Multi-spindle synchronous servo control method and system for threading machine

Abstract

The invention provides a threading machine multi-spindle synchronous servo control method and system. According to the method, dynamic torque fluctuation data of the worm gear transmission shaft of the threading machine are collected in real time, a torque frequency spectrum matrix is generated through frequency spectrum transformation, and fluctuation frequency spectrum features are extracted; a non-contact worm gear torque sensor is integrated on a transmission shaft, and a dynamic sampling strategy is triggered based on fluctuation characteristics; recording magnetic induction intensity change and torsion deformation, and generating a coupling state identifier of the torque and the rotating speed according to a coupling relationship; a fluctuation spectrum phase offset feature is obtained, a lagging risk area exceeding the synchronization tolerance is positioned based on a coupling identifier, and a phase offset is extracted as a spindle dynamic compensation vector; and analyzing the compensation vector to generate a synchronization error suppression instruction, and adjusting the output torque of the servo shaft in real time to suppress the multi-spindle synchronization error. According to the application, microsecond worm gear transmission lag dynamic compensation and multi-spindle synchronous error compression are realized, and the service life of equipment is prolonged.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of multi-spindle synchronous servo control, and particularly relates to a tapping machine multi-spindle synchronous servo control method and system. BACKGROUND

[0002] As the core equipment for thread processing, the tapping machine needs to meet the requirements of high precision and high efficiency for thread cutting in batch manufacturing of oil pipes and automobile parts. The multi-spindle cooperative operation faces severe challenges: first, dynamic load disturbance easily causes spindle speed fluctuation, resulting in thread lead error; second, phase synchronization accuracy needs to be controlled at the millisecond level to avoid thread misalignment or tooth disorder caused by angle deviation of multi-spindles in high-speed rotation; third, thermal deformation and mechanical resonance will further amplify multi-axis trajectory deviation. The traditional mechanical gear synchronization method has poor anti-disturbance performance due to rigid coupling, and cannot compensate for nonlinear disturbance in real time, so there is an urgent need for a servo control method that can adapt to load changes and suppress phase cumulative error online to ensure that multi-spindles remain strictly synchronized in position, speed and torque under variable working conditions and eliminate the risk of thread rejection in batch processing.

[0003] The current targeted solution is a master-slave cooperative control architecture based on a virtual spindle: the system takes the virtual spindle as the global synchronization reference, and each physical spindle receives the phase instruction of the virtual spindle through the high-speed bus, and dynamically corrects the relative position error between the slave axes by using a cross-coupling compensator; at the same time, a load observer is integrated to estimate the cutting torque disturbance in real time, which is fed forward to the current loop to offset the influence of torque fluctuation on synchronization accuracy. This scheme builds a full closed-loop control through high-resolution encoder feedback, significantly improving the synchronization performance of multi-spindles in steady-state conditions. However, the core deficiency is that the decoupling of the virtual instruction source and the dynamic response of the physical axis is not complete, and when a sudden load disturbance causes a single-axis transient step-out, the phase error will spread to the whole domain through the coupling network, and the system lacks the ability to adaptively suppress the frequency drift of mechanical resonance, resulting in a sharp decline in synchronization stability under high-speed conditions. SUMMARY

[0004] The present application provides a tapping machine multi-spindle synchronous servo control method and system to solve the problems of existing technology such as worm gear transmission lag compensation inaccuracy, multi-spindle synchronization error propagation and loss of control of equipment wear.

[0005] In a first aspect, the present application provides a tapping machine multi-spindle synchronous servo control method, comprising: real-time acquisition of dynamic torque fluctuation data of a worm gear transmission shaft of the tapping machine, generation of a torque frequency spectrum matrix by performing frequency spectrum transformation on the dynamic torque fluctuation data, and extraction of fluctuation frequency spectrum features from the torque frequency spectrum matrix; integration of a non-contact worm gear torque sensor on the worm gear transmission shaft, and triggering of the non-contact worm gear torque sensor based on the fluctuation frequency spectrum features to perform a dynamic sampling strategy; Record the changes in magnetic field strength and torsional deformation of the worm gear drive shaft under the dynamic sampling strategy, and generate a coupling state identifier of torque and speed of the worm gear drive shaft based on the coupling relationship between the changes in magnetic field strength and torsional deformation. The phase offset feature of the fluctuation spectrum is obtained, and the hysteresis risk zone where the phase offset feature exceeds the preset synchronization tolerance is located based on the coupling state identifier. The phase offset corresponding to the hysteresis risk zone is extracted as the main axis dynamic compensation vector. The main spindle dynamic compensation vector is analyzed to generate a synchronization error suppression command. The output torque command of each servo axis of the thread cutting machine is adjusted in real time according to the synchronization error suppression command to suppress the multi-spindle synchronization error caused by the worm gear drive lag.

[0006] Optionally, dynamic torque fluctuation data of the worm gear drive shaft of the thread-cutting machine is acquired in real time, the dynamic torque fluctuation data is subjected to spectral transformation to generate a torque spectrum matrix, and fluctuation spectrum features are extracted from the torque spectrum matrix, including: The torque change value during the rotation of the worm gear drive shaft of the thread cutting machine is continuously collected, and the torque change value is arranged in time sequence to form a dynamic torque fluctuation data sequence. The dynamic torque fluctuation data sequence is divided into multiple data segments according to a fixed time window, and a time-domain to frequency-domain conversion operation is performed on each data segment to generate a spectral distribution segment. The spectral distribution segments are spliced ​​together in chronological order to form a torque spectrum matrix, and the spectral intensity values ​​of the torque spectrum matrix are extracted. The frequency components whose spectral intensity values ​​exceed a preset background noise threshold are identified, and the spectral intensity values ​​corresponding to the frequency components are recorded as fluctuation spectral features.

[0007] Optionally, a non-contact worm gear torque sensor is integrated on the worm gear drive shaft, and the non-contact worm gear torque sensor is triggered based on the fluctuation spectrum characteristics to execute a dynamic sampling strategy, including: A non-contact worm gear torque sensor is integrated on the worm gear drive shaft, and the spectral intensity value recorded in the fluctuation spectrum characteristics is read. When the spectral intensity value exceeds a preset trigger threshold, an activation command for the non-contact worm gear torque sensor is generated. According to the activation command, the non-contact worm gear torque sensor is activated to perform a dynamic scanning operation on the magnetic field of the worm gear drive shaft surface to acquire an initial magnetic field reference reading. The sampling frequency parameter is set according to the frequency component whose spectral intensity value exceeds the trigger threshold, and the magnetic field sensing probe of the non-contact worm gear torque sensor is controlled to perform magnetic field sampling operation according to the sampling frequency parameter to record the real-time magnetic field intensity change sequence. The initial magnetic field reference readings are integrated with the real-time magnetic field intensity change sequence to form a dynamic sampling strategy.

[0008] Optionally, a sampling frequency parameter is set based on the frequency components whose spectral intensity values ​​exceed a trigger threshold, and the magnetic field sensing probe of the non-contact worm gear torque sensor is controlled to perform a magnetic field sampling operation according to the sampling frequency parameter to record the real-time magnetic field intensity change sequence, including: Extract the frequency components exceeding the trigger threshold from the fluctuation spectrum features, and combine the frequency components with a preset sampling multiple coefficient to generate sampling frequency parameters; The sampling frequency parameter is written into the magnetic field induction probe control register of the non-contact worm gear torque sensor to set the probe sampling interval time. The magnetic field sensing probe of the non-contact worm gear torque sensor is activated to perform magnetic field strength acquisition operation at the sampling interval, and the real-time raw magnetic field strength reading output by the magnetic field sensing probe is recorded at the same time. The difference between the original real-time magnetic field strength reading and the initial magnetic field reference reading is calculated as the change in real-time magnetic field strength. All real-time magnetic field intensity changes are arranged in chronological order of sampling time to form a sequence of real-time magnetic field intensity changes.

[0009] Optionally, the changes in magnetic flux density and torsional deformation of the worm gear drive shaft under the dynamic sampling strategy are recorded, and a coupling state identifier of torque and rotational speed of the worm gear drive shaft is generated based on the coupling relationship between the changes in magnetic flux density and torsional deformation, including: Mark the target point on the surface of the worm gear drive shaft, and capture the displacement change of the target point as a torsional deformation sequence using a high-speed vision sensor. The magnetic field strength change sequence and the torsional deformation sequence are aligned at the same time point to form paired data units; The ratio of the change in magnetic field strength to the torsional deformation in the paired data unit is calculated as the coupling coefficient between the change in magnetic field strength and the torsional deformation; When the coupling coefficient exceeds the preset material property calibration threshold, the corresponding time point is marked as an abnormal coupling state point. The abnormal coupling state points are integrated to generate a coupling state identifier that characterizes the abnormal relationship between torque and speed of the worm gear drive shaft.

[0010] Optionally, the phase shift characteristics of the fluctuation spectrum are obtained, and the hysteresis risk region where the phase shift characteristics exceed the preset synchronization tolerance is located based on the coupling state identifier. The phase shift amount corresponding to the hysteresis risk region is extracted as the principal axis dynamic compensation vector, including: Extract the phase angle change values ​​corresponding to the frequency components recorded in the wave spectrum features, and arrange the phase angle change values ​​in chronological order to form a phase shift feature sequence; Within the time interval corresponding to the abnormal coupling state point marked by the coupling state identifier, locate the phase angle change value at the same time point in the phase offset feature sequence; The phase angle change value is compared with a preset synchronization tolerance threshold. When the phase angle change value exceeds the synchronization tolerance threshold, the corresponding time point is marked as a lag risk point. Integrate the time intervals containing consecutive lag risk points, and mark the working stages of the worm gear drive shaft covered by the time intervals as lag risk zones; The average value of the phase angle change within the hysteresis risk zone is extracted as the principal axis dynamic compensation vector.

[0011] Optionally, the spindle dynamic compensation vector is parsed to generate a synchronization error suppression command, and the output torque command of each servo axis of the thread cutting machine is adjusted in real time according to the synchronization error suppression command to suppress the multi-spindle synchronization error caused by worm gear drive lag, including: The spindle dynamic compensation vector is combined with a preset torque correction coefficient to generate a torque correction value; Obtain the current output torque command of each servo axis of the thread cutting machine, and superimpose the torque correction value onto the output torque command of the corresponding servo axis to form an updated output torque command; A synchronization error suppression command is generated based on the updated output torque command, and the synchronization error suppression command is transmitted to the servo axis controller of the thread cutting machine. The servo axis controller executes a synchronization error suppression command, which adjusts the output torque of each servo axis according to the synchronization error suppression command to suppress the multi-spindle synchronization error caused by the worm gear drive lag.

[0012] Secondly, this application provides a multi-spindle synchronous servo control system for a thread-cutting machine, comprising: The acquisition module is used to acquire dynamic torque fluctuation data of the worm gear drive shaft of the thread cutting machine in real time, perform spectral transformation on the dynamic torque fluctuation data to generate a torque spectrum matrix, and extract fluctuation spectrum features from the torque spectrum matrix. A triggering module is used to integrate a non-contact worm gear torque sensor on the worm gear drive shaft, and to trigger the non-contact worm gear torque sensor based on the fluctuation spectrum characteristics to execute a dynamic sampling strategy; The generation module is used to record the changes in magnetic field strength and torsional deformation of the worm gear drive shaft under the dynamic sampling strategy, and generate a coupling state identifier of torque and speed of the worm gear drive shaft based on the coupling relationship between the changes in magnetic field strength and torsional deformation. The positioning module is used to acquire the phase offset features of the fluctuation spectrum features, locate the hysteresis risk zone where the phase offset features exceed the preset synchronization tolerance based on the coupling state identifier, and extract the phase offset amount corresponding to the hysteresis risk zone as the main axis dynamic compensation vector. The adjustment module is used to parse the main spindle dynamic compensation vector to generate a synchronization error suppression command, and adjust the output torque command of each servo axis of the thread cutting machine in real time according to the synchronization error suppression command to suppress the multi-spindle synchronization error caused by worm gear drive lag.

[0013] Thirdly, this application provides a computing device, including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are to be invoked and executed by the processing component to implement a multi-spindle synchronous servo control method for a thread-cutting machine as described in the first aspect above.

[0014] Fourthly, this application provides a computer storage medium storing a computer program, which, when executed by a computer, implements a multi-spindle synchronous servo control method for a thread-cutting machine as described in the first aspect.

[0015] This application's technical solution achieves high-precision synchronous control of multiple spindles in a thread-cutting machine through dynamic torque monitoring and intelligent compensation technology. Specifically, torque fluctuation feature extraction based on spectrum analysis significantly improves the sensitivity of transmission state recognition; non-contact sensing and magnetic-deformation coupling analysis ensure the reliability of dynamic sampling data; and a phase offset compensation mechanism effectively suppresses multi-spindle synchronization errors. This method overcomes the control limitations of traditional mechanical transmissions, achieving fully closed-loop intelligent control from torque fluctuation detection to real-time compensation, significantly improving the machining accuracy and equipment operational stability of the thread-cutting machine, and providing a highly synchronous spindle control solution for complex thread machining.

[0016] Furthermore, high-precision synchronous control of multiple spindles in a thread-cutting machine is achieved through dynamic torque monitoring and intelligent compensation technology. Specifically, torque fluctuation feature extraction based on spectrum analysis significantly improves the sensitivity of transmission state recognition; non-contact sensing and magnetic-deformation coupling analysis ensure the reliability of dynamic sampling data; and a phase offset compensation mechanism effectively suppresses multi-spindle synchronization errors. This method overcomes the control limitations of traditional mechanical transmissions, achieving fully closed-loop intelligent control from torque fluctuation detection to real-time compensation, significantly improving the machining accuracy and operational stability of the thread-cutting machine, and providing a highly synchronous spindle control solution for complex thread machining.

[0017] These or other aspects of this application will become more apparent in the following description of the embodiments. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 A flowchart of a multi-spindle synchronous servo control method for a thread-cutting machine provided in this application is shown; Figure 2 This application provides a schematic diagram of the structure of a multi-spindle synchronous servo control system for a thread-cutting machine. Figure 3 A schematic diagram of the structure of a computing device provided in this application is shown. Detailed Implementation

[0020] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0021] In some of the processes described in the specification, claims, and accompanying drawings of this application, multiple operations appearing in a specific order are included. However, it should be clearly understood that these operations may not be executed in the order they appear herein, or may be executed in parallel. The operation numbers, such as 101, 102, etc., are merely used to distinguish different operations and do not themselves represent any execution order. Furthermore, these processes may include more or fewer operations, and these operations may be executed sequentially or in parallel. It should be noted that the descriptions such as "first," "second," etc., in this document are used to distinguish different messages, devices, modules, etc., and do not represent a chronological order, nor do they limit "first" and "second" to different types.

[0022] Researchers have discovered a fundamental bottleneck in multi-spindle synchronous control technology for thread cutting machines: while a collaborative architecture based on virtual spindles can improve steady-state synchronization accuracy, its lack of dynamic disturbance decoupling and failure to suppress resonance leads to system instability under high-speed conditions. Specifically, instantaneous loss of synchronization in a single axis caused by sudden load disturbances propagates into global phase accumulation error through the coupling network, and the drift of the mechanical resonant frequency exacerbates the trajectory deviation amplification effect, resulting in a sharp increase in the thread mis-threading rate during batch processing. This contradiction stems from the blind spot in the collaborative perception of dynamic torque fluctuations and phase shifts in the worm gear transmission system, necessitating the construction of a closed-loop disturbance rejection architecture from the mechanical transmission layer to the control layer.

[0023] To address the aforementioned challenges, this invention proposes a multi-spindle synchronous servo control method for a thread-cutting machine. Its innovation lies in overcoming the limitations of virtual spindle control by using torque fluctuation characteristics to drive dynamic sampling and magnetic-deformation coupling analysis. Specifically: Real-time acquisition of dynamic torque fluctuation data from the worm gear drive shaft is performed, and the fluctuation spectrum characteristics are extracted through spectrum transformation; a non-contact torque sensor is triggered to execute a dynamic sampling strategy, simultaneously capturing changes in magnetic field strength and torsional deformation; a torque-speed status indicator is generated based on the magnetic-deformation coupling relationship; the hysteresis risk zone exceeding the synchronization tolerance is located through phase offset characteristics, and the phase offset is extracted as a dynamic compensation vector; the compensation vector is analyzed to generate a synchronization error suppression command, adjusting the servo shaft output torque in real time. This method overturns the traditional master-slave control paradigm: torque spectrum feature-driven sampling achieves millisecond-level perception of nonlinear disturbances in the transmission system for the first time; magnetic induction-deformation coupling identification accurately quantifies the impact of resonant frequency drift on phase synchronization through deep integration of mechanical state and electrical signals; dynamic compensation mechanism forms a closed-loop control chain of "fluctuation perception-state interpretation-offset positioning-real-time suppression", providing essential guarantees for high-precision thread machining from mechanical transmission disturbances to multi-axis collaborative stability.

[0024] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0025] Figure 1 This application provides a flowchart of a multi-spindle synchronous servo control method for a thread-cutting machine, as shown in the embodiments of this application. Figure 1 As shown, the method includes: 101. Real-time acquisition of dynamic torque fluctuation data of the worm gear drive shaft of the thread cutting machine, spectral transformation of the dynamic torque fluctuation data to generate a torque spectrum matrix, and extraction of fluctuation spectrum features from the torque spectrum matrix.

[0026] Optionally, step 101 may specifically include the following steps: 1011. Continuously collect the torque change value during the rotation of the worm gear drive shaft of the thread cutting machine, and arrange the torque change value in time sequence to form a dynamic torque fluctuation data sequence.

[0027] 1012. Divide the dynamic torque fluctuation data sequence into multiple data segments according to a fixed time window, and perform a time-domain to frequency-domain conversion operation on each data segment to generate a spectral distribution segment.

[0028] 1013. The spectral distribution segments are spliced ​​together in chronological order to form a torque spectrum matrix, and the spectral intensity values ​​of the torque spectrum matrix are extracted.

[0029] 1014. Identify the frequency components whose spectral intensity values ​​exceed a preset background noise threshold, and record the spectral intensity values ​​corresponding to the frequency components as fluctuation spectral features.

[0030] In the above scheme, a thread-cutting machine refers to a machine tool used for machining threads. A worm gear drive shaft refers to the rotating shaft of a worm gear drive. Dynamic torque fluctuation data refers to data on torque changes over time. Spectrum transformation refers to the method of converting a time-domain signal into a frequency-domain signal. A torque spectrum matrix refers to a matrix reflecting the spectral characteristics of torque. Fluctuation spectrum characteristics refer to the characteristic components in the spectrum. Torque change value refers to the instantaneous change in torque. A dynamic torque fluctuation data sequence refers to torque data arranged in time. A time window refers to the time range of data segments. A spectrum distribution segment refers to short-time spectrum data. A spectrum intensity value refers to the intensity of a spectrum component. A preset background noise threshold refers to the critical value for judging a valid signal. Frequency components refer to the frequency components in the spectrum.

[0031] In this embodiment, the system first continuously collects torque changes during rotation using a dynamic torque sensor mounted on the worm gear drive shaft of the thread-cutting machine. The data acquisition module, based on strain gauge electrical measurement technology, uses Wheatstone bridges composed of resistance strain gauges bonded to the surface of the drive shaft. When the drive shaft undergoes minute deformation under torque, the resistance value of the strain gauges changes in real time. A signal conditioning circuit converts this resistance change into a voltage signal output. A timing recorder captures the voltage signal at a fixed sampling frequency and arranges it in timestamp order as a dynamic torque fluctuation data sequence, forming the original time-domain signal stream. This process relies on high-precision analog-to-digital conversion technology to ensure data real-time performance and integrity.

[0032] Subsequently, the system divides the dynamic torque fluctuation data sequence into data segments with fixed time windows and performs time-frequency conversion on each data segment: the window segmenter divides the continuous data stream into multiple data segments according to a preset duration, each segment containing an equal number of sampling points. The spectrum converter applies a Fast Fourier Transform algorithm to each data segment, decomposing the time-domain torque signal into frequency components and generating a spectral distribution segment containing amplitude and phase information. This process suppresses spectral leakage through windowing functions (such as the Hanning window) to ensure the accuracy of the frequency domain energy distribution.

[0033] Next, the system stitches together spectral distribution segments in chronological order to form a torque spectrum matrix and extracts spectral intensity values. The matrix construction module arranges the spectral distribution segments of each data segment into a two-dimensional matrix according to the acquisition time order. The rows represent the time axis, and the columns represent the frequency axis. The matrix elements are the amplitudes at the corresponding time and frequency points. The intensity extractor traverses each frequency point in the matrix and takes the square of its amplitude as the spectral intensity value (quantifying the energy magnitude of the frequency component within a specific time window). This process integrates the spatiotemporal torque fluctuation characteristics based on time-frequency matrix reconstruction technology.

[0034] Finally, the system identifies frequency components whose spectral intensity exceeds a preset background noise threshold and records their intensity as fluctuation spectral features: a noise threshold filter compares the spectral intensity value of each frequency point with the preset background noise threshold (the background noise energy calibrated through an unloaded experiment). The feature extractor filters all frequency components exceeding the limit, recording their frequency values ​​and corresponding spectral intensity values ​​to form a set of fluctuation spectral features. This process uses a peak search algorithm to automatically locate significant frequency components and eliminate environmental interference.

[0035] In practical applications, in the scenario of synchronous control of multi-spindle threading on a threading machine, strain gauge torque sensors are installed on the worm gear drive shaft of the threading machine to continuously collect the torque change values ​​during the rotation of the worm gear drive shaft (such as the periodic load fluctuations of the worm gear drive bearing when the spindle rotates at high speed). The torque change values ​​are arranged in chronological order to form a dynamic torque fluctuation data sequence (step 1011). Then, the dynamic torque fluctuation data sequence is divided into multiple data segments according to a fixed time window (such as dividing a single feed cycle of threading with a millisecond-level window). A conversion operation from the time domain to the frequency domain is performed on each data segment to generate a spectrum distribution segment (step 1012). Next, the spectrum distribution segments are spliced ​​in chronological order to form a torque spectrum matrix (such as arranging the frequency domain energy distribution of each segment according to the processing cycle time sequence), and the spectrum intensity value of the torque spectrum matrix is ​​extracted (step 1013). Finally, frequency components with spectrum intensity values ​​exceeding a preset background noise threshold are identified (such as retaining characteristic frequencies related to spindle speed after excluding environmental vibration noise), and the spectrum intensity value corresponding to the frequency component is recorded as the fluctuation spectrum feature (step 1014). The fluctuation spectrum feature is used to diagnose the synchronization status of multiple spindles: the dynamic torque fluctuation data sequence reflects the sudden load change of the worm gear drive shaft in thread cutting (such as the instantaneous torque surge when the tool cuts into the workpiece), the spectrum distribution segment reveals the frequency response pattern of a specific machining stage (such as high-frequency harmonics in the spindle acceleration stage), the torque spectrum matrix maps the abnormal frequency points of the entire machining cycle through time-frequency correlation (such as low-frequency resonance in the thread finishing stage), and the fluctuation spectrum feature quantifies the root cause of synchronization error (such as the torque fluctuation spectrum intensity value of a spindle exceeding the standard due to transmission backlash), providing a basis for error compensation for subsequent multi-spindle servo control.

[0036] The scheme described in step 101 above achieves refined spectral analysis and feature extraction of torque fluctuations in the worm gear drive shaft. Through continuous acquisition and time-frequency conversion processing of dynamic torque data, a torque spectrum matrix reflecting the dynamic characteristics of the transmission system is innovatively constructed. This technology employs an adaptive noise threshold identification algorithm to accurately extract engineering-significant fluctuation spectrum features, providing key frequency domain indicators for subsequent condition monitoring. Through time window segmentation and spectrum stitching techniques, a seamless transition from transient analysis to continuous monitoring is achieved, establishing a scientific frequency domain feature library for transmission system fault early warning.

[0037] 102. Integrate a non-contact worm gear torque sensor on the worm gear drive shaft, and trigger the non-contact worm gear torque sensor based on the fluctuation spectrum characteristics to execute a dynamic sampling strategy.

[0038] Optionally, step 102 may specifically include the following steps: 1021. A non-contact worm gear torque sensor is integrated on the worm gear transmission shaft, and the spectral intensity value recorded in the fluctuation spectrum characteristics is read. When the spectral intensity value exceeds a preset trigger threshold, an activation command for the non-contact worm gear torque sensor is generated.

[0039] 1022. According to the activation command, the non-contact worm gear torque sensor is activated to perform a dynamic scanning operation on the magnetic field of the worm gear drive shaft surface to acquire an initial magnetic field reference reading.

[0040] 1023. Set the sampling frequency parameter according to the frequency component whose spectral intensity value exceeds the trigger threshold, and control the magnetic field sensing probe of the non-contact worm gear torque sensor to perform magnetic field sampling operation according to the sampling frequency parameter to record the real-time magnetic field intensity change sequence.

[0041] Step 1023 may specifically include the following processes: extracting frequency components exceeding the trigger threshold from the fluctuation spectrum characteristics, and combining the frequency components with a preset sampling multiple coefficient to generate sampling frequency parameters; writing the sampling frequency parameters into the magnetic field sensing probe control register of the non-contact worm gear torque sensor to set the probe sampling interval time; starting the magnetic field sensing probe of the non-contact worm gear torque sensor to perform magnetic field strength acquisition operation according to the sampling interval time, and simultaneously recording the real-time magnetic field strength raw reading output by the magnetic field sensing probe; calculating the difference between the real-time magnetic field strength raw reading and the initial magnetic field reference reading as the real-time magnetic field strength change; arranging all real-time magnetic field strength changes in order of sampling time to form a real-time magnetic field strength change sequence.

[0042] 1024. Integrate the initial magnetic field reference reading with the real-time magnetic field intensity change sequence to form a dynamic sampling strategy.

[0043] In the above scheme, a non-contact worm gear torque sensor refers to a sensor that can measure torque without contact. Dynamic sampling strategy refers to a sampling scheme adjusted according to requirements. Spectrum intensity value refers to the intensity of a spectral component. Preset trigger threshold refers to the critical value for activating the sensor. Activation command refers to the control signal that starts the sensor. Dynamic scanning operation refers to the process of scanning the magnetic field. Initial magnetic field reference reading refers to the initial magnetic field intensity value. Sampling frequency parameter refers to the setting parameter of the sampling frequency. Magnetic field induction probe refers to the sensor component that detects the magnetic field. Magnetic field sampling operation refers to the process of acquiring magnetic field data. Real-time magnetic field intensity change sequence refers to the time-series data of magnetic field changes. Frequency component refers to the frequency components in the spectrum. Preset sampling multiple coefficient refers to the coefficient used to calculate the sampling frequency. Control register refers to the register that controls the sensor parameters. Sampling interval time refers to the time interval between two samplings. Real-time raw magnetic field intensity reading refers to the raw measured value of the magnetic field intensity.

[0044] In this embodiment, the system first integrates a non-contact worm gear torque sensor on the worm gear drive shaft: the sensor deployment module fixes the non-contact worm gear torque sensor to the surface of the drive shaft, and its internal Hall element maintains a sensing distance from the magnetic field of the shaft surface. The feature triggering module reads the spectral intensity value recorded in the fluctuation spectrum features extracted in step 101 in real time, and compares the spectral intensity value of each frequency component with a preset trigger threshold (set according to the equipment safety operation standard) through a threshold comparator. When the spectral intensity value of any frequency component exceeds the preset trigger threshold, the instruction generator generates an activation instruction for the non-contact worm gear torque sensor, starting the subsequent dynamic sampling process.

[0045] Subsequently, the system initiates the sensor's dynamic scanning operation based on the activation command: after receiving the activation command, the command parser controls the magnetic field scanning unit of the non-contact worm gear torque sensor to perform a full-range magnetic field scan on the surface of the worm gear drive shaft. The reference acquisition unit records the original distribution state of the magnetic field on the shaft surface and generates an initial magnetic field reference reading (i.e., a reference value of the magnetic field strength under no load or steady state). This process is based on magnetic field calibration technology, providing a reference for subsequent real-time change calculations.

[0046] Next, the system sets the sampling frequency parameters based on the out-of-limit frequency components and performs magnetic field sampling: the frequency parameter generator extracts the frequency components exceeding the trigger threshold from the fluctuation spectrum characteristics, multiplies their frequency values ​​by a preset sampling multiple coefficient (such as a multiple of the Nyquist sampling rate), and generates the sampling frequency parameters. The parameter configurator writes these parameters into the control register of the sensor's magnetic field induction probe, setting the probe's sampling interval time. The probe controller starts the magnetic field induction probe to perform high-frequency magnetic field strength acquisition at this interval, outputting the raw real-time magnetic field strength reading. The change calculator calculates the difference between each real-time reading and the initial magnetic field reference reading to obtain the real-time magnetic field strength change. The sequence builder arranges all changes by timestamp, forming a real-time magnetic field strength change sequence, completely recording the dynamic fluctuations of torque.

[0047] Finally, the system integrates the baseline readings and the change sequence to form a dynamic sampling strategy: the strategy synthesis engine uses the initial magnetic field baseline reading as a static benchmark and correlates it with the dynamic fluctuation data represented by the real-time magnetic field intensity change sequence. The data encapsulator binds the two into a structured data packet, generating a dynamic sampling strategy that includes a time-magnetic field change mapping relationship. This strategy is subsequently used to drive the sensor to switch sampling modes on demand (e.g., low-frequency sampling in steady state, high-frequency sampling in abnormal state), achieving resource optimization and accurate monitoring.

[0048] In practical applications, in the dynamic load compensation scenario of multi-spindle thread cutting on a threading machine, a non-contact worm gear torque sensor (a wireless measurement device based on the principle of magnetoelectric induction) is integrated on the worm gear drive shaft of the threading machine. The dynamic sampling strategy is triggered by the fluctuation spectrum characteristics extracted in step 101 (such as specific frequency components reflecting abnormal wear of the drive shaft)—reading the spectral intensity values ​​recorded in the fluctuation spectrum characteristics. When the spectral intensity value of a certain frequency component exceeds a preset trigger threshold (such as a sudden increase in harmonic intensity caused by fatigue cracks on the worm gear tooth surface), an activation command for the non-contact worm gear torque sensor is generated (step 1021). Subsequently, based on the activation command, the non-contact worm gear torque sensor initiates a dynamic scanning operation of the magnetic field on the surface of the worm gear drive shaft, and an initial magnetic field reference reading under no-load conditions is acquired through a Hall probe. Step 1022); Next, extract the frequency components exceeding the trigger threshold from the fluctuation spectrum characteristics (such as high-frequency disturbances related to the worm gear meshing clearance), combine the frequency components with the preset sampling multiplier coefficient to generate sampling frequency parameters, write the sampling frequency parameters into the magnetic field induction probe control register of the non-contact worm gear torque sensor to set the probe sampling interval time, start the magnetic field induction probe to perform magnetic field strength acquisition operation according to the sampling interval time, record the real-time magnetic field strength raw reading, calculate the difference between the real-time magnetic field strength raw reading and the initial magnetic field reference reading as the real-time magnetic field strength change, and arrange them in chronological order to form a real-time magnetic field strength change sequence (Step 1023); Finally, integrate the initial magnetic field reference reading and the real-time magnetic field strength change sequence to generate a complete dynamic sampling strategy (Step 1024). This strategy triggers sensor activation by exceeding the spectral intensity value limit, and the adaptive setting of sampling frequency parameters ensures accurate capture of high-frequency disturbances. The real-time magnetic field intensity change sequence quantifies load transients (such as the instantaneous torque impact of the tool cutting into the workpiece corresponding to the sequence peak), providing a real-time compensation basis for the multi-spindle synchronous servo system. The dynamic sampling strategy drives the slave shaft torque feedforward control, eliminating the master-slave shaft phase deviation caused by worm gear wear. The magnetic field intensity change maps the mechanical transmission stiffness degradation trend, linking with the predictive maintenance system to warn of the lifespan of key components.

[0049] The scheme described in step 102 above achieves dynamic adjustment of the intelligent sensing sampling strategy based on spectral characteristics. An innovative trigger-based dynamic sampling scheme is designed through a linkage mechanism between a non-contact torque sensor and spectral characteristics. This technology employs a frequency component-adaptive sampling parameter generation algorithm to ensure precise matching between magnetic field measurement and torque fluctuations. The innovative dual-track recording mode of reference reading and change not only preserves the static reference benchmark but also fully captures the dynamic change process, providing a high-quality data foundation for torque-magnetic field relationship analysis.

[0050] 103. Record the changes in magnetic field strength and torsional deformation of the worm gear drive shaft under the dynamic sampling strategy, and generate a coupling state identifier of torque and speed of the worm gear drive shaft based on the coupling relationship between the changes in magnetic field strength and torsional deformation.

[0051] Optionally, step 103 may specifically include the following steps: 1031. Mark the target point on the surface of the worm gear drive shaft, and capture the displacement change of the target point as a torsional deformation sequence using a high-speed vision sensor.

[0052] 1032. Align the magnetic field strength change sequence with the torsional deformation sequence at the same time point to form paired data units.

[0053] 1033. Calculate the ratio of the change in magnetic field strength to the torsional deformation in the paired data unit as the coupling coefficient between the change in magnetic field strength and the torsional deformation.

[0054] 1034. When the coupling coefficient exceeds the preset material property calibration threshold, the corresponding time point is marked as an abnormal coupling state point.

[0055] 1035. Integrate the abnormal coupling state points to generate a coupling state identifier that characterizes the abnormal relationship between torque and speed of the worm gear drive shaft.

[0056] In the above scheme, the target point refers to the reference point marked on the worm gear drive shaft. A high-speed vision sensor refers to a sensor that captures images at high speed. Displacement change refers to the change in position. Torsional deformation sequence refers to the time-series data of torsional deformation. Paired data units refer to interrelated data combinations. Coupling coefficient refers to a coefficient reflecting the relationship between magnetic field strength and deformation. Preset material property calibration threshold refers to the critical value for judging anomalies. An abnormal coupling state point refers to the time point of coupling anomalies. A coupling state identifier refers to a symbol identifying the coupling state. The coupling state of torque and speed refers to the correlation between torque and speed.

[0057] In this embodiment, the system first marks the target point on the surface of the worm gear drive shaft: the marking and positioning module uses a high-reflectivity material (such as a reflective ceramic sheet) to create the target point at a specific location on the drive shaft surface, ensuring that it can be stably identified by optical equipment during rotation. The vision capture system uses a high-speed vision sensor (such as a thousand-frame-rate industrial camera) to continuously capture images of the rotating shaft with microsecond-level exposure times, and extracts the spatial coordinates of the target point in each frame image through a sub-pixel edge detection algorithm. The displacement resolver compares the angular displacement deviation of the target point at adjacent time points, calculates the circumferential displacement difference as the torsional deformation, and arranges them in chronological order to form a torsional deformation sequence, quantifying the instantaneous elastic deformation of the shaft under torque.

[0058] Subsequently, the system aligns the magnetic field strength change sequence and the torsional deformation sequence by time point: the time synchronizer receives the real-time magnetic field strength change sequence (from the non-contact torque sensor) and the torsional deformation sequence (from the high-speed vision sensor) generated in step 102, and unifies the time reference of the two data streams through high-precision clock stamp alignment technology. The data pairing engine uses millisecond-level time windows as units to bind the magnetic field strength change and torsional deformation collected at the same moment into paired data units, forming a spatiotemporally correlated data structure. This process is based on multi-source sensor data fusion technology to ensure the synchronization and comparability of physical parameters.

[0059] Next, the system calculates the ratio of the change in magnetic field strength to the torsional deformation in each paired data cell: the coefficient generator traverses each paired data cell, extracting the values ​​of the change in magnetic field strength and the torsional deformation within the cell. The physical relationship analysis module uses this ratio as the coupling coefficient between the change in magnetic field strength and torsional deformation (i.e., the change in magnetic field required per unit deformation), which characterizes the magnetoelastic response of the material under torque load. This process, based on the principle of magnetoelastic effect, establishes a quantitative correlation between magnetic field fluctuations and mechanical deformation through a linear proportional model.

[0060] Then, the system checks whether the coupling coefficient exceeds the preset material property calibration threshold: the threshold comparator calls the preset material property calibration threshold (such as the safe range of magnetoelastic response for copper alloy worm gears) pre-calibrated in the materials laboratory and compares it point by point with the real-time calculated coupling coefficient. When the coefficient exceeds the upper limit of the threshold or falls below the lower limit, the anomaly marking module records the time point as an abnormal coupling state point, indicating that the torque-deformation relationship deviates from the normal physical properties of the material at that moment. This process uses a threshold over-limit triggering mechanism to accurately locate the critical moment of material fatigue or structural anomaly.

[0061] Finally, the system integrates abnormal coupled state points to generate coupled state identifiers: the identifier synthesis engine summarizes the timestamps and deviation data of all abnormal coupled state points, and merges discrete abnormal points into continuous abnormal intervals according to temporal continuity using a state clustering algorithm. The data encapsulator generates coupled state identifiers containing the abnormal period, peak deviation, and duration. These identifiers use binary encoding or color labels (such as red warnings) to characterize the abnormal torque-speed correlation of the worm gear drive shaft, providing a structured anomaly report for subsequent maintenance decisions.

[0062] In practical applications, in the health monitoring scenario of worm gear drive shafts in multi-spindle thread cutting on thread turning machines, a dynamic sampling strategy is implemented for the worm gear drive shaft during the operation of the thread turning machine. First, a target point (such as circumferentially distributed sub-millimeter-level reflective marks) is laser-etched on the surface of the drive shaft. The circumferential displacement change of the target point is captured at a microsecond-level frame rate by a high-speed vision sensor, generating a torsional deformation sequence under axial torque load (step 1031). Simultaneously, the real-time magnetic field strength change sequence output in step 102 is acquired. The two are aligned with the timestamps of the acquisition system clock signal to form paired data units (each unit contains the magnetic field strength change and torsional deformation at the same moment) (step 1). 032); Calculate the instantaneous ratio of the change in magnetic field strength to the torsional deformation based on the paired data units as the coupling coefficient between the change in magnetic field strength and the torsional deformation (this coefficient reflects the magnetic-force response characteristics of the material) (step 1033); When the coupling coefficient exceeds the preset material characteristic calibration threshold (according to the experimental calibration of the magnetic permeability and elastic modulus of the worm gear material), mark this moment as an abnormal coupling state point (such as a sudden drop in coefficient indicating magnetic-force decoupling caused by material fatigue) (step 1034); Finally, integrate the time distribution and intensity of all abnormal coupling state points to generate coupling state identifiers (such as green / yellow / red three-level identifiers corresponding to normal / warning / fault states) (step 1035). This identifier reveals mechanical degradation through the abnormal correlation between the torsional deformation sequence and the change in magnetic field strength: abnormal displacement at the target point (such as a sudden increase in axial torsion) exposes the propagation of microcracks in the drive shaft, and the threshold of the coupling relationship coefficient exceeds the standard, triggering a material failure warning (such as the plastic deformation in the stress concentration area of ​​the worm gear tooth root corresponding to the red identifier). The coupling state identifier is input into the multi-spindle synchronous servo system to correct the slave shaft torque command in real time to compensate for the stiffness decay of the main drive shaft and avoid multi-spindle synchronization misalignment caused by the performance degradation of a single shaft.

[0063] The scheme described in step 103 above enables online identification and anomaly detection of torque-deformation coupling states. By fusing multimodal data from magnetic field strength and visual measurements, an innovative coupling relationship model reflecting material properties is constructed. This technology employs a time-aligned paired data analysis method to accurately quantify the dynamic correlation strength between magnetic field changes and mechanical deformation. Through intelligent comparison using preset material thresholds, automatic marking of abnormal coupling states is achieved, providing a new technical approach for assessing the health status of transmission systems.

[0064] 104. Obtain the phase offset feature of the fluctuation spectrum feature, locate the hysteresis risk zone where the phase offset feature exceeds the preset synchronization tolerance based on the coupling state identifier, and extract the phase offset amount corresponding to the hysteresis risk zone as the main axis dynamic compensation vector.

[0065] Optionally, step 104 may specifically include the following steps: 1041. Extract the phase angle change values ​​corresponding to the frequency components recorded in the wave spectrum features, and arrange the phase angle change values ​​in chronological order to form a phase offset feature sequence.

[0066] 1042. Within the time interval corresponding to the abnormal coupling state point marked by the coupling state identifier, locate the phase angle change value at the same time point in the phase offset feature sequence.

[0067] 1043. Compare the phase angle change value with a preset synchronization tolerance threshold. When the phase angle change value exceeds the synchronization tolerance threshold, mark the corresponding time point as a lag risk point.

[0068] 1044. Integrate the time intervals where continuous lag risk points are located, and mark the working stage of the worm gear drive shaft covered by the time interval as the lag risk zone.

[0069] 1045. Extract the average value of the phase angle change within the hysteresis risk zone as the principal axis dynamic compensation vector.

[0070] In the above scheme, phase offset characteristic refers to the characteristic of phase change. Hysteresis risk zone refers to the area where hysteresis risk exists. Main spindle dynamic compensation vector refers to the vector used for dynamic compensation. Phase angle change value refers to the amount of change in phase angle. Phase offset characteristic sequence refers to the time-series data of phase offset. Preset synchronization tolerance threshold refers to the critical value for judging synchronization anomalies. Hysteresis risk point refers to the time point where hysteresis risk exists. Working stage refers to the working state of the worm gear drive shaft. Average value refers to the average of the values.

[0071] In this embodiment, the system first extracts the phase angle change values ​​corresponding to the frequency components recorded in the wave spectrum features: the phase resolver reads the phase information (i.e., the angle offset of the signal in the time-frequency domain) of each frequency component from the wave spectrum feature set generated in step 101. The sequence builder arranges all phase angle change values ​​in the order of the timestamps of the original signal to form a time-continuous phase offset feature sequence. This process is based on Fourier phase dewinding technology to eliminate phase jump interference and ensure the continuity of angle changes.

[0072] Subsequently, the system locates the same time point in the phase offset feature sequence within the time interval corresponding to the abnormal coupling state point marked by the coupling state identifier: the spatiotemporal alignment engine receives the coupling state identifier (containing the time interval information of the abnormal coupling state point) generated in step 103 and performs precise time-axis matching with the phase offset feature sequence. The data locator extracts the phase angle change value corresponding to the time point that completely coincides with the abnormal coupling state point in the phase offset feature sequence. This process ensures millisecond-level alignment between the magnetic-elastic anomaly period and the phase offset data through high-precision clock synchronization technology.

[0073] Next, the system compares the phase angle change value with the preset synchronization tolerance threshold and marks the hysteresis risk point: the threshold comparator calls the preset synchronization tolerance threshold calibrated by the engineering. The risk marking module compares the located phase angle change value with the threshold point by point: if the phase angle change value exceeds the upper limit of the threshold or is lower than the lower limit, the time point is recorded as a hysteresis risk point. This process uses out-of-bounds trigger logic to accurately identify the critical moment when the drive shaft torque-speed loses synchronization.

[0074] Then, the system integrates the time intervals containing continuous lag risk points and marks them as lag risk regions: an interval clustering algorithm scans all discrete lag risk points and merges adjacent points based on time continuity. The region marker defines the merged continuous time period as a lag risk region, representing that the worm gear drive shaft is continuously in a torque transmission delay state during this time period. This process uses time window sliding detection technology to avoid misjudgment due to instantaneous noise and focus on the true risk period.

[0075] Finally, the system extracts the average value of the phase angle change within the hysteresis risk zone as the master shaft dynamic compensation vector: the vector generator calculates the arithmetic mean of all phase angle change values ​​within the marked hysteresis risk zone time range. The data encapsulator outputs this average value as the master shaft dynamic compensation vector, quantifying the average phase hysteresis of the drive shaft within this risk zone. This process uses moving average filtering technology to suppress fluctuation noise, ensuring the stability and representativeness of the compensation vector.

[0076] In practical applications, in the phase compensation control scenario of multi-spindle thread cutting on a threading machine, when the multi-spindle of the threading machine is synchronously machining high-hardness alloy materials, based on the fluctuation spectrum characteristics obtained in step 101 (such as specific frequency components reflecting abnormal spindle load), the phase angle change values ​​corresponding to the frequency components recorded in the fluctuation spectrum characteristics (such as the instantaneous phase offset of the worm gear meshing frequency component) are extracted, and the phase angle change values ​​are arranged in chronological order to form a phase offset feature sequence (step 1041); then, based on the coupling state identifier generated in step 103 (such as a red identifier marking abnormal magnetic-force decoupling of the worm gear drive shaft), within the time interval corresponding to the abnormal coupling state point marked by the coupling state identifier (such as when local wear of the worm gear causes...), (Step 1042) During periods of abnormal material stress, locate the phase angle change value at the same time point in the phase offset feature sequence; then compare the phase angle change value with the preset synchronization tolerance threshold (this threshold is set according to the multi-axis synchronization accuracy requirements of the thread cutting machine). When the phase angle change value is detected to exceed the synchronization tolerance threshold (e.g., the phase lag exceeds the safety margin), mark the corresponding time point as a lag risk point (Step 1043); further integrate the time intervals where continuous lag risk points are located, and mark the working stage of the worm gear drive shaft covered by the time interval as the lag risk area (Step 1044); finally, extract the average value of the phase angle change value in the lag risk area as the main shaft dynamic compensation vector (Step 1045). This vector achieves precise positioning through the linkage of phase offset feature sequence and coupling state identifier: abnormal coupling state point exposes phase drift caused by worm gear wear (such as increased transmission clearance causing lag in the slave shaft response), the lag risk area maps the synchronization failure period under high load conditions (such as master-slave phase mismatch during the thread finishing stage), and the master shaft dynamic compensation vector quantifies the amount of phase offset that needs to be compensated (such as a negative value indicating that the slave shaft needs to be adjusted ahead), inputs the master-slave control architecture of the multi-axis servo system, and corrects the slave shaft phase command in real time to eliminate synchronization error.

[0077] The scheme described in step 104 above achieves precise location of phase lag risk and calculation of compensation. Through correlation analysis of spectral phase and coupling state, it innovatively identifies lag risk regions affecting synchronization accuracy. This technology employs a judgment algorithm combining tolerance thresholds and continuous intervals, ensuring both the sensitivity of risk identification and the reliability of the result determination. The extracted average phase offset is used as a dynamic compensation vector, providing a precise adjustment basis for subsequent synchronization control.

[0078] 105. Analyze the main spindle dynamic compensation vector to generate a synchronization error suppression command, and adjust the output torque command of each servo axis of the thread cutting machine in real time according to the synchronization error suppression command to suppress the multi-spindle synchronization error caused by the worm gear drive lag.

[0079] Optionally, step 105 may specifically include the following steps: 1051. Combine the spindle dynamic compensation vector with a preset torque correction coefficient to generate a torque correction value.

[0080] 1052. Obtain the current output torque command of each servo axis of the thread cutting machine, and superimpose the torque correction value onto the output torque command of the corresponding servo axis to form an updated output torque command.

[0081] 1053. Generate a synchronization error suppression command based on the updated output torque command, and transmit the synchronization error suppression command to the thread cutting machine servo axis controller.

[0082] 1054. By executing the synchronization error suppression command through the servo axis controller, the output torque of each servo axis is adjusted according to the synchronization error suppression command to suppress the multi-spindle synchronization error caused by the worm gear drive lag.

[0083] In the above scheme, the synchronization error suppression command refers to the control command that suppresses synchronization error. A servo axis refers to a controlled motion axis. The output torque command refers to the command that controls the torque output. The torque correction value refers to the amount of torque adjustment. The preset torque correction coefficient refers to the coefficient used to calculate the correction value. The servo axis controller refers to the device that controls the servo axis. Multi-spindle synchronization error refers to the synchronization deviation between multiple spindles. Worm gear drive hysteresis refers to the delay phenomenon of worm gear drive.

[0084] In this embodiment, the system first combines the spindle dynamic compensation vector with a preset torque correction coefficient: the coefficient matching module receives the spindle dynamic compensation vector (quantifying the average value of the phase lag of the worm gear drive shaft) generated in step 104, and calls the preset torque correction coefficient (a pre-set ratio value based on material stiffness and transmission efficiency) calibrated in engineering. The correction value generator multiplies the two values ​​to generate a torque correction value, which represents the torque increment or decrement required to compensate for the phase lag. This process is based on dynamic scaling technology to ensure a linear correspondence between the compensation amount and the degree of lag.

[0085] Subsequently, the system acquires the current output torque commands of each servo axis of the thread-cutting machine and superimposes torque correction values: the command acquisition unit reads the raw data of the output torque commands of all servo axes of the thread-cutting machine in real time. The command superposition engine maps the torque correction values ​​generated in step 1051 to the corresponding servo axes according to the axis number, and adds them to the original command values ​​through arithmetic superposition operations to form the updated output torque commands. This process adopts dynamic command injection technology to ensure that the correction values ​​are seamlessly integrated into the control flow, while maintaining the independence and synchronization of commands for each axis.

[0086] Next, the system generates a synchronization error suppression command based on the updated output torque command and transmits it to the controller: the command compiler parses the updated output torque command set for all servo axes and converts it into a binary control code stream using the command encoding protocol. The command distributor transmits the encoded data packet as a synchronization error suppression command to each servo axis controller of the thread-cutting machine via the real-time industrial Ethernet bus. This process relies on low-latency communication technology to ensure that the command is synchronously distributed across controllers within milliseconds.

[0087] Finally, the system executes synchronization error suppression commands through the servo axis controller to adjust the output torque: the command execution unit receives the synchronization error suppression command within the servo axis controller and converts the torque command into a quadrature-axis current compensation amount (based on the motor electromagnetic torque equation) through the current loop solver. The power drive module injects the compensation amount into the inverter modulation signal to adjust the output torque of each servo motor in real time. This process is based on direct torque compensation technology, enabling the worm gear drive shaft to obtain torque correction instantaneously during load changes, effectively suppressing multi-spindle synchronization errors caused by worm gear drive lag.

[0088] In practical applications, during the process of machining oil pipe threads on a multi-spindle threading machine to suppress synchronous errors, when a phase lag is detected due to local wear of the worm gear drive shaft (the spindle dynamic compensation vector output in step 104 is negative), the system analyzes the spindle dynamic compensation vector (this vector quantifies the phase offset that the spindle needs to compensate for), combines the spindle dynamic compensation vector with a preset torque correction coefficient (this coefficient is calibrated based on the worm gear material stiffness and transmission ratio), and generates a torque correction value (step 1051). Subsequently, it obtains the current output torque command of each servo axis of the threading machine (such as the preset torque value of the slave axis during the thread cutting stage), and adds the torque correction value to the output torque of the corresponding servo axis. On the torque command (such as increasing torque output to the slave axis in the hysteresis risk zone), an updated output torque command is formed (step 1052); then, a synchronization error suppression command (the command includes the torque adjustment amplitude and the duration of action) is generated based on the updated output torque command, and the synchronization error suppression command is transmitted to the thread cutting machine servo axis controller via real-time industrial Ethernet (step 1053); finally, the synchronization error suppression command is executed by the servo axis controller, so that the output torque of each servo axis is adjusted according to the synchronization error suppression command (such as increasing the slave axis torque to compensate for the mechanical hysteresis of the main spindle worm gear drive), suppressing the multi-main spindle synchronization error caused by the worm gear drive hysteresis (such as eliminating the pitch deviation in the thread finishing stage) (step 1054). This process achieves active error suppression through the dynamic injection of torque correction values: the spindle dynamic compensation vector is derived from the phase offset characteristics (refer to the hysteresis risk zone location in step 104), the preset torque correction coefficient maps the mechanical transmission characteristics (such as the nonlinear relationship between worm gear wear and torque compensation), the updated output torque command drives the servo axis controller to perform feedforward compensation, and the real-time execution of the synchronization error suppression command ensures that multiple spindles maintain strict synchronization throughout the thread machining process (refer to the error elimination principle of cross-coupling control).

[0089] The solution described in step 105 above achieves closed-loop suppression and dynamic compensation of multi-spindle synchronization errors. An innovative real-time error suppression system is constructed through intelligent conversion of compensation vectors to torque commands. This technology employs a command-superimposed torque adjustment scheme, achieving precise compensation without affecting the original control architecture. The innovative servo-axis collaborative control mechanism effectively suppresses multi-axis synchronization errors caused by worm gear drive lag, significantly improving the machining accuracy and stability of the thread turning machine.

[0090] The following are specific examples for steps 101 to 105: In the scenario of worm gear wear compensation for oil pipe threading machines, the worm gear drive shaft of the threading machine experiences tooth surface wear due to long-term high-load operation during the oil pipe threading process. The system first collects dynamic torque fluctuation data of the worm gear drive shaft in real time (capturing the torque transients during shaft rotation through a magnetoelectric sensor), and arranges the torque change values ​​in chronological order to form a dynamic torque fluctuation data sequence. Then, the sequence is divided into multiple data segments according to a fixed time window, and a Fourier transform is performed on each data segment to generate a spectral distribution fragment, which is then spliced ​​into a torque spectrum matrix in chronological order. Next, the spectral intensity values ​​of the torque spectrum matrix are extracted, and frequency components with spectral intensity values ​​exceeding a preset background noise threshold are identified and their spectral intensity values ​​are recorded as fluctuation spectrum features. Based on this characteristic, a non-contact worm gear torque sensor is integrated on the worm gear drive shaft. When the spectral intensity value of a specific frequency component exceeds a preset trigger threshold, an activation command is generated. According to the activation command, the non-contact worm gear torque sensor starts a dynamic scanning operation of the magnetic field on the surface of the worm gear drive shaft, acquiring the initial magnetic field reference reading under no-load conditions. The frequency component exceeding the trigger threshold in the fluctuation spectrum characteristics is extracted and combined with a preset sampling multiple coefficient to generate a sampling frequency parameter, which is written into the magnetic field induction probe control register to set the probe sampling interval time. The magnetic field induction probe is started to perform magnetic field intensity acquisition operation according to the sampling interval time, recording the real-time magnetic field intensity raw reading. The difference between the original reading and the initial magnetic field reference reading is calculated as the real-time magnetic field intensity change, and sorted by time to form a real-time magnetic field intensity change sequence. Finally, the initial magnetic field reference reading and the real-time magnetic field intensity change sequence are integrated to form a dynamic sampling strategy. When implementing this strategy, a target point is laser-marked on the drive shaft surface. A high-speed vision sensor captures the displacement change at the target point, generating a torsional deformation sequence. The magnetic field strength change sequence and the torsional deformation sequence are aligned by timestamps to form paired data units. The ratio of the two in the unit is calculated as the coupling coefficient between magnetic field strength change and torsional deformation. When the coefficient exceeds a preset material characteristic calibration threshold (calibrated based on the bronze material of the worm gear), that moment is marked as an abnormal coupling state point (e.g., a sudden drop in the coefficient indicates plastic deformation of the material). All abnormal points are integrated to generate coupling state identifiers (e.g., a red identifier corresponds to tooth root fatigue failure). Then, the phase angle change values ​​corresponding to the frequency components in the wave spectrum are extracted and sorted by time to form a phase offset feature sequence. Based on the coupling state identifiers, the phase angle change values ​​within the time period corresponding to the abnormal coupling state point are located and compared with a preset synchronization tolerance threshold (set according to thread accuracy requirements). When the phase lag exceeds the limit, a lag risk point is marked. The time periods containing consecutive lag points are integrated as lag risk zones (e.g., the final stage of pipe threading), and the average value of the phase angle change values ​​within this zone is extracted as the spindle dynamic compensation vector.Finally, the dynamic compensation vector of the main spindle is analyzed (negative values ​​indicate that the slave axis needs to be compensated in advance), and it is combined with the preset torque correction coefficient (calibrated according to the transmission stiffness) to generate the torque correction value; the current output torque command of each servo axis of the thread cutting machine is obtained, and the correction value is superimposed on the command of the corresponding slave axis to form the updated output torque command; based on this, a synchronization error suppression command is generated and transmitted to the servo axis controller via real-time industrial Ethernet; the controller executes the synchronization error suppression command, adjusts the output torque of the slave axis (such as increasing the torque of the slave axis to offset the wear lag of the main spindle worm gear), suppresses the multi-spindle synchronization error caused by the worm gear transmission lag, and ensures the consistency of the thread pitch.

[0091] Figure 2 This application provides a schematic diagram of the structure of a multi-spindle synchronous servo control system for a thread-cutting machine, as shown in the embodiment of the present application. Figure 2 As shown, the system includes: The acquisition module 21 is used to acquire dynamic torque fluctuation data of the worm gear drive shaft of the thread cutting machine in real time, perform spectral transformation on the dynamic torque fluctuation data to generate a torque spectrum matrix, and extract fluctuation spectrum features from the torque spectrum matrix. Trigger module 22 is used to integrate a non-contact worm gear torque sensor on the worm gear drive shaft, and trigger the non-contact worm gear torque sensor based on the fluctuation spectrum characteristics to execute a dynamic sampling strategy; The generation module 23 is used to record the changes in magnetic field strength and torsional deformation of the worm gear drive shaft under the dynamic sampling strategy, and generate a coupling state identifier of torque and speed of the worm gear drive shaft based on the coupling relationship between the changes in magnetic field strength and torsional deformation. The positioning module 24 is used to acquire the phase offset feature of the fluctuation spectrum feature, locate the hysteresis risk area where the phase offset feature exceeds the preset synchronization tolerance based on the coupling state identifier, and extract the phase offset amount corresponding to the hysteresis risk area as the main axis dynamic compensation vector. The adjustment module 25 is used to parse the main spindle dynamic compensation vector to generate a synchronization error suppression command, and adjust the output torque command of each servo axis of the thread cutting machine in real time according to the synchronization error suppression command to suppress the multi-spindle synchronization error caused by the worm gear drive lag.

[0092] Figure 2 The aforementioned multi-spindle synchronous servo control system for a thread-cutting machine can execute... Figure 1 The implementation principle and technical effects of the multi-spindle synchronous servo control method for a thread-cutting machine described in the illustrated embodiment will not be repeated here. The specific methods by which each module and unit performs operations in the multi-spindle synchronous servo control system for a thread-cutting machine described in the above embodiments have been described in detail in the embodiments related to this method, and will not be elaborated upon here.

[0093] In one possible design,Figure 2 The multi-spindle synchronous servo control system for a thread-cutting machine shown in the embodiment can be implemented as a computing device, such as... Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32; The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are invoked and executed by the processing component 32.

[0094] The processing component 32 is used for the above Figure 1 The embodiment describes a multi-spindle synchronous servo control method for a thread-cutting machine.

[0095] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above-described method. Alternatively, the processing component may be implemented as one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above-described method.

[0096] Storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0097] Of course, computing devices may also include other components, such as input / output interfaces, display components, communication components, etc.

[0098] Input / output interfaces provide interfaces between processing components and peripheral interface modules, which can be output devices, input devices, etc.

[0099] The communication components are configured to facilitate wired or wireless communication between computing devices and other devices.

[0100] The computing device can be a physical device or an elastic computing host provided by a cloud computing platform. In this case, the computing device can refer to a cloud server, and the aforementioned processing components, storage components, etc., can be basic server resources rented or purchased from the cloud computing platform.

[0101] This application also provides a computer storage medium storing a computer program, which, when executed by a computer, can perform the above-described functions. Figure 1The embodiment shown illustrates a multi-spindle synchronous servo control method for a thread-cutting machine.

[0102] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0103] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0104] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0105] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A multi-spindle synchronous servo control method for a thread-cutting machine, characterized in that, include: The dynamic torque fluctuation data of the worm gear drive shaft of the thread cutting machine is collected in real time. The dynamic torque fluctuation data is subjected to spectrum transformation to generate a torque spectrum matrix, and the fluctuation spectrum features are extracted from the torque spectrum matrix. A non-contact worm gear torque sensor is integrated on the worm gear drive shaft, and the non-contact worm gear torque sensor is triggered based on the fluctuation spectrum characteristics to execute a dynamic sampling strategy; Record the changes in magnetic field strength and torsional deformation of the worm gear drive shaft under the dynamic sampling strategy, and generate a coupling state identifier of torque and speed of the worm gear drive shaft based on the coupling relationship between the changes in magnetic field strength and torsional deformation. The phase offset feature of the fluctuation spectrum is obtained, and the hysteresis risk zone where the phase offset feature exceeds the preset synchronization tolerance is located based on the coupling state identifier. The phase offset corresponding to the hysteresis risk zone is extracted as the main axis dynamic compensation vector. The main spindle dynamic compensation vector is analyzed to generate a synchronization error suppression command. The output torque command of each servo axis of the thread cutting machine is adjusted in real time according to the synchronization error suppression command to suppress the multi-spindle synchronization error caused by the worm gear drive lag.

2. The method according to claim 1, characterized in that, Real-time acquisition of dynamic torque fluctuation data of the worm gear drive shaft of the thread-cutting machine; spectral transformation of the dynamic torque fluctuation data to generate a torque spectrum matrix; and extraction of fluctuation spectrum features from the torque spectrum matrix, including: The torque change value during the rotation of the worm gear drive shaft of the thread cutting machine is continuously collected, and the torque change value is arranged in time sequence to form a dynamic torque fluctuation data sequence. The dynamic torque fluctuation data sequence is divided into multiple data segments according to a fixed time window, and a time-domain to frequency-domain conversion operation is performed on each data segment to generate a spectral distribution segment. The spectral distribution segments are spliced ​​together in chronological order to form a torque spectrum matrix, and the spectral intensity values ​​of the torque spectrum matrix are extracted. The frequency components whose spectral intensity values ​​exceed a preset background noise threshold are identified, and the spectral intensity values ​​corresponding to the frequency components are recorded as fluctuation spectral features.

3. The method according to claim 1, characterized in that, A non-contact worm gear torque sensor is integrated on the worm gear drive shaft, and the non-contact worm gear torque sensor is triggered based on the fluctuation spectrum characteristics to execute a dynamic sampling strategy, including: A non-contact worm gear torque sensor is integrated on the worm gear drive shaft, and the spectral intensity value recorded in the fluctuation spectrum characteristics is read. When the spectral intensity value exceeds a preset trigger threshold, an activation command for the non-contact worm gear torque sensor is generated. According to the activation command, the non-contact worm gear torque sensor is activated to perform a dynamic scanning operation on the magnetic field of the worm gear drive shaft surface to acquire an initial magnetic field reference reading. The sampling frequency parameter is set according to the frequency component whose spectral intensity value exceeds the trigger threshold, and the magnetic field sensing probe of the non-contact worm gear torque sensor is controlled to perform magnetic field sampling operation according to the sampling frequency parameter to record the real-time magnetic field intensity change sequence. The initial magnetic field reference readings are integrated with the real-time magnetic field intensity change sequence to form a dynamic sampling strategy.

4. The method according to claim 3, characterized in that, The sampling frequency parameter is set according to the frequency component whose spectral intensity value exceeds the trigger threshold, and the magnetic field sensing probe of the non-contact worm gear torque sensor is controlled to perform magnetic field sampling operation according to the sampling frequency parameter to record the real-time magnetic field intensity change sequence, including: Extract the frequency components exceeding the trigger threshold from the fluctuation spectrum features, and combine the frequency components with a preset sampling multiple coefficient to generate sampling frequency parameters; The sampling frequency parameter is written into the magnetic field induction probe control register of the non-contact worm gear torque sensor to set the probe sampling interval time. The magnetic field sensing probe of the non-contact worm gear torque sensor is activated to perform magnetic field strength acquisition operation at the sampling interval, and the real-time raw magnetic field strength reading output by the magnetic field sensing probe is recorded at the same time. The difference between the original real-time magnetic field strength reading and the initial magnetic field reference reading is calculated as the change in real-time magnetic field strength. All real-time magnetic field intensity changes are arranged in chronological order of sampling time to form a sequence of real-time magnetic field intensity changes.

5. The method according to claim 1, characterized in that, The changes in magnetic flux density and torsional deformation of the worm gear drive shaft under the dynamic sampling strategy are recorded, and a coupling state identifier for the torque and rotational speed of the worm gear drive shaft is generated based on the coupling relationship between the changes in magnetic flux density and torsional deformation, including: Mark the target point on the surface of the worm gear drive shaft, and capture the displacement change of the target point as a torsional deformation sequence using a high-speed vision sensor. The magnetic field strength change sequence and the torsional deformation sequence are aligned at the same time point to form paired data units; The ratio of the change in magnetic field strength to the torsional deformation in the paired data unit is calculated as the coupling coefficient between the change in magnetic field strength and the torsional deformation; When the coupling coefficient exceeds the preset material property calibration threshold, the corresponding time point is marked as an abnormal coupling state point. The abnormal coupling state points are integrated to generate a coupling state identifier that characterizes the abnormal relationship between torque and speed of the worm gear drive shaft.

6. The method according to claim 1, characterized in that, The phase shift characteristics of the fluctuation spectrum are obtained, and the hysteresis risk region where the phase shift characteristics exceed the preset synchronization tolerance is located based on the coupling state identifier. The phase shift amount corresponding to the hysteresis risk region is extracted as the principal axis dynamic compensation vector, including: Extract the phase angle change values ​​corresponding to the frequency components recorded in the wave spectrum features, and arrange the phase angle change values ​​in chronological order to form a phase shift feature sequence; Within the time interval corresponding to the abnormal coupling state point marked by the coupling state identifier, locate the phase angle change value at the same time point in the phase offset feature sequence; The phase angle change value is compared with a preset synchronization tolerance threshold. When the phase angle change value exceeds the synchronization tolerance threshold, the corresponding time point is marked as a lag risk point. Integrate the time intervals containing consecutive lag risk points, and mark the working stages of the worm gear drive shaft covered by the time intervals as lag risk zones; The average value of the phase angle change within the hysteresis risk zone is extracted as the principal axis dynamic compensation vector.

7. The method according to claim 1, characterized in that, The dynamic compensation vector of the main spindle is analyzed to generate a synchronization error suppression command. Based on this command, the output torque commands of each servo axis of the thread cutting machine are adjusted in real time to suppress multi-spindle synchronization errors caused by worm gear drive lag. This includes: The spindle dynamic compensation vector is combined with a preset torque correction coefficient to generate a torque correction value; Obtain the current output torque command of each servo axis of the thread cutting machine, and superimpose the torque correction value onto the output torque command of the corresponding servo axis to form an updated output torque command; A synchronization error suppression command is generated based on the updated output torque command, and the synchronization error suppression command is transmitted to the servo axis controller of the thread cutting machine. The servo axis controller executes a synchronization error suppression command, which adjusts the output torque of each servo axis according to the synchronization error suppression command to suppress the multi-spindle synchronization error caused by the worm gear drive lag.

8. A multi-spindle synchronous servo control system for a thread-cutting machine, characterized in that, include: The acquisition module is used to acquire dynamic torque fluctuation data of the worm gear drive shaft of the thread cutting machine in real time, perform spectral transformation on the dynamic torque fluctuation data to generate a torque spectrum matrix, and extract fluctuation spectrum features from the torque spectrum matrix. A triggering module is used to integrate a non-contact worm gear torque sensor on the worm gear drive shaft, and to trigger the non-contact worm gear torque sensor based on the fluctuation spectrum characteristics to execute a dynamic sampling strategy; The generation module is used to record the changes in magnetic field strength and torsional deformation of the worm gear drive shaft under the dynamic sampling strategy, and generate a coupling state identifier of torque and speed of the worm gear drive shaft based on the coupling relationship between the changes in magnetic field strength and torsional deformation. The positioning module is used to acquire the phase offset features of the fluctuation spectrum features, locate the hysteresis risk zone where the phase offset features exceed the preset synchronization tolerance based on the coupling state identifier, and extract the phase offset amount corresponding to the hysteresis risk zone as the main axis dynamic compensation vector. The adjustment module is used to parse the main spindle dynamic compensation vector to generate a synchronization error suppression command, and adjust the output torque command of each servo axis of the thread cutting machine in real time according to the synchronization error suppression command to suppress the multi-spindle synchronization error caused by worm gear drive lag.

9. A computing device, characterized in that, It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are invoked and executed by the processing component to implement a multi-spindle synchronous servo control method for a thread-cutting machine as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that, The device contains a computer program that, when executed by a computer, implements a multi-spindle synchronous servo control method for a thread-cutting machine as described in any one of claims 1 to 7.

Images