperature maintenance mode will be triggered, and probe temperature-magnetic coupling interference quantitative analysis will be performed. Otherwise, the sample directional cooling capacity adjustment will continue. If the output power of the cooling device is greater than the preset maximum cooling power, or the compensation power of the heating device is greater than the preset maximum compensation power, and the temperature control adaptation prediction result still does not meet the low temperature stability discrimination condition, an ultra-low temperature control failure prompt will be sent, and the operating status data of the cooling device and heating device, such as the current output power, temperature control adaptation prediction result, power adjustment times, and cumulative control time, will be sent to the probe test data control center to facilitate the staff to troubleshoot the fault and optimize the temperature control parameters. The preset maximum cooling power and preset maximum compensation power are both preset by the preset personnel.

[0025] Furthermore, the specific process of the constant temperature maintenance mode is as follows: The temperature deviation value and temperature change rate are input into the PID controller to obtain the temperature-coordinated control quantity for constant temperature regulation of the ultra-low temperature field; the temperature-coordinated control quantity represents a dimensionless normalized control signal, with a value range of [-1, 1]. A positive value indicates that the current temperature is lower than the preset ultra-low temperature target value, and a negative value indicates that the current temperature is higher than the preset ultra-low temperature target value. The preset ultra-low temperature target value is set in advance by preset personnel; the temperature deviation value is represented by the result of the difference calculation between the current instantaneous temperature value and the preset ultra-low temperature target value; the temperature change rate during the temperature sampling period is obtained based on the central difference method, and the specific calculation formula is as follows:

[0026] Where T(t) i ) represents the rate of temperature change at sampling time i, t i T(t) represents the time point of the i-th sampling time.i+1 T(t) represents the instantaneous temperature value at the (i+1)th sampling time. i-1 ) represents the instantaneous temperature value at the (i-1)th sampling time, and Δt is a fixed sampling time interval.

[0027] The temperature-coordinated control signal is sent to the probe test data control center, outputting cooling power adjustment commands and heating power adjustment commands. The cooling power adjustment command is used to adjust the control signal of the cooling device's output power to control the amount of cooling. The heating power adjustment command is used to adjust the control signal of the heating device's compensation power to supplement heat and offset the overshoot of cooling. Based on the cooling power adjustment command, a constant output constraint is applied to the cooling device to suppress the excessive cooling trend in the extremely low temperature environment. The constant output constraint means setting the output power of the cooling device to a preset steady-state power range corresponding to the current temperature deviation. When the temperature deviation is within the preset steady-state power range, the cooling power is forced to remain unchanged. Based on the heating power adjustment command, a micro-balanced temperature compensation adjustment is performed on the heating device to reduce environmental heat intrusion, device heating, and localization caused by magnetic environment coupling. Temperature drift; Micro-balanced temperature compensation adjustment means that the output power of the heating device is set to a preset fixed heating ratio of the current power of the cooling device. The heating output is only activated when the temperature deviates from the preset ultra-low temperature target value and the direction of deviation is opposite to the direction of the cooling power. The heating power is applied in the form of discrete pulses; The application of heating power in the form of discrete pulses means that the driving signal of the heating device consists of a series of voltage pulse sequences with fixed time width and fixed amplitude. Each pulse corresponds to a preset fixed heating energy, such as a single pulse providing 0.01J of heat. By increasing or decreasing the number of pulses, the heating energy is discretized and stepped, thereby reducing the temperature overshoot or oscillation that may be caused by continuous power output; The constant temperature maintenance mode runs continuously until the test task of the sample under test is completed or a new temperature setting command is received.

[0028] It is important to note that the preset temperature adjustment mapping table on which the sample-oriented cooling capacity adjustment relies in this embodiment is pre-constructed by ultra-low temperature testing experts in conjunction with massive amounts of historical temperature control data and stored in the ultra-low temperature strong magnetic field test parameter library. When the system performs sample-oriented cooling capacity adjustment, it can directly import the temperature control adaptation prediction results, including the temperature deviation coefficient, the predicted value of temperature fluctuation trend, and the ultra-low temperature stability evaluation index, into the mapping table, quickly match and output the corresponding cooling power adjustment step size and heating compensation power adjustment step size, and ensure that the cooling capacity adjustment is accurately adapted to the current temperature control requirements.

[0029] Specifically, the preset temperature regulation mapping table adopts a three-dimensional relational table structure. The horizontal dimension of the table is marked as the temperature deviation coefficient range. For example, the range where the temperature deviation coefficient is not less than the preset temperature deviation reference value is marked as the temperature deviation range, and the range where the temperature deviation coefficient is less than the preset temperature deviation reference value is marked as the temperature stable range. The vertical dimension is marked as the temperature fluctuation trend prediction value range. For example, the range where the temperature fluctuation trend prediction value is greater than the preset first-level fluctuation reference value is marked as the high temperature fluctuation range, and the range where the temperature fluctuation trend prediction value is not greater than the preset first-level fluctuation reference value and not less than the preset fluctuation reference value is marked as the low temperature fluctuation range. The intervals are categorized as follows: intervals where the predicted temperature fluctuation trend value is less than the preset fluctuation reference value are marked as stable temperature intervals, etc. The depth dimension is marked as the corresponding intervals for the ultra-low temperature stability assessment indicators. For example, intervals where the ultra-low temperature stability assessment indicator is not greater than the preset stability reference value are marked as low indicator intervals, and intervals where the ultra-low temperature stability assessment indicator is greater than the preset stability reference value are marked as high indicator intervals, etc. The preset first-level fluctuation reference value is set in advance by designated personnel and is greater than the preset fluctuation reference value. The three-dimensional cross-node data in the table represent the cooling power adjustment step size and heating compensation power adjustment step size under the corresponding parameter combinations; mapping. The table construction process fully relies on multi-dimensional historical temperature control data resources, covering temperature control parameter combinations under different temperature deviations, fluctuation trends, and stability levels. Each parameter group is assigned a comprehensive quantitative score based on the temperature deviation coefficient and fluctuation trend. The comprehensive quantitative score is obtained by first mapping the fluctuation trend to a pre-defined interval and assigning a corresponding coefficient. For example, if the fluctuation trend is in a high temperature fluctuation range, the corresponding coefficient is 0.8. Then, the temperature deviation coefficient, temperature overshoot coefficient, and fluctuation trend are multiplied to obtain the comprehensive quantitative score. For example, when the temperature deviation coefficient is in the temperature deviation range and the fluctuation trend is in the high temperature fluctuation range... To quickly suppress the temperature rise trend, a larger cooling power adjustment step size and a smaller heating compensation step size are matched. At the same time, the actual application effect data of the adjustment step size in each historical temperature control scenario are recorded. Correlation analysis methods, such as Pearson correlation coefficient, are used to calculate the correlation coefficient of each parameter combination. This eliminates abnormal parameter combinations with correlation coefficients less than the preset correlation threshold caused by sensor instantaneous failure, electromagnetic interference, etc., and retains the parameter correspondence with statistical significance. This ensures the stability of the mapping table output results and the adaptability to the temperature control scenario. The preset correlation threshold is identified by the average value of the correlation coefficients of the parameter combinations corresponding to the historical time period.

[0030] In this embodiment, by using sample-oriented cooling capacity adjustment and constant temperature maintenance mode, it is helpful to prioritize the supply of cooling capacity to the core test area under the limited cooling power of the portable probe testing system, reduce temperature control redundancy caused by cooling capacity consumption in non-core areas, reduce temperature fluctuation amplitude and overshoot frequency, improve the long-term stability of the ultra-low temperature environment, and achieve precise control of the temperature of the test area of ​​the sample to be tested and efficient energy consumption management, thereby providing a stable and uniform ultra-low temperature field foundation for subsequent probe testing.

[0031] Further, the probe temperature-magnetic coupling interference quantification analysis is performed as follows: Step S11: Collect the DC resistance values ​​of the probe leads, contact electrodes, and both ends of the signal transmission path; simultaneously collect the instantaneous temperature value and current magnetic field strength value of the test area of ​​the sample under test; construct a probe temperature-magnetic multidimensional sequence including timestamps. For example, the probe temperature-magnetic multidimensional sequence is {[R(t1),T(t1),B(t1)],[R(t2),T(t2),B(t2)],…,[R(tN),T(tN),B(t1)],…,[R(tN),T(tN),B(t1)],…,[R(tN),T(tN),B(t1)],…,[R(tN),T(tN),B(tN)],…,[R(tN),T(tN),B(tN)],…,[R(tN),T(tN),T(tN),B(tN)],…,[R(tN),T(tN),T(tN),T(tN)],…,[R ... (tN]}, where R(tq) is the DC resistance value measured at the q-th sampling time, in ohms, T(tq) is the instantaneous temperature value collected synchronously, in Kelvin, B(tq) is the magnetic field strength value collected synchronously, in Tesla, q=1,2,...,N, q represents the sequence number of the current sampling point, and N represents the total number of sampling points; Step S12: Perform sliding window linear fitting on the DC resistance value based on the sliding window least squares linear fitting algorithm to obtain the resistance change rate; Based on Pearson correlation The Pearson correlation coefficient between the resistivity change rate sequence and the temperature change rate sequence is obtained using a numerical algorithm. This coefficient serves as a time-varying resistivity drift characteristic index, used to quantify the linear correlation between resistivity and temperature fluctuations. The value range is [-1, 1]. The closer the absolute value of the resistivity drift characteristic index is to 1, the more temperature-driven the resistivity drift. Power spectral density is estimated from the original signals of the feedback loops from the force sensor and piezoelectric ceramics using the Welch periodogram method, yielding the power spectral density function. Within a preset interference-sensitive frequency band, the interference energy concentration coefficient is calculated. This coefficient is used as a temperature-magnetic interference characteristic index. A larger index indicates that the temperature-magnetic noise energy is more concentrated within the preset interference-sensitive frequency band, resulting in more severe pollution of the feedback signal. The preset interference-sensitive frequency band refers to the frequency bands covered by integer multiples of the carrier frequency of the force sensor and the integer multiples of the piezoelectric ceramic drive scanning frequency. These bands represent the areas where temperature-magnetic noise energy is most concentrated. The specific calculation formula for the interference energy concentration coefficient is as follows:

[0032] Where η represents the interference energy concentration coefficient, and P(f) is the power spectral density function estimated by the Welch periodogram method, with units of V. 2 / Hz, flow and f high These are the upper and lower boundary frequencies of the preset interference-sensitive frequency band, in Hz and f. max The highest frequency is half of the original signal sampling frequency. The numerator represents the total noise energy within the preset interference-sensitive frequency band, and the denominator represents the total energy of the original voltage signal across the entire frequency band.

[0033] Step S13: Compare the resistivity time-varying drift characteristic index and the temperature-magnetic interference characteristic index with the corresponding interference comparison reference values; if both the resistivity time-varying drift characteristic index and the temperature-magnetic interference characteristic index are less than the corresponding interference comparison reference values, it is determined that the current signal link is not subject to significant resistivity drift and temperature-magnetic interference, and online compensation for probe spatial drift is performed. The preset drift tolerance reference value is represented by the average value of the preset temperature-magnetic interference reference values ​​over a historical time period, and the preset temperature-magnetic interference reference value is represented by the average value of the temperature-magnetic interference characteristic index over a historical time period; otherwise, it is determined that there is temperature-magnetic coupling interference in the current probe test environment, and temperature-magnetic coupling interference collaborative compensation measures are taken; the interference comparison reference values ​​include the preset drift tolerance reference value and the preset temperature-magnetic interference reference value.

[0034] In this embodiment, quantitative analysis of probe temperature-magnetic coupling interference helps to accurately distinguish between two different types of interference: resistivity time-varying drift and electromagnetic noise induced by strong magnetic fields. This reduces the risk of misjudgment of compensation strategies due to interference aliasing, reduces unnecessary compensation calculation overhead, improves the pertinence and execution efficiency of subsequent collaborative compensation measures, and enables rapid identification and quantitative characterization of interference types.

[0035] like Figure 3 The diagram shown is a logic diagram of a temperature and magnetic field coordinated control multi-degree-of-freedom probe testing system with coordinated temperature and magnetic field control, which is composed of... Figure 3 It can be seen that: when implementing temperature-magnetic coupling interference collaborative compensation measures, if the resistivity time-varying drift characteristic index is less than the preset drift tolerance reference value, and the temperature-magnetic interference characteristic index is not less than the preset temperature-magnetic interference reference value, then a temperature-magnetic adaptive notch filter collaborative filtering strategy is adopted. If the resistivity time-varying drift characteristic index is not less than the preset resistance drift tolerance reference value, and the temperature-magnetic interference characteristic index is less than the preset temperature-magnetic interference reference value, then resistivity time-varying adaptive compensation is adopted. If the resistivity time-varying drift characteristic index is not less than the preset drift tolerance reference value, and the temperature-magnetic interference characteristic index is not less than the preset temperature-magnetic interference reference value, then resistivity time-varying adaptive compensation and temperature-magnetic adaptive notch filter collaborative filtering strategies are performed in sequence.

[0036] Furthermore, the specific process of the temperature-magnetic coupling interference collaborative compensation measure is as follows: If the resistivity time-varying drift characteristic index is less than the preset drift tolerance reference value, and the temperature-magnetic interference characteristic index is not less than the preset temperature-magnetic interference reference value, it indicates that there is currently only electromagnetic coupling interference induced by a strong magnetic field, such as parasitic Hall voltage, eddy current interference, electromagnetic coupling noise, etc., and no significant resistivity drift. In this case, a temperature-magnetic adaptive notch filter collaborative filtering strategy is adopted. If the resistivity time-varying drift characteristic index is not less than the preset resistance drift tolerance reference value, and the temperature-magnetic interference characteristic index is less than the preset temperature-magnetic interference reference value, it indicates that there is currently only resistivity time-varying drift interference dominated by extremely low temperatures, and no significant electromagnetic coupling interference. In this case, a resistivity time-varying drift characteristic index is adopted. Adaptive compensation; if the resistivity time-varying drift characteristic index is not less than the preset drift tolerance reference value, and the temperature and magnetic interference characteristic index is not less than the preset temperature and magnetic interference reference value, it indicates that there is currently both resistivity time-varying drift interference dominated by extremely low temperature and electromagnetic coupling interference induced by strong magnetic field. Then, the resistivity time-varying adaptive compensation and temperature and magnetic adaptive notch filtering collaborative filtering strategy are performed in sequence. This helps to reduce the overall distortion of signal amplitude caused by resistivity drift first, so that the subsequent temperature and magnetic adaptive notch filtering can suppress band noise based on more accurate signal amplitude, improve the stability and convergence speed of the two interference collaborative compensation, and ensure that the feedback signal is restored to a clean state in both amplitude and spectrum dimensions.

[0037] To address resistivity time-varying drift interference dominated by extremely low temperatures, resistivity time-varying adaptive compensation is implemented, dynamically generating resistivity compensation coefficients for feedback signal amplitude correction. These coefficients are dynamically generated as follows: the probe path resistance is obtained based on the total resistance of the current test loop; the probe path resistance is represented by the difference between the total test loop resistance and the preset loop resistance; the total test loop resistance is represented by the division between the excitation current value monitored by the current source and the voltage value monitored by the nanovoltmeter; the preset loop resistance is pre-set by designated personnel; the ratio of the nominal resistance value to the probe path resistance is used as the resistivity compensation coefficient; the nominal resistance value is the initial resistance value of the probe leads, contact electrodes, and signal transmission path under standard room temperature conditions, obtained through the initial measurement before testing; and the amplitude corresponding to the resistivity compensation coefficient is used as the coefficient. As an adjustment step size, the voltage amplitude of the feedback signal is gradually adjusted in the direction that the resistivity compensation coefficient approaches 1 to offset the signal amplitude attenuation or gain caused by resistivity time-varying drift, ensuring the fidelity of the feedback signal transmission. The resistivity time-varying drift characteristic index is continuously monitored. When the resistivity time-varying drift characteristic index is less than the preset drift tolerance reference value, online probe space drift compensation is adopted. Otherwise, resistivity time-varying adaptive compensation continues to be performed. When the voltage amplitude of the feedback signal is greater than the preset voltage reference value, if the resistivity time-varying drift characteristic index is still not less than the preset drift tolerance reference value, a resistivity compensation failure prompt is sent, and the current probe path resistance value, resistivity compensation coefficient, and drift characteristic index are sent to the probe test data control center. The preset voltage reference value is represented by the average voltage amplitude of the feedback signal over a historical time period.

[0038] Furthermore, the specific process of the thermomagnetic adaptive notch filtering collaborative filtering strategy is as follows: The feedback voltage signal, the driving carrier signal monitored by the force sensor, and the piezoelectric ceramic scanning driving signal in the feedback signal are used as inputs to three channels, which are then input to the independent component analysis algorithm. Three independent components are output, labeled as the mechanical component, displacement component, and thermomagnetic induced noise component, respectively. The mechanical component and displacement component constitute the pure feedback signal. Based on the separated thermomagnetic induced noise component, the characteristic frequency points of the thermomagnetic interference are identified. Specifically, the Welch periodogram method is used to estimate the power spectral density of the separated thermomagnetic induced noise component, obtaining the thermomagnetic power spectral density function, and extracting the local maxima points in the thermomagnetic power spectral density function. The specific process for obtaining the local maxima points is as follows: the thermomagnetic power spectral density function is... Values ​​are taken at scattered frequency points. For each frequency point, a neighborhood window is formed by expanding a preset number of adjacent frequency points to the left and right, with the center frequency point as the center frequency point. The preset number is set in advance by a preset person. If the amplitude of the center frequency point is greater than the amplitude of all other frequency points in its neighborhood window, the center frequency point is determined to be a local maximum point. All local maximum points are sorted in descending order of amplitude, and a preset number of frequency points are selected in the order from front to back as the interference characteristic frequency point set. For example, if the temperature magnetic power spectral density function has a peak value of 0.5 at 50Hz, a peak value of 0.8 at 120Hz, and a peak value of 0.3 at 230Hz, and the preset number of frequency points is 2, then the interference characteristic frequency point set is {120Hz, 50Hz}, where the preset number of frequency points is set in advance by a preset person.

[0039] Specifically, the Independent Component Analysis (ICA) algorithm employs a fixed-point iterative algorithm based on maximizing negative entropy (FastICA), which performs the following steps: First, the input signals of the three channels are centered by subtracting their respective means, making the mean of each channel signal zero. Then, the centered signals are whitened by calculating the eigenvalues ​​and eigenvectors of the covariance matrix through principal component analysis. The signal is projected onto the space spanned by the eigenvectors and scaled using the reciprocal of the square root of the eigenvalues ​​to obtain the whitened signal matrix. Next, a separation matrix is ​​randomly initialized, for example, with dimensions of three by three, and then... The following fixed-point iterative formula updates each column of the separation matrix: Calculate the expected value of the negative entropy gradient corresponding to the current column, solve for the new column vector using Newton's method, and then orthogonalize the separation matrix, for example, through symmetric orthogonalization or successive orthogonalization. Repeat the above iterative process until the separation matrix converges, that is, the cosine of the angle between the column vectors of two adjacent iterations is less than the preset convergence threshold. After convergence, multiply the whitened signal by the separation matrix to obtain three independent components, which are labeled as the mechanical component, displacement component, and thermomagnetic induced noise component, respectively. The preset convergence threshold is represented by the average value of the cosine of the angle between the column vectors over a historical time period.

[0040] A dynamically generated and real-time updated notch filter bank is constructed. The notch filter bank consists of multiple parallel second-order infinite impulse response (IRR) notch filters. Each filter corresponds to a specific interference characteristic frequency point, used to suppress noise signals within a preset interference-sensitive frequency band while preserving the original characteristics of the effective signal, avoiding signal distortion caused by over-filtering. Based on each frequency point in the interference characteristic frequency point set, the corresponding notch filter coefficients are obtained using the second-order IRR notch filter formula, thereby dynamically generating a filter bank composed of parallel notch filters at preset frequency points. The center frequency of each notch filter is precisely aligned with the interference characteristic frequency point set. For each frequency point, the bandwidth coefficient is preset to a fixed value by a pre-set operator to suppress narrowband noise near that frequency point. The temperature and magnetic interference characteristic index is continuously monitored. When the index is less than the preset temperature and magnetic interference reference value, it is determined that the interference has been effectively suppressed, and the current notch filter bank parameters are maintained and operation continues. Otherwise, the temperature and magnetic adaptive notch filter collaborative filtering strategy is re-executed, and the notch filter bank parameters are updated. When the number of executions exceeds the preset maximum number of iterations, if the temperature and magnetic interference characteristic index is still not less than the preset temperature and magnetic interference reference value, online compensation for probe spatial drift is performed. The preset maximum number of iterations is set in advance by a pre-set operator.

[0041] In this embodiment, the temperature-magnetic coupling interference synergistic compensation measures help reduce signal amplitude distortion caused by resistivity time-varying drift and narrowband electromagnetic noise induced by strong magnetic fields, reduce residual interference energy in the feedback signal, reduce contact feature misjudgment and spatial drift accumulation caused by signal contamination, improve the signal-to-noise ratio and fidelity of the force sensor and piezoelectric ceramic displacement feedback signal, and achieve synchronous purification of the feedback signal in both amplitude and spectrum dimensions.

[0042] Furthermore, the specific process of online compensation for probe spatial drift is as follows: Obtain a pure feedback signal, including mechanical and displacement components; perform second-order difference operations on the pure feedback signal based on the second-order central difference algorithm to obtain the second-order derivative sequence of contact resistance; the contact resistance waveform refers to the curve showing the change in the ratio of probe loop voltage to excitation current over time, acquired by the four-wire method, during the probe's uniform approach from a position away from the sample to the surface of the sample to be tested, with a sampling frequency of not less than 1kHz; extract the time corresponding to the first sampling point in the second-order derivative sequence of contact resistance where the absolute value of each data point is greater than the preset abrupt change reference value as the time of occurrence of the contact resistance abrupt change point, and determine it as the probe contact feature anchor point. The preset abrupt change reference value is represented by the average of the absolute values ​​of each data point in the second-order derivative sequence of contact resistance over a historical time period; using the probe contact feature anchor point as a reference, according to the preset approach speed and contact time deviation of the probe, according to the formula... The probe spatial displacement deviation between the probe tip and the surface of the sample under test is calculated in reverse. The sign of Δd indicates whether the actual contact position is ahead or behind the theoretical position, Δt represents the contact time deviation, and v represents the preset approximation velocity. The contact time deviation represents the difference between the current probe contacting the feature anchor point and the preset contact time. Based on the probe spatial displacement deviation, the zero-point offset in multi-degree-of-freedom probe driving commands, such as piezoelectric ceramic driving commands, is dynamically corrected. The corrected zero point is used as the spatial reference for subsequent probe fine-tuning, thereby reducing time-varying nonlinear drift caused by residual noise masking or thermal drift accumulation. The multi-degree-of-freedom probe driving commands are driving signals used to control probe movement, including but not limited to piezoelectric ceramic driving voltage and stepper motor pulse count. This embodiment uses piezoelectric ceramic as an example. The zero-point offset refers to the offset value in the multi-degree-of-freedom probe driving command corresponding to the probe zero position, i.e., the initial reference position. After correction, the actual contact position is aligned with the position of the multi-degree-of-freedom probe driving command. The corrected zero point is used as the spatial reference for subsequent probe fine-tuning, thereby reducing time-varying nonlinear drift caused by residual noise masking or thermal drift accumulation.

[0043] Furthermore, the specific process of probe-driven nonlinear self-calibration is as follows: The piezoelectric ceramic driving voltage is input into a recursive least squares algorithm to obtain the displacement sensor feedback value; a linear regression model is used to linearly fit the piezoelectric ceramic driving voltage and the displacement sensor feedback value to obtain a fitted voltage-displacement mapping curve; the open-loop driving voltage of the voltage-displacement mapping curve is obtained based on the curve, specifically as follows: based on the current preset probe desired displacement, the open-loop driving voltage is obtained by inverse solving the fitted voltage-displacement mapping curve, used to compensate for the voltage-displacement relationship distortion caused by the magnetostrictive effect; the open-loop driving voltage is superimposed as a feedforward compensation term onto the output of the PID controller, specifically: the control quantity output by the PID controller is added to the feedforward compensation term to obtain the final driving voltage, making... The voltage-displacement distortion caused by magnetostriction in a strong magnetic field environment can be corrected in real time. The displacement feedback value monitored by the current displacement sensor, the force feedback value monitored by the current force sensor, the preset desired position, and the preset desired force are input into the model predictive controller. The controller outputs an optimized driving voltage sequence for a preset future time period and executes only the first voltage value in the optimized driving voltage sequence. After executing this voltage value, the piezoelectric ceramic actuator drives the probe to produce a corresponding displacement change according to the voltage value. This displacement change is fed back to the model predictive controller in real time through the displacement sensor. At the same time, the force sensor updates the contact force value synchronously, forming a closed-loop control. The model predictive controller rereads the latest displacement feedback value and force feedback value in the next sampling cycle to achieve coordinated and precise control of probe position and contact force. The model predictive control (MPC) can simultaneously meet the requirements of probe positioning accuracy, contact force safety range (e.g., 0.1mN~1mN), and voltage change rate constraint, thereby achieving precise probe positioning and closed-loop control of contact force. This fundamentally solves the problems of positioning deviation, pressure runaway, and low test accuracy caused by residual noise masking probe contact characteristics.

[0044] In this embodiment, online compensation for probe spatial drift and nonlinear self-calibration of probe drive help reduce probe positioning deviation caused by residual noise masking probe contact characteristics and thermal drift accumulation, reduce the probability of contact pressure runaway and contact resistance fluctuation, improve the positioning accuracy and contact stability of probe in extremely low temperature and strong magnetic field coupling environment, and realize highly reliable and highly consistent automated testing of multi-degree-of-freedom probes.

[0045] like Figure 4 The diagram shown is of a portable probe testing system device provided by a multi-degree-of-freedom probe testing system with coordinated temperature and magnetic field control. Figure 4It is known that the portable probe testing system is smaller than 4 inches in size and can be placed in various vacuum equipment such as glove boxes to achieve vacuum transfer of samples. The portable probe testing system provides a vacuum-sealed environment, which can be maintained for more than 12 hours during testing. The portable probe testing system includes a probe station 1, a vacuum valve 2, a sealing cover (not shown in the figure), a testing system (not shown in the figure), and an external testing interface 3. The probe station 1 carries the multi-degree-of-freedom probe station assembly and the sample to be tested. Its drive mechanism is linked to the probe station 1, enabling the probe to complete multi-axis motion and attitude adjustment in the three-dimensional space above the probe station. The external testing interface 3 is sealed and installed on the side wall of the testing system. Its inner end connects to the probe signal end of the multi-degree-of-freedom probe station assembly, and its outer end connects to external testing instruments such as source meters and network analyzers, achieving low-noise transmission of signals between the internal cavity and the external testing system. The sealing cover and vacuum valve 2 are used to provide a vacuum environment inside the portable probe testing system. The probe station 1, as the core platform for sample carrying and probe positioning, is used to fix the sample to be tested and provides a basic carrier for temperature and magnetic field coordinated control. The multi-degree-of-freedom probe station assembly consists of a piezoelectric ceramic actuator, a stepper motor, and a multi-degree-of-freedom probe station assembly. The system comprises a motor and other drive mechanisms, along with the probe body. The probe stage has at least four probe modules, enabling multi-dimensional attitude adjustments such as X / Y / Z three-axis translation, rotation, and rocking. The motion motor can be an electromagnetic motor, stepper motor, piezoelectric inertial stage, piezoelectric scanning stage, or inchworm-like motion stage, ensuring precise probe positioning and contact with the sample. The testing system integrates a temperature control module and a magnetic field generation module. The temperature control module is equipped with heating and cooling devices that transfer heat to the sample through excellent thermal conductivity, achieving controllable temperature adjustment. Heating devices include, but are not limited to, heating rods, heating plates, and heating wires. The device includes, but is not limited to, Peltier and liquid nitrogen; a magnetic field generating module, such as an electromagnet or superconducting magnet, is used to provide a controllable magnetic field environment and achieve coordinated control with temperature; an external test interface 3 is used to enable signal communication between the probe inside the cavity and external test instruments, ensuring low-noise signal transmission; the housing of the portable probe test system can be equipped with a viewing window for easy observation of sample morphology and optical testing, and can be equipped with an antenna for microwave signal detection. A light-transmitting structure can be selected, and an optical module can be installed on the top or side of the sealed cover to perform variable-temperature optical experiments such as infrared spectroscopy and Raman spectroscopy.

[0046] Example 2, based on Example 1, serves as a second alternative to probe-driven nonlinear self-calibration. When the magnetic field strength changes rapidly over time, for example, when the magnetic field scanning rate exceeds a preset rate threshold during a sweep test, the voltage-displacement mapping of the piezoelectric ceramic exhibits time-varying nonlinear characteristics due to the coupling effect of magnetostriction and dynamic eddy current damping. The recursive least squares linear model struggles to track its dynamic changes in real time, potentially leading to feedforward compensation lag. The specific process of probe-driven nonlinear self-calibration is as follows: The preset desired displacement and magnetic field strength values ​​are input into a preset single-hidden-layer feedforward neural network, which outputs the corresponding open-loop driving voltage. The preset desired displacement is pre-set by a pre-defined operator, and the preset rate threshold is represented by the average value of the magnetic field scanning rate over a historical time period. The open-loop driving voltage is superimposed on the output of the PID controller to compensate for dynamic distortion under rapid sweeping. Finally, the current displacement feedback value, current force feedback value, preset desired position, and preset desired force are input into the model predictive control. In the device, a preset future time period optimized driving voltage sequence is output, and only the first voltage value in the optimized driving voltage sequence is executed. After executing this voltage value, the piezoelectric ceramic actuator drives the probe to generate a corresponding displacement change according to the voltage value. This displacement change is fed back to the model predictive controller in real time through the displacement sensor. At the same time, the force sensor updates the contact force value synchronously, forming a closed-loop control. The specific training process of the preset single hidden layer feedforward neural network is as follows: First, the driving voltage and corresponding displacement feedback data of the piezoelectric ceramic under different magnetic field strengths are collected to form a training sample set. The input of each sample is the preset expected displacement and magnetic field strength, and the output is the actual required open-loop driving voltage. Next, the connection weights and biases from the input layer to the hidden layer are randomly initialized. The number of hidden layer nodes is preset to a fixed value. The extreme learning machine algorithm is used, that is, the input weights and hidden layer biases are randomly fixed, and the output layer weights are calculated analytically by the least squares method to minimize the mean square error between the network output and the real driving voltage, thereby completing the training.

[0047] In this embodiment, probe-driven nonlinear self-calibration helps to correct the nonlinear distortion of voltage and displacement of piezoelectric ceramics in a strong magnetic field environment online, reduce positioning errors caused by magnetostriction and hysteresis effects, reduce the impact of feedforward compensation lag on rapid field sweep testing, improve the displacement tracking accuracy and response speed of the probe in the process of dynamic magnetic field changes, and realize high linearity and high consistency control of the piezoelectric drive system under extreme physical field conditions, thereby ensuring the positioning capability of multi-degree-of-freedom probes in complex temperature-magnetic coupling environments.

[0048] like Figure 5 The diagram shown is an overall structural schematic of a portable probe testing system device provided by a multi-degree-of-freedom probe testing system with coordinated temperature and magnetic field control. Figure 5It is known that: the sample 4 to be tested is placed inside a sealed vacuum testing chamber, and a sealing cover 5 is provided on the top of the chamber. The entire portable probe testing system is located inside a vacuum device such as a glove box 6, and is used for sample clamping and probe pre-alignment operations in an inert gas environment. The sealing cover can be selected to be entirely transparent or partially transparent at the top. A vacuum valve is installed on the side wall of the portable probe testing system, which is connected to an external vacuum device to evacuate the inside of the chamber and maintain a high vacuum environment. The chamber is equipped with a probe stage and a multi-degree-of-freedom probe assembly. The probe can move in the three-dimensional space X, Y, and Z directions to achieve precise positioning and contact with the sample. The entire portable probe testing system operates under vacuum sealing conditions, which can isolate atmospheric interference and provide a stable testing space for the sample.

[0049] like Figure 6 The diagram shown is a sample loading operation flowchart for a multi-degree-of-freedom probe testing system with coordinated temperature and magnetic field control. Figure 6 It can be seen that: first, the sample to be tested is prepared in the glove box, then the test box is sent into the glove box and opened, the sample to be tested is placed in the sample stage of the test box, the probe is moved to the appropriate position, the test box is closed and the vacuum valve is closed, then the test box is taken out from the glove box, and finally the inside of the test box is evacuated to a vacuum state and the vacuum valve is closed, thus completing the sample loading and probe pre-positioning operation before the test.

[0050] It should be noted that the multi-degree-of-freedom probe testing system with coordinated temperature and magnetic field control provided in this embodiment is also suitable for high-temperature testing scenarios, enabling controllable heating and temperature monitoring of samples over a wide temperature range, from extremely low to high temperatures. Figure 7 The figure shown is a temperature rise test line graph of a multi-degree-of-freedom probe testing system with coordinated temperature and magnetic field control. Figure 7 It can be seen that under heating and temperature control conditions, the sample temperature gradually increases over time, and the highest temperature of the test box can reach 550K, with the temperature controllable throughout the process. The horizontal axis of the figure represents time in minutes, and the vertical axis represents sample temperature in Kelvin. The curve shows the process of the sample temperature gradually rising from the initial temperature to the high temperature region, indicating that the system can be compatible with multi-field high-precision temperature controllers, such as the Kelvinion series, to achieve stable heating and precise temperature control over a wide temperature range, based on the synergistic regulation of extremely low temperature and strong magnetic field.

[0051] The above description is only an optional embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A multi-degree-of-freedom probe testing system with coordinated temperature and magnetic field control, characterized in that, include: Temperature adaptation and control module, temperature and magnetic interference co-compensation module, and probe control module: The temperature adaptation and control module is used to: perform a probe testing temperature adaptability test to evaluate the stability of the portable probe testing system in an extremely low temperature environment under volume-constrained conditions, and execute sample directional cooling adjustment and isothermal maintenance mode based on the test results. The temperature and magnetic interference co-compensation module is used to: perform quantitative analysis of probe temperature and magnetic coupling interference after the probe test temperature adaptability test is completed, and implement temperature and magnetic coupling interference co-compensation measures based on the analysis results. The probe control module is used to: perform online compensation for probe spatial drift after the probe temperature-magnetic coupling interference quantification analysis is completed, and perform probe-driven nonlinear self-calibration after the probe spatial drift online compensation is completed.

2. The multi-degree-of-freedom probe testing system with coordinated temperature and magnetic field control as described in claim 1, characterized in that, The specific process for verifying the temperature compatibility of the probe test is as follows: Obtain the temperature test data of the test area of ​​the sample to be tested; The temperature test data of the sample to be tested includes temperature distribution data and cooling device operating status data; The temperature distribution data includes ambient temperature, regional temperature gradient, and temperature fluctuation amplitude. The operating status data of the cooling device includes the output power of the cooling device and the heat exchange response rate. The ambient temperature represents the average temperature value of each monitoring point inside the test chamber of the portable probe testing system. The regional temperature gradient is represented by calculating the difference between the instantaneous temperature value of each point and the instantaneous temperature value of the adjacent points, calculating the ratio between the result of the difference calculation and the distance between the corresponding adjacent points, and averaging all the ratio calculation results. The temperature fluctuation amplitude is represented by the result of the difference calculation between the highest and lowest temperature values ​​in the test chamber of the portable probe test system during the temperature sampling period. The heat exchange response rate is expressed by calculating the difference between the current temperature value and the initial temperature inside the test chamber of the portable probe test system, and then dividing the result of the difference calculation by the adjustment time. The temperature test data of the sample to be tested is preprocessed by noise reduction and normalization to obtain the preprocessed standardized sample temperature field adaptation data. The standardized sample temperature field adaptation data is input into the preset temperature control correlation model, and the temperature control adaptation prediction results are output. The temperature control adaptation prediction results include the temperature deviation coefficient, temperature fluctuation trend prediction value, and ultra-low temperature stability evaluation index of the test area of ​​the sample to be tested. The temperature control adaptation status is determined based on the temperature control adaptation prediction results. The specific process is as follows: Determine whether the temperature control adaptation prediction result meets the low temperature stability discrimination condition. If so, trigger the constant temperature maintenance mode and perform probe temperature magnetic coupling interference quantitative analysis. Otherwise, trigger sample directional cooling adjustment. The low-temperature stability discrimination criteria indicate that the temperature deviation coefficient is less than the preset temperature deviation reference value, the temperature fluctuation trend prediction value is less than the preset fluctuation reference value, and the ultra-low temperature stability evaluation index is greater than the preset stability reference value.

3. The multi-degree-of-freedom probe testing system with coordinated temperature and magnetic field control as described in claim 1, characterized in that, The specific process for directional cooling of the sample is as follows: Input the temperature control adaptation prediction results into the preset temperature adjustment mapping table for querying, and obtain the cooling power adjustment step size and heating compensation power adjustment step size; The cooling power adjustment step size represents the adjustment range of the cooling device's output power within a single control cycle. The heating compensation power adjustment step size represents the adjustment range of the output compensation power of the heating device within a single control cycle. The output power of the cooling device is adjusted step by step in the direction of decreasing the temperature deviation coefficient by adjusting the cooling power step by step, and the compensation power of the heating device is adjusted step by step in the direction of decreasing the predicted value of temperature fluctuation trend by adjusting the heating compensation power step by step. The temperature control adaptation prediction results are continuously monitored. If the temperature control adaptation prediction results meet the low temperature stability discrimination conditions, the current cooling power and heating compensation power are maintained to adjust the temperature. When the temperature of the sample test area is in the preset ultra-low temperature range within a preset number of control cycles, and the temperature control adaptation prediction results meet the low temperature stability discrimination conditions, the constant temperature maintenance mode is triggered, and probe temperature magnetic coupling interference quantitative analysis is performed. Conversely, if the sample directional cooling adjustment is not performed, the process will continue.

4. The multi-degree-of-freedom probe testing system with coordinated temperature and magnetic field control as described in claim 3, characterized in that, The specific process of the constant temperature maintenance mode is as follows: The temperature deviation value and the rate of temperature change are input into the PID controller to obtain the temperature coordinated regulation control quantity; The temperature deviation value is represented by the result of a difference calculation between the current instantaneous temperature value and the preset ultra-low temperature target value; The rate of temperature change during the temperature sampling period was obtained using the central difference method. The temperature coordinated regulation control quantity is sent to the probe test data control center, and cooling power adjustment command and heating power adjustment command are output. The cooling device is constrained to maintain a constant output based on the cooling power adjustment command. The heating device is adjusted for minor balance and temperature compensation based on the heating power adjustment command. The constant temperature maintenance mode continues to run until the test task of the sample to be tested is completed or a new temperature setting command is received.

5. The multi-degree-of-freedom probe testing system with coordinated temperature and magnetic field control as described in claim 3, characterized in that, The quantitative analysis of probe temperature-magnetic coupling interference is performed as follows: The DC resistance values ​​of the probe leads, contact electrodes and both ends of the signal transmission path are collected. The instantaneous temperature value and the current magnetic field strength value of the test area of ​​the sample under test are collected simultaneously to construct a probe temperature-magnetic multidimensional sequence containing timestamps. The DC resistance value is linearly fitted using a sliding window least squares linear fitting algorithm to obtain the resistance change rate. The Pearson correlation coefficient between the rate of change of resistivity and the rate of change of temperature is obtained based on the Pearson correlation coefficient algorithm and used as an indicator of the time-varying drift characteristic of resistivity. The power spectral density of the original signals from feedback loops such as force sensors and piezoelectric ceramics is estimated to obtain the power spectral density function. Within the preset interference sensitive frequency band, the interference energy concentration coefficient is calculated and used as a characteristic index of thermomagnetic interference. The resistivity time-varying drift characteristic index is compared with a preset drift tolerance reference value, and the temperature and magnetic interference characteristic index is compared with a preset temperature and magnetic interference reference value. If both the resistivity time-varying drift characteristic index and the thermomagnetic interference characteristic index are less than the corresponding interference comparison reference value, then online compensation for probe spatial drift is performed. Conversely, measures should be taken to compensate for temperature and magnetic coupling interference.

6. The multi-degree-of-freedom probe testing system with coordinated temperature and magnetic field control as described in claim 5, characterized in that, The specific process of the temperature-magnetic coupling interference collaborative compensation measure is as follows: If the resistivity time-varying drift characteristic index is less than the preset drift tolerance reference value, and the temperature and magnetic interference characteristic index is not less than the preset temperature and magnetic interference reference value, the temperature and magnetic adaptive notch filter collaborative filtering strategy is adopted. If the resistivity time-varying drift characteristic index is not less than the preset resistance drift tolerance reference value, and the temperature and magnetic interference characteristic index is less than the preset temperature and magnetic interference reference value, resistivity time-varying adaptive compensation is adopted. If the resistivity time-varying drift characteristic index is not less than the preset drift tolerance reference value, and the temperature and magnetic interference characteristic index is not less than the preset temperature and magnetic interference reference value, then the resistivity time-varying adaptive compensation and temperature and magnetic adaptive notch filtering collaborative filtering strategy are performed in sequence. To address the resistivity time-varying drift interference dominated by extremely low temperatures, the resistivity time-varying adaptive compensation is performed to dynamically generate resistivity compensation coefficients. Using the amplitude corresponding to the resistivity compensation coefficient as the adjustment step size, the voltage amplitude of the feedback signal is adjusted step by step in the direction that the resistivity compensation coefficient approaches 1. The resistivity time-varying drift characteristic index is continuously monitored. When the resistivity time-varying drift characteristic index is less than the preset drift tolerance reference value, online probe space drift compensation is adopted. Otherwise, resistivity time-varying adaptive compensation is continued.

7. The multi-degree-of-freedom probe testing system with coordinated temperature and magnetic field control as described in claim 6, characterized in that, The specific process of the temperature-magnetic adaptive notch filter collaborative filtering strategy is as follows: The feedback voltage signal, drive carrier signal, and piezoelectric ceramic scanning drive signal in the feedback signal are used as inputs to three channels, which are then input to the independent component analysis algorithm. The algorithm outputs three independent components, which are labeled as mechanical component, displacement component, and thermomagnetic induced noise component, respectively. Based on the separated thermomagnetic induced noise components, the characteristic frequency points of thermomagnetic interference are identified. The specific identification process is as follows: The power spectral density of the separated thermomagnetic induced noise component is estimated to obtain the thermomagnetic power spectral density function, and the local maxima in the thermomagnetic power spectral density function are extracted. Dynamically generate notch filter banks that can be updated in real time; The temperature and magnetic interference characteristic index is continuously monitored. When the index is less than the preset temperature and magnetic interference reference value, it is determined that the interference has been effectively suppressed. The current notch filter bank parameters are maintained and the operation continues. Otherwise, the temperature and magnetic adaptive notch collaborative filtering strategy is re-executed and the notch filter bank parameters are updated.

8. The multi-degree-of-freedom probe testing system with coordinated temperature and magnetic field control as described in claim 6, characterized in that, The specific process of online compensation for probe spatial drift is as follows: Obtain a clean feedback signal, including mechanical and displacement components; Perform a second-order difference operation on the contact resistance waveform in the pure feedback signal to obtain the second-order derivative sequence of the contact resistance; The moment corresponding to the first sampling point in the second derivative sequence of contact resistance that has an absolute value greater than the preset abrupt change reference value is extracted as the moment when the contact resistance abrupt change point occurs, and is determined as the probe contact feature anchor point. Based on the probe contact feature anchor point, the probe spatial displacement deviation between the probe tip and the surface of the sample to be tested is calculated in reverse according to the probe's preset approach speed and contact time deviation. Based on the probe spatial displacement deviation, the multi-degree-of-freedom probe driving command is dynamically corrected and the corrected zero point is used as the spatial reference for subsequent probe fine-tuning. The corrected zero point is used as the spatial reference for subsequent probe fine-tuning, thereby reducing time-varying nonlinear drift caused by residual noise masking or thermal drift accumulation.

9. The multi-degree-of-freedom probe testing system with coordinated temperature and magnetic field control as described in claim 1, characterized in that, The specific process of probe-driven nonlinear self-calibration is as follows: The piezoelectric ceramic driving voltage is input into a recursive least squares algorithm to obtain the displacement sensor feedback value; The piezoelectric ceramic driving voltage and displacement sensor feedback value are linearly fitted based on a linear regression model to obtain the fitted voltage-displacement mapping curve; The open-loop driving voltage of the voltage-displacement mapping curve is obtained based on the voltage-displacement mapping curve. The specific acquisition process is as follows: Based on the current preset probe expected displacement, the open-loop driving voltage is obtained by inverse solution using the currently fitted voltage-displacement mapping curve; The open-loop drive voltage is added as a feedforward compensation term to the output of the PID controller, specifically: The control quantity output by the PID controller is added to the feedforward compensation term to obtain the final drive voltage; The displacement feedback value, force feedback value, preset desired position, and preset desired force are input into the model predictive controller, which outputs an optimized driving voltage sequence for a preset future time period and executes only the first voltage value in the optimized driving voltage sequence. The model predictive controller rereads the latest displacement and force feedback values ​​in the next sampling period.

10. The multi-degree-of-freedom probe testing system with coordinated temperature and magnetic field control as described in claim 9, characterized in that, The probe-driven nonlinear self-calibration also includes: The preset desired displacement and magnetic field strength values ​​are input into a preset single hidden layer feedforward neural network, and the corresponding open-loop driving voltage is output. The open-loop drive voltage is superimposed on the output of the PID controller; The current displacement feedback value, the current force feedback value, the preset desired position, and the preset desired force are input into the model prediction controller, which outputs a preset future time period optimized driving voltage sequence and executes only the first voltage value in the optimized driving voltage sequence.

Images