A power tong control system with adaptive torque control function and method thereof

Abstract

The application provides a power tong control system with an adaptive torque control function and a method thereof, and belongs to the technical field of intelligent control of pipe connection equipment. The method comprises the following steps: collecting original acoustic signals, synchronously collecting original vibration signals, performing band-pass filtering and power frequency interference removal processing on the two types of original signals respectively, and generating a noise-reduced acoustic vibration basic data set; performing time domain feature extraction and frequency domain decomposition processing on the acoustic vibration basic data set, separating out thread engagement micro-bulge scraping characteristic components and transmission chain torsional vibration characteristic components, and generating an acoustic vibration separation characteristic data set; and by integrating acoustic emission and torsional vibration decoupling technology into the whole process control, the identification and response speed of the power tong to the precursors of misalignment, misconnection and slipping teeth is greatly improved, and the occurrence probability of pipe thread sticking faults is effectively reduced.

Description

Technical Field

[0001] This invention proposes a power clamp control system and method with adaptive torque control function, belonging to the field of intelligent control technology for pipe connection equipment. Background Technology

[0002] In current oil drilling and workover operations, power tongs are the core equipment for uncoupling operations on the tubing string, and their control accuracy directly determines the quality of the pipe thread connection and the safety of downhole operations. Currently, most mainstream power tongs use a conversion method of multiplying the system oil pressure by the hydraulic motor displacement to obtain the torque value. This method does not cover actual operating condition variables such as slippage between the jaw plate and the tubing body, gear transmission backlash, and pressure loss along the hydraulic lines. The final output indicated torque can deviate from the actual torque of the threaded pair by 18% to 25%, directly causing the optimal shoulder torque specified by the API standard to be systematically and artificially inflated, leading to frequent pipe thread sticking failures after long-term operation.

[0003] Current power tongs rely on manual experience to identify the threading status during the threading process. They can only detect obvious threading errors through operating sounds or physical deviations of the tongs, and cannot intervene in the early stages of thread misalignment. Automated models continue to apply pressure without identifying thread misalignment, causing plowing of the male and female thread sides. The resulting metal debris remains on the sealing surface, creating a potential leak for subsequent downhole operations.

[0004] Conventional power clamps employ a fixed-ratio pressure boosting logic at the end of the threading process, resulting in a rigid impact on the pipe end contact shoulder. This can easily cause micro-plastic crushing of the metal-to-metal sealing surface of special threads such as VAMTOP and FOX. In low-temperature winter operating scenarios, the consistency of thread grease increases significantly with decreasing temperature. Under the same torque input, the threaded pair cannot achieve sufficient creep fit. The existing control system lacks a corresponding compensation mechanism, leading to numerous hidden faults in pipe string connections that appear tight but actually fail to seal properly. This makes on-site quality control extremely difficult. Summary of the Invention

[0005] This invention provides a power caliper control system and method with adaptive torque control function to solve the problems mentioned in the background art above: This invention proposes a power caliper control method with adaptive torque control function, the method comprising: S1. Acquire the original acoustic wave signal and simultaneously acquire the original vibration signal. Perform bandpass filtering and power frequency interference removal on the two types of original signals respectively to generate a noise-reduced acoustic and vibration basic dataset. Perform time-domain feature extraction and frequency-domain decomposition on the acoustic and vibration basic dataset to separate the thread meshing micro-protrusion scraping feature component and the transmission chain torsional vibration feature component to generate an acoustic and vibration separation feature dataset. S2. Perform feature mapping processing on the acoustic-vibration separation feature dataset, match the thread meshing micro-protrusion scraping feature component to the preset acoustic fingerprint library, match the transmission chain torsional vibration feature component to the preset torsional vibration mode library, and generate an initial feature matching result set; perform time-series correlation verification processing on the initial feature matching result set to form a stable acoustic-vibration fusion feature result set; S3. Perform state determination processing on the acoustic vibration fusion feature result set to generate current operating state identification data; perform boundary condition verification processing on the current operating state identification data, and combine the real-time collected clamp rotation speed and hydraulic system flow data to lock the operating range boundary of the current state and generate state range locking data. S4. Perform anomaly feature comparison processing on the state interval locking data, compare the real-time acoustic and vibration features with the preset misalignment, slippage and tooth slippage anomaly feature templates one by one, and generate anomaly feature trigger marker data; perform priority sorting processing on the anomaly feature trigger marker data, divide the response order according to the anomaly hazard level, and generate anomaly response execution sequence data. S5. Perform control command generation processing on the abnormal response execution sequence data, output matching hydraulic valve group adjustment commands corresponding to different abnormal types, and generate real-time dynamic adjustment command data; perform timing calibration processing on the real-time dynamic adjustment command data, match the response window of the current operating state, and generate the final directly executable power clamp control command data. S6. Perform working condition adaptation processing on the final directly executable power clamp control command data to generate refined pressure control data with temperature compensation; perform output rate limiting processing on the refined pressure control data with temperature compensation to constrain the pressure rise gradient and generate final execution control data without rigid impact.

[0006] This invention proposes a power caliper control system with adaptive torque control function, the system comprising: One or more processors; Memory, used to store one or more programs; Wherein, when the one or more programs are executed by the one or more processors, the one or more processors are made to implement the method described in any one of the above.

[0007] The beneficial effects of this invention are as follows: By integrating acoustic emission and torsional vibration decoupling technology into the entire process control, the recognition and response speed of the power clamp for signs of misalignment, incorrect threading, and stripping is significantly improved, effectively reducing the probability of pipe thread sticking failure. Abandoning the traditional fixed-slope pressure boosting logic, and combining it with the thread grease rheological temperature compensation curve, the rigid impact at the moment of shoulder contact is reduced, avoiding the problem of micro-plastic crushing of special thread metal sealing surfaces. A new millisecond-level reverse unloading response mechanism can quickly cut off power when instantaneous slippage is detected, preventing thread tooth breakage damage without interrupting the continuous operation of the normal threading process. The entire process replaces the coarse calculation method of approximating torque with hydraulic pressure, narrowing the deviation range between the indicated torque and the actual thread pair torque, reducing the hidden quality problem of false tightness and real leakage under low-temperature conditions. It eliminates the need for adding a large number of external detection components, adapting to the modification and upgrading of existing in-service power clamps, controlling modification costs, and directly connecting to the original hydraulic system to complete the functional implementation. Attached Figure Description

[0008] Figure 1 This is a diagram illustrating the steps of the method described in this invention; Figure 2 This is a detailed flowchart of step S5 in this invention. Detailed Implementation

[0009] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0010] One embodiment of the present invention, such as Figure 1 As shown, a power caliper control method with adaptive torque control function is disclosed, the method comprising: S1. Acquire the original acoustic wave signal output by the high-frequency acoustic emission array embedded in the gearbox during the operation of the power clamp, and simultaneously acquire the original vibration signal output by the microelectromechanical angular accelerometer built into the jaw plate seat. Perform bandpass filtering and power frequency interference removal on the two types of original signals respectively to generate a noise-reduced acoustic and vibration basic dataset. Perform time-domain feature extraction and frequency-domain decomposition on the acoustic and vibration basic dataset to separate the thread meshing micro-protrusion scraping feature component and the transmission chain torsional vibration feature component to generate an acoustic and vibration separation feature dataset. S2. Perform feature mapping processing on the acoustic-vibration separation feature dataset, match the thread meshing micro-protrusion scraping feature component to the preset acoustic fingerprint library, match the transmission chain torsional vibration feature component to the preset torsional vibration mode library, and generate an initial feature matching result set; perform time-series correlation verification processing on the initial feature matching result set, remove invalid matching items that are falsely triggered by single-frame signals, and generate a stable acoustic-vibration fusion feature result set. S3. Perform state determination processing on the acoustic vibration fusion feature result set, and map the current power clamp operating state to the preset four-stage state set of lead-in, alignment, thread engagement shoulder, and interference hold, and generate the current operating state identification data; perform boundary condition verification processing on the current operating state identification data, and combine the real-time collected clamp body speed and hydraulic system flow data to lock the operating range boundary of the current state and generate state range locking data. S4. Perform anomaly feature comparison processing on the state interval locking data, compare the real-time acoustic and vibration features with the preset misalignment, slippage and tooth slippage anomaly feature templates one by one, and generate anomaly feature trigger marker data; perform priority sorting processing on the anomaly feature trigger marker data, divide the response order according to the anomaly hazard level, and generate anomaly response execution sequence data. S5. Perform control command generation processing on the abnormal response execution sequence data, output matching hydraulic valve group adjustment commands corresponding to different abnormal types, and generate real-time dynamic adjustment command data; perform timing calibration processing on the real-time dynamic adjustment command data, match the response window of the current operating state, and generate the final directly executable power clamp control command data. S6. Perform working condition adaptation processing on the final directly executable power clamp control command data, and modify the preset pressure output curve parameters by combining the real-time collected ambient temperature data to generate refined pressure control data with temperature compensation; perform output rate limiting processing on the refined pressure control data with temperature compensation to constrain the pressure rise gradient and generate final execution control data without rigid impact.

[0011] The working principle and effects of the above technical solution are as follows: By synchronously acquiring and layering feature decomposition of acoustic and vibration signals throughout the entire operation of the power clamp, it can accurately capture subtle changes in the thread engagement and transmission chain operation, improving the accuracy of power clamp operation status identification. Through multi-dimensional feature matching and timing verification, it effectively filters out judgment deviations caused by false signal triggering, reducing misjudgments of equipment operation status. Through graded anomaly identification and orderly response control modes, it improves the efficiency of handling abnormal operating conditions and avoids equipment damage caused by persistent faults such as misalignment, incorrect threading, and stripping. By combining dynamic torque adjustment based on operating conditions and ambient temperature, it optimizes the hydraulic pressure output curve, reduces the rigid impact of the hydraulic system, and improves the stability and precision of power clamp torque control. This solution not only adapts to the operational needs of different stages of power clamp operation but also significantly reduces equipment failure losses, effectively extends equipment lifespan, and ensures continuous and stable on-site operations.

[0012] In one embodiment of the present invention, S1 includes: S11. Deploy a multi-channel synchronous acquisition link, uniformly set the sampling frame rate and sampling bit width of the acoustic emission array and the microelectromechanical angular accelerometer, synchronously start the data acquisition process of the two types of sensing elements, continuously record the original acoustic waveform sequence and original vibration acceleration time sequence of the entire process output from the start of the power clamp to torque saturation, and summarize them to form the original acoustic vibration mixed acquisition package without interference suppression. S12. Divide the sound wave signal and vibration signal into their own dedicated filter frequency bands, configure the corresponding finite-length unit impulse response filter operators, traverse all waveform data in the original sound and vibration mixed acquisition package point by point, filter out the power grid coupled power frequency harmonics and low-frequency environmental noise of the equipment casing, complete the baseline drift correction of the two types of sensor data, and integrate them to obtain a regular waveform data group with noise interference removed. S13. Divide the regular waveform data group into a fixed-length sliding analysis window. Calculate four types of time-domain quantization indicators, namely peak factor, kurtosis, impulse factor, and root mean square, segment by segment within the window. Simultaneously, perform a fast Fourier transform on the waveform within the window to decompose the full-domain spectrum, split the energy distribution curves corresponding to different frequency intervals, complete the extraction of full-dimensional feature parameters, and produce a complete time-domain and frequency-domain joint feature matrix. S14. Distinguish between the mid-to-high frequency narrow band energy segment corresponding to the friction vibration of the thread meshing contact surface and the low-frequency continuous spectrum segment corresponding to the torsional oscillation of the reduction gear transmission shaft system. Set an energy threshold to complete the interval segmentation of the two types of feature components, remove the coupled noise components that are superimposed between the two types of components, and archive the separated independent feature components in a unified manner to generate an acoustic-vibration separation feature dataset.

[0013] The working principle and effects of the above technical solution are as follows: By using a multi-channel synchronous acquisition link to uniformly acquire the acoustic and vibration signals of the entire power clamp process, it can comprehensively cover various operating conditions of the equipment, avoiding the problem of incomplete status analysis due to missing local data. Through targeted frequency band filtering and baseline correction processing, it can effectively eliminate environmental noise and power frequency harmonic interference, reduce the proportion of invalid noise in the original sensor data, and improve the purity of the basic data. By using a sliding window to jointly extract multiple time-domain indicators and full-domain frequency-domain features, it can fully explore the hidden operating details within the acoustic and vibration signals, enhancing the comprehensiveness of equipment condition feature extraction. By using frequency band energy threshold segmentation to remove coupled noise and separate two types of core feature components, it can accurately distinguish between thread friction characteristics and transmission torsional vibration characteristics, ensuring the basic accuracy of subsequent equipment status identification and fault judgment, providing reliable data support for subsequent refined control.

[0014] In one embodiment of the present invention, S2 includes: S21. Extract the spectral peak distribution, pulse duration, energy decay rate parameters of each group of thread scraping components in the acoustic-vibration separation feature dataset, construct a multi-dimensional feature description vector, traverse all standard thread meshing sample vectors in the acoustic fingerprint database, calculate the cosine similarity values ​​between vectors, and record the index codes corresponding to all samples that meet the similarity threshold. S22. Extract the inherent resonant frequency, harmonic amplitude, attenuation period parameters of each group of transmission torsional vibration components in the acoustic-vibration separation feature dataset, construct the torsional vibration mode description vector, traverse all standard mode sample vectors of transmission systems in the torsional vibration mode library, calculate the Euclidean distance between vectors, and record the index codes corresponding to all samples that meet the distance constraint threshold. S23. Integrate the thread scraping matching index code and the transmission torsional vibration matching index code, complete the same frame binding of the two sets of matching data according to the acquisition timestamp order, store all matching sample numbers and corresponding similarity quantification values ​​corresponding to a single frame of sensor data, and summarize to obtain an initial feature matching result set covering the entire running sequence. S24. Extract five consecutive frames of matching data along the time axis to form a time-series verification unit. Compare the overlap ratio of matching sample numbers of adjacent frames within the unit. Eliminate isolated matching records that only appear in a single frame and have no corresponding matching samples in consecutive frames before and after. Retain valid matching entries that continuously overlap in multiple frames. Reorganize the remaining valid entries in chronological order to generate a stable acoustic-vibration fusion feature result set.

[0015] The working principle and effects of the above technical solution are as follows: By constructing multi-dimensional feature vectors for thread scraping and transmission torsional vibration respectively and matching them with corresponding standard sample libraries, the state discrimination dimensions of different components can be refined, improving the accuracy of power clamp operation feature matching. By binding the two types of matching data in a time-series dimension, the full-time-series operation feature information of the equipment can be completely retained, reducing state misjudgments caused by the omission of local features. Through the filtering method of cross-validation of multi-frame continuous data, invalid matching records caused by instantaneous interference during the acquisition process can be filtered out, significantly reducing the judgment error caused by single-point signal fluctuations. This processing method can take into account the dual operating states of the thread meshing part and the transmission chain structure, making the equipment operating condition identification more comprehensive and detailed, and ensuring the stability of the final feature results, providing reliable pre-data support for subsequent equipment state judgment and anomaly identification.

[0016] In one embodiment of the present invention, step S22 includes: Read the complete transmission chain torsional vibration characteristic component time series data in the acoustic-vibration separation feature dataset, extract the four core quantization parameters corresponding to each frame signal, namely the inherent resonant frequency, fundamental harmonic amplitude, oscillation decay period, and frequency band energy ratio, and summarize them to form the original torsional vibration parameter set. The original set of torsional vibration parameters is normalized and unified to eliminate the differences in numerical magnitude between different parameter dimensions. Based on the normalized parameter values, a multi-dimensional standardized torsional vibration mode description vector is constructed to complete the digital representation of a single frame torsional vibration signal. Retrieve all standard modal sample vectors of the transmission system stored in the preset torsional vibration modal library, and perform a full-domain comparison operation between the real-time torsional vibration modal description vector and each standard sample vector in chronological order, and calculate the Euclidean distance quantization value between the two sets of vectors in batches. Compare all Euclidean distance calculation results with the system's preset distance constraint threshold, filter out all matching samples whose values ​​are lower than the constraint threshold, and mark the modality category and sequence position information corresponding to this type of sample; All qualified standard modal samples are uniformly recorded with their own index codes. The codes are then archived in an orderly manner according to the signal acquisition sequence to generate a transmission torsional vibration component modal matching index dataset.

[0017] The working principle and effects of the above technical solution are as follows: By comprehensively extracting torsional vibration time-series data of the transmission chain through multiple core parameters, it can completely capture subtle vibration changes in the transmission system and improve the completeness of torsional vibration data acquisition. Normalization processing unifies the numerical magnitudes of various parameters, eliminating comparison biases caused by differences in parameter dimensions, making the digital representation of torsional vibration signals more regular and uniform. Sample matching is performed through global vector comparison and Euclidean distance calculation, refining the identification accuracy of transmission modes and reducing matching omissions caused by traditional single-parameter comparisons. Threshold filtering and time-series archiving processing remove sample data with substandard matching accuracy, retaining accurate and effective modal matching information. This entire processing method can accurately identify various operating modes of the transmission chain and stably output time-coherent matching data, avoiding misjudgments of operating conditions caused by ignoring weak torsional vibration characteristics, and providing solid data support for equipment transmission status analysis.

[0018] In one embodiment of the present invention, S3 includes: S31. Extract the three types of change patterns of energy change trend, frequency drift amplitude and pulse occurrence frequency in the continuous time sequence features within the acoustic-vibration fusion feature result set. Compare the unique feature change pattern templates of the four operation stages, calculate the matching degree of the current time sequence feature with the templates of each stage, and select the stage with the highest matching degree as the current operation stage. S32. Assign a unique digital code to the selected operating stage, and synchronously record all acoustic and vibration characteristic quantization mean, extreme value, fluctuation range and auxiliary parameters within the time period corresponding to the code. Encapsulate the stage code and auxiliary parameters into a single time-series record, continuously store all collection time period records, and generate current operating status identification data. S33. Simultaneously read the instantaneous speed value output by the rotary encoder of the power clamp spindle and the real-time flow value output by the hydraulic main oil circuit flow sensor. Align the two sets of working condition data with the same timestamp and embed them into the current operating status identifier data to form a composite time-series data packet that binds the acoustic and vibration characteristics with the hydraulic mechanical working condition. S34. Statistically analyze the fluctuation range of the rotational speed value and the fluctuation range of the hydraulic flow value corresponding to the same operating stage code, eliminate extreme outliers caused by instantaneous impacts, retain the value range that is stable over a long period of time as the exclusive operating boundary of this stage, bind and archive the stage code with the corresponding rotational speed and flow boundary values ​​one by one, and generate state range locked data.

[0019] The working principle and effects of the above technical solution are as follows: By dividing the power clamp operation stages through multi-dimensional time-series feature fitting and comparison, the granularity of state differentiation throughout the entire equipment operation process can be refined, improving the accuracy of equipment operation stage judgment and reducing identification errors caused by coarse state division. By encoding and binding various acoustic and vibration characteristic parameters to form standardized state data, the equipment operation information can be organized, making the state records of each operation stage more complete and standardized. By synchronously binding real-time operating data of speed and flow, a deep integration of vibration characteristics and hydraulic operating parameters can be achieved, enhancing the comprehensiveness of equipment state perception. By eliminating instantaneous abnormal values ​​and locking stable operating ranges, the problem of misjudgment of ranges caused by instantaneous fluctuations in operating conditions can be avoided. The entire processing method can accurately distinguish each operation process of the power clamp, adapt to the control requirements of different stages, and provide accurate operating condition boundary basis for subsequent anomaly identification and torque regulation, ensuring the stability of equipment operation.

[0020] In one embodiment of the present invention, step S4 includes: S41. The thread scraping component and transmission torsional vibration component contained in each time unit of the state interval locking data are split, and the three types of anomaly discrimination parameters of peak offset, energy anomaly gain and pulse density of each component are extracted to construct a single-frame real-time anomaly discrimination parameter vector. S42. Sequentially retrieve the standard abnormal feature templates corresponding to the three types of faults: misalignment, incorrect engagement, and slippage. Calculate the quantized deviation between the real-time abnormality discrimination parameter vector and the template vector for each type. Set a fixed deviation threshold to determine whether a single type of abnormality meets the triggering standard. Assign an independent triggering mark code to all abnormal types that meet the triggering standard. S43. Synchronously bind and store the trigger mark code with the speed and flow boundary values ​​of the corresponding time unit, distinguish between two types of trigger scenarios: single parameter over-limit and multi-parameter synchronous over-limit, fully record the duration deviation of the time window for each anomaly, summarize all time unit records, and generate anomaly feature trigger mark data. S44. Define the four-level hazard classification standard: slipped teeth correspond to the highest level, mis-deduction corresponds to the second highest level, off-deduction corresponds to the basic level, and slight parameter fluctuations correspond to the lowest level. Traverse all trigger mark codes in the abnormal feature trigger mark data and match the corresponding hazard level sequence number for each mark. S45. Reorder all anomaly markers from highest to lowest according to the hazard level number. Within the same level, sort them again from longest to shortest according to the duration of the anomaly. Mark the corresponding anomaly type and duration window after each sorted record. Store all anomaly entries completely and orderly, and generate anomaly response execution sequence data.

[0021] The working principle and effects of the above technical solution are as follows: By constructing standardized vectors using multi-dimensional anomaly discrimination parameters for fault comparison, it can comprehensively capture subtle abnormal operating conditions of the power pliers, improving the precision of identifying various thread faults. Abnormal operating conditions are filtered through template matching and threshold judgment, accurately distinguishing different fault types such as misaligned threads, incorrect threads, and stripped threads, reducing misjudgments caused by fluctuations in normal operating conditions. By binding operating parameters and refining anomaly triggering scenarios, the timing information of anomaly occurrences is fully preserved, enhancing the precision and accuracy of equipment fault tracing. By classifying hazard levels into four levels and using dual sorting rules to streamline the anomaly response sequence, high-risk faults can be prioritized, avoiding equipment damage and operational failures caused by delayed handling of serious faults. This entire approach ensures timely investigation and rectification of minor anomalies while focusing on rapid response to major faults, effectively improving the safety and operational stability of power pliers operations.

[0022] In one embodiment of the present invention, S42 includes: Retrieve the standard abnormal feature templates for misaligned thread failure, incorrect thread failure, and slipped thread failure built into the system, fully load the standard parameter vector data corresponding to each type of abnormal template, and generate an integrated dataset of multiple types of abnormal templates. The single-frame real-time anomaly discrimination parameter vector is matched with various standard template vectors in the multi-type anomaly template integrated dataset in turn, and the vector deviation values ​​corresponding to each type of fault are output in batches to generate a multi-dimensional vector deviation statistical data set. Read the system's preset fixed deviation threshold, compare the various deviation values ​​within the multi-dimensional vector deviation statistics dataset with the fixed deviation threshold in batches, filter out the abnormal types whose deviation values ​​meet the trigger conditions, and generate valid abnormal type filtering data. The various anomaly types covered by the effective anomaly type screening data are coded, and a unique numerical tag code is assigned to each anomaly that meets the triggering conditions. The coding and archiving of all effective anomalies are completed, and the original anomaly trigger data with unique codes is generated.

[0023] The working principle and effects of the above technical solution are as follows: By integrating multiple types of power clamp fault standard feature templates to perform vector matching calculations, targeted identification of different thread faults can be achieved, improving the accuracy of fault type differentiation. By batch calculating the deviation values ​​between real-time parameter vectors and standard templates, the degree of equipment operating condition abnormality can be quantified, reducing the one-sidedness caused by relying on a single signal to judge faults. By using a unified threshold to complete the screening of abnormal states, the judgment criteria for various faults are unified, avoiding missed and false judgments caused by inconsistent judgment scales. By completing dedicated coding and archiving of valid abnormalities, the storage format of various abnormal data is standardized, facilitating subsequent fault hierarchy sorting and response sequence arrangement. The entire processing method can accurately capture various subtle operating condition abnormalities, improving the overall accuracy of fault identification, and also organize abnormal data information, providing a solid and standardized data foundation for subsequent graded response and precise control.

[0024] One embodiment of the present invention, such as Figure 2 As shown, S5 includes: S51. Read the first-order exception marker code in the exception response execution sequence data, retrieve the exception type-specific adjustment logic that matches the code, and split the three sets of adjustment parameters corresponding to the opening adjustment amplitude, adjustment direction, and adjustment duration of the three types of valves: main pressure regulating valve, proportional flow valve, and directional valve. S52. Convert the three types of valve adjustment parameters into digital level control codes that the controller can recognize, and synchronously write the torque reduction amplitude, speed limit threshold, flow throttling ratio and auxiliary constraint parameters corresponding to the abnormality. Package and store all control codes and auxiliary constraint parameters according to the timestamp of the abnormality occurrence, and generate a single adjustment instruction unit corresponding to the abnormality. S53. Traverse all exception flag codes in the exception response execution sequence data, generate independent adjustment instruction units one by one, stack all instruction units in sorting order, record the start execution time and duration of each instruction unit, and integrate all units to form real-time dynamic adjustment instruction data of multi-instruction stack. S54. Retrieve the running stage boundary interval corresponding to the current timing unit within the locked data of the status interval, extract the start and end nodes of the time window that allows control instructions to intervene and adjust in this stage, remove instruction units that exceed the start and end nodes of the window, and correct the start execution time point of the instruction units remaining inside the window to adapt to the timing range of the window. S55. Unify and standardize the output timing interval of various valve control codes within the instruction unit, eliminate the control signal superposition conflict caused by the synchronous output of multiple instructions, verify that the values ​​of all adjustment parameters do not exceed the adjustment limits allowed by the hydraulic system hardware, and encapsulate all instruction units after standardization to generate the final directly executable power clamp control instruction data.

[0025] The working principle and effects of the above technical solution are as follows: By classifying and decomposing the adjustment parameters of various valve components and generating independent instruction units, the control logic adapted to the abnormal output of different equipment can be significantly improved, enhancing the control targeting of the power clamp under abnormal operating conditions. By binding multi-dimensional constraint parameters and regularizing instruction data according to timestamps, the output standards of various control parameters can be unified, reducing the operating condition adaptation deviation caused by a single adjustment instruction. By matching the effective adjustment window of the equipment operation phase and filtering and correcting the instructions, the operational disturbance caused by ineffective period control can be avoided, ensuring that the control action conforms to the actual operating rhythm of the equipment. By regularizing the instruction output timing and verifying the hardware adaptation range, signal conflicts caused by the superposition of multiple instructions are eliminated, preventing hardware wear problems caused by over-parameter adjustment. The entire processing method can achieve layered and precise control of various faults, and ensure the smooth and orderly adjustment process of the hydraulic system, effectively improving the overall control accuracy and operational stability of the power clamp.

[0026] In one embodiment of the present invention, S51 includes: Read the complete time sequence entries inside the abnormal response execution sequence data, extract the first abnormal marker code and the corresponding abnormal level time sequence attribute information of the sequence, and generate target abnormal location data; identify the abnormal type of the target abnormal location data, match it with the valve group adjustment rule system corresponding to various faults built into the system, and generate a dedicated adjustment logic dataset. The dedicated control logic dataset is decomposed and parsed to separate the core parameters such as the opening change range, valve core movement direction, and continuous working duration that are adapted to the operating conditions of the main pressure regulating valve, and generate the main pressure regulating valve control parameter group. The dedicated control logic dataset is split into dimensions to extract the flow control range, dynamic compensation duration, and steady-state holding parameters of the proportional flow valve, generating a proportional flow valve control parameter set. The dedicated control logic dataset is finely decomposed into parameters, separating the station switching direction, switching action duration, and state holding period parameters of the directional valve to generate the directional valve control parameter group; the complete parameter groups corresponding to the three types of valves are integrated to complete the summary of multi-valve group coordinated control parameters and generate integrated data of global valve group control parameters.

[0027] The working principle and effects of the above technical solution are as follows: By accurately locating the highest priority equipment anomalies and identifying specific fault types, the valve group adjustment logic can be aligned with actual working conditions, improving the matching degree of fault control. By layering and decomposing various operating parameters of the main pressure regulating valve, proportional flow valve, and directional valve, the control dimensions of different valve components are refined, enhancing the precision of multi-valve group coordinated adjustment. Extracting the action amplitude, running time, and state holding parameters of various valve bodies separately covers all core points of hydraulic system adjustment, reducing the problem of inadequate adjustment caused by missing parameters. By integrating multiple valve group parameters to form complete adjustment data, the output standard of multi-device coordinated control can be unified. This entire processing method can output differentiated adjustment schemes for different abnormal working conditions, eliminating the control deviation caused by a single adjustment mode adapting to all faults, and ensuring more coordinated linkage of hydraulic valve groups, improving the stability and accuracy of power clamp adjustment under abnormal working conditions.

[0028] In one embodiment of the present invention, step S6 includes: S61. Deploy distributed temperature sensing nodes around the clamp body to synchronously collect three sets of temperature values: hydraulic oil tank temperature, equipment shell temperature, and ambient air temperature. Calculate the weighted average of the three sets of values ​​to obtain the equivalent ambient temperature of the power clamp's overall operation. Embed the final directly executable power clamp control command data into the data with the same timestamp. S62. Retrieve the hydraulic oil viscosity correction coefficient and seal elastic decay correction coefficient corresponding to different temperature ranges, match the corresponding correction coefficient according to the equivalent ambient temperature, and synchronously scale and correct the inflection point peak slope parameter of the preset basic pressure output curve in the control command to eliminate the hydraulic output deviation caused by temperature changes. S63. Merge and store all node values ​​of the pressure curve after scaling correction with the original valve adjustment control code torque constraint parameters, label the equivalent ambient temperature value and correction coefficient value corresponding to each record, retain the complete parameter record of the temperature correction process, and generate refined pressure control data with temperature compensation. S64. Extract all instantaneous pressure change nodes within the refined pressure control data with temperature compensation, calculate the pressure change rate between adjacent nodes, compare it with the upper limit of the allowable smooth change rate of the hydraulic cylinder, and identify pressure abrupt change nodes that exceed the rate limit. S65. The identified pressure mutation nodes are split into multi-level intermediate pressure nodes, the time span of pressure change is lengthened, the pressure difference between any adjacent time nodes is limited to the threshold corresponding to the steady change rate, the smooth transition rewriting of all mutation nodes is completed, the rewritten pressure curve is verified to have no instantaneous steep increase or decrease segments, the rewritten complete set of control parameters is integrated, and the final execution control data without rigid impact is generated.

[0029] The working principle and effects of the above technical solution are as follows: By collecting and weighting temperature data at multiple points to calculate the equivalent operating temperature, the actual working environment of the power clamp can be accurately reflected, reducing data deviations caused by a single temperature measurement method. By adjusting the pressure curve parameters according to the temperature-corresponding correction coefficient, the hydraulic output deviation caused by changes in oil temperature and seal characteristics can be offset, significantly improving the pressure control accuracy under different temperature conditions. By completely retaining the temperature correction parameter records, the pressure regulation process is traceable, ensuring the consistency of each working condition adjustment. By identifying pressure change points and adding multiple transition intervals, the rate of pressure change can be slowed down, eliminating instantaneous pressure shocks in the hydraulic system. The entire processing method can adapt to complex and variable on-site temperature environments, weakening the impact of temperature fluctuations on operational accuracy, and buffering the mechanical vibration generated by pressure regulation, reducing equipment wear, and effectively improving the long-term stability and service life of the power clamp.

[0030] In one embodiment of the present invention, S62 includes: Read the equivalent ambient temperature of the power clamp's overall operation from the embedded control command data, traverse the system's preset global temperature range gradient, match the current temperature to the standard temperature range segment, and generate temperature range matching data. The temperature range matching data is processed by coefficient retrieval, and the hydraulic oil viscosity correction coefficient and seal elastic decay correction coefficient adapted to the corresponding temperature range are extracted to generate a two-dimensional temperature correction coefficient dataset. The parameter adaptation process is performed on the two-dimensional temperature correction coefficient dataset. The correction coefficient is applied to the basic pressure output curve contained in the control command. The three core parameters of curve inflection point position, peak value, and slope are scaled and adjusted as a whole to generate the original pressure curve data after parameter correction. The original pressure curve data after parameter correction is processed to offset the deviations in hydraulic medium flow characteristics and sealing structure deformation caused by ambient temperature fluctuations, thereby optimizing the pressure output accuracy and generating standardized pressure curve data after temperature adaptation.

[0031] The working principle and effects of the above technical solution are as follows: By matching the standard range corresponding to the equivalent temperature to complete the coefficient retrieval, both temperature interference factors of hydraulic medium and sealing components can be taken into account simultaneously, enhancing the coverage of temperature interference correction. By simultaneously adjusting three core parameters—inflection point, peak value, and slope—of the pressure curve, the overall output offset caused by temperature is completely corrected, improving the fit of pressure output under different environments. Through a two-step processing mode of layered scaling of curve parameters and superimposed deviation correction, the situation where a single parameter correction cannot offset the composite deformation deviation is reduced, avoiding the problem of sudden fluctuations in torque output when switching between high and low temperatures. The entire process can weaken the control errors caused by low-temperature oil viscosity and high-temperature seal loosening, while maintaining the continuous stability of the entire pressure change curve, allowing the power clamp to maintain a stable and accurate torque output effect in working environments with significant temperature differences throughout the year.

[0032] According to one embodiment of the present invention, a power caliper control system with adaptive torque control function includes: One or more processors; Memory, used to store one or more programs; Wherein, when the one or more programs are executed by the one or more processors, the one or more processors are made to implement the method described in any one of the above.

[0033] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A power caliper control method with adaptive torque control function, characterized in that, The method includes: S1. Acquire the original acoustic wave signal and simultaneously acquire the original vibration signal. Perform bandpass filtering and power frequency interference removal on the two types of original signals respectively to generate a noise-reduced acoustic and vibration basic dataset. Perform time-domain feature extraction and frequency-domain decomposition on the acoustic and vibration basic dataset to separate the thread meshing micro-protrusion scraping feature component and the transmission chain torsional vibration feature component to generate an acoustic and vibration separation feature dataset. S2. Perform feature mapping processing on the acoustic-vibration separation feature dataset, match the thread meshing micro-protrusion scraping feature component to the preset acoustic fingerprint library, match the transmission chain torsional vibration feature component to the preset torsional vibration mode library, and generate an initial feature matching result set; perform time-series correlation verification processing on the initial feature matching result set to generate a stable acoustic-vibration fusion feature result set; S3. Perform state determination processing on the acoustic vibration fusion feature result set to generate current operating state identification data; perform boundary condition verification processing on the current operating state identification data, and combine the real-time collected clamp rotation speed and hydraulic system flow data to lock the operating range boundary of the current state and generate state range locking data. S4. Perform anomaly feature comparison processing on the state interval locked data, compare the real-time acoustic and vibration features with the preset misalignment, slippage and tooth slippage anomaly feature templates one by one, and generate anomaly feature trigger marker data; perform priority sorting processing on the anomaly feature trigger marker data, divide the response order according to the anomaly hazard level, and generate anomaly response execution sequence data. S5. Perform control command generation processing on the abnormal response execution sequence data, output matching hydraulic valve group adjustment commands corresponding to different abnormal types, and generate real-time dynamic adjustment command data; perform timing calibration processing on the real-time dynamic adjustment command data, match the response window of the current operating state, and generate the final directly executable power clamp control command data. S6. Perform working condition adaptation processing on the final directly executable power clamp control command data to generate refined pressure control data with temperature compensation; perform output rate limiting processing on the refined pressure control data with temperature compensation to constrain the pressure rise gradient and generate final execution control data without rigid impact.

2. The power caliper control method with adaptive torque control function according to claim 1, characterized in that, S1 includes: S11. Deploy a multi-channel synchronous acquisition link, uniformly set the sampling frame rate and sampling bit width of the acoustic emission array and the microelectromechanical angular accelerometer, synchronously start the data acquisition process of the two types of sensing elements, continuously record the original acoustic waveform sequence and original vibration acceleration time sequence of the entire process output from no-load start to torque saturation of the power clamp, and summarize them to form the original acoustic vibration mixed acquisition package without interference suppression. S12. Divide the acoustic signal and vibration signal into their own dedicated filter frequency bands, configure the corresponding finite-length unit impulse response filter operators, traverse all waveform data in the original acoustic-vibration mixed acquisition package point by point, filter out the power grid coupled power frequency harmonics and the low-frequency environmental noise of the equipment casing, complete the baseline drift correction of the two types of sensor data, and integrate them to obtain a regular waveform data group with noise interference removed. S13. Divide the regular waveform data group into a fixed-length sliding analysis window. Calculate four types of time-domain quantization indicators, namely peak factor, kurtosis, impulse factor, and root mean square, segment by segment within the window. Simultaneously, perform a fast Fourier transform on the waveform within the window to decompose the full-domain spectrum, split the energy distribution curves corresponding to different frequency intervals, complete the extraction of full-dimensional feature parameters, and produce a complete time-domain and frequency-domain joint feature matrix. S14. Distinguish between the mid-to-high frequency narrow band energy segment corresponding to the friction vibration of the thread meshing contact surface and the low-frequency continuous spectrum segment corresponding to the torsional oscillation of the reduction gear transmission shaft system. Set an energy threshold to complete the interval segmentation of the two types of feature components, remove the coupled noise components that are superimposed between the two types of components, and archive the separated independent feature components in a unified manner to generate an acoustic-vibration separation feature dataset.

3. The power caliper control method with adaptive torque control function according to claim 1, characterized in that, The S2 includes: S21. Extract the spectral peak distribution, pulse duration, energy decay rate parameters of each group of thread scraping components in the acoustic-vibration separation feature dataset, construct a multi-dimensional feature description vector, traverse all standard thread meshing sample vectors in the acoustic fingerprint database, calculate the cosine similarity values ​​between vectors, and record the index codes corresponding to all samples that meet the similarity threshold. S22. Extract the inherent resonant frequency, harmonic amplitude, attenuation period parameters of each group of transmission torsional vibration components in the acoustic-vibration separation feature dataset, construct the torsional vibration mode description vector, traverse all standard mode sample vectors of transmission systems in the torsional vibration mode library, calculate the Euclidean distance between vectors, and record the index codes corresponding to all samples that meet the distance constraint threshold. S23. Integrate the thread scraping matching index code and the transmission torsional vibration matching index code, complete the same frame binding of the two sets of matching data according to the acquisition timestamp order, store all matching sample numbers and corresponding similarity quantification values ​​corresponding to a single frame of sensor data, and summarize to obtain an initial feature matching result set covering the entire running sequence. S24. Extract five consecutive frames of matching data along the time axis to form a time-series verification unit. Compare the overlap ratio of matching sample numbers of adjacent frames within the unit. Eliminate isolated matching records that only appear in a single frame and have no corresponding matching samples in consecutive frames before and after. Retain valid matching entries that continuously overlap in multiple frames. Reorganize the remaining valid entries in chronological order to generate a stable acoustic-vibration fusion feature result set.

4. The power caliper control method with adaptive torque control function according to claim 3, characterized in that, S22 includes: Read the complete transmission chain torsional vibration characteristic component time series data in the acoustic-vibration separation feature dataset, extract the four core quantization parameters corresponding to each frame signal, namely the inherent resonant frequency, fundamental harmonic amplitude, oscillation decay period, and frequency band energy ratio, and summarize them to form the original torsional vibration parameter set. The original set of torsional vibration parameters is normalized and unified. Based on the normalized parameter values, a multi-dimensional standardized torsional vibration mode description vector is constructed to complete the digital representation of a single frame torsional vibration signal. Retrieve all standard modal sample vectors of the transmission system stored in the preset torsional vibration modal library, and perform a full-domain comparison operation between the real-time torsional vibration modal description vector and each standard sample vector in chronological order, and calculate the Euclidean distance quantization value between the two sets of vectors in batches. Compare all Euclidean distance calculation results with the system's preset distance constraint threshold, filter out all matching samples whose values ​​are lower than the constraint threshold, and label the modality category and sequence position information corresponding to this type of sample; All qualified standard modal samples are uniformly recorded with their own index codes. The codes are then archived in an orderly manner according to the signal acquisition sequence to generate a transmission torsional vibration component modal matching index dataset.

5. The power caliper control method with adaptive torque control function according to claim 1, characterized in that, The S3 includes: S31. Extract the three types of change patterns of energy change trend, frequency drift amplitude and pulse occurrence frequency in the continuous time sequence features within the acoustic-vibration fusion feature result set. Compare the unique feature change pattern templates of the four operation stages, calculate the matching degree of the current time sequence feature with the templates of each stage, and select the stage with the highest matching degree as the current operation stage. S32. Assign a unique digital code to the selected operating stage, and synchronously record the quantized mean, extreme value, fluctuation range and auxiliary parameters of all acoustic and vibration characteristics within the time period corresponding to the code. Encapsulate the stage code and auxiliary parameters into a single time-series record, continuously store all collection time period records, and generate the current operating status identification data. S33. Synchronously read the instantaneous speed value output by the rotary encoder of the power clamp spindle and the real-time flow value output by the hydraulic main oil circuit flow sensor. Align the two sets of working condition data with the same timestamp and embed them into the current operating status identifier data to form a composite time-series data packet that binds the acoustic and vibration characteristics with the hydraulic mechanical working condition. S34. Statistically analyze the fluctuation range of the rotational speed value and the fluctuation range of the hydraulic flow value corresponding to the same operating stage code, eliminate extreme outliers caused by instantaneous impacts, retain the value range that is stable over a long period of time as the exclusive operating boundary of this stage, bind and archive the stage code with the corresponding rotational speed and flow boundary values ​​one by one, and generate state range locking data.

6. The power caliper control method with adaptive torque control function according to claim 1, characterized in that, The S4 includes: S41. The thread scraping component and transmission torsional vibration component contained in each time unit of the state interval locking data are split, and the three types of anomaly discrimination parameters of peak offset, energy anomaly gain and pulse density of each component are extracted to construct a single-frame real-time anomaly discrimination parameter vector. S42. Sequentially retrieve the standard abnormal feature templates corresponding to the three types of faults: misalignment, incorrect engagement, and slippage. Calculate the quantized deviation between the real-time abnormality discrimination parameter vector and the template vector for each type. Set a fixed deviation threshold to determine whether a single type of abnormality meets the triggering standard. Assign an independent triggering mark code to all abnormal types that meet the triggering standard. S43. Synchronously bind and store the trigger mark code with the speed and flow boundary values ​​of the corresponding time unit, distinguish between two types of trigger scenarios: single parameter over-limit and multi-parameter synchronous over-limit, fully record the duration deviation of the time window for each anomaly, summarize all time unit records, and generate anomaly feature trigger mark data. S44. Define the four-level hazard classification standard: slipped teeth correspond to the highest level, mis-deduction corresponds to the second highest level, off-deduction corresponds to the basic level, and slight parameter fluctuations correspond to the lowest level. Traverse all trigger mark codes in the abnormal feature trigger mark data and match the corresponding hazard level sequence number for each mark. S45. Reorder all anomaly markers from highest to lowest according to the hazard level number. Within the same level, sort them again from longest to shortest according to the duration of the anomaly. Mark the corresponding anomaly type and duration window after each sorted record. Store all anomaly entries completely and orderly, and generate anomaly response execution sequence data.

7. The power caliper control method with adaptive torque control function according to claim 1, characterized in that, The S5 includes: S51. Read the first-order exception marker code in the exception response execution sequence data, retrieve the exception type-specific adjustment logic that matches the code, and split the three sets of adjustment parameters corresponding to the opening adjustment amplitude, adjustment direction, and adjustment duration of the three types of valves: main pressure regulating valve, proportional flow valve, and directional valve. S52. Convert the three types of valve adjustment parameters into digital level control codes that the controller can recognize, and synchronously write the torque reduction amplitude, speed limit threshold, flow throttling ratio and auxiliary constraint parameters corresponding to the abnormality. Package and store all control codes and auxiliary constraint parameters according to the timestamp of the abnormality occurrence, and generate a single adjustment instruction unit corresponding to the abnormality. S53. Traverse all exception flag codes in the exception response execution sequence data, generate independent adjustment instruction units one by one, stack all instruction units in sorting order, record the start execution time and duration of each instruction unit, and integrate all units to form real-time dynamic adjustment instruction data of multi-instruction stack. S54. Retrieve the running stage boundary interval corresponding to the current timing unit within the locked data of the status interval, extract the start and end nodes of the time window that allows control instructions to intervene and adjust in this stage, remove instruction units that exceed the start and end nodes of the window, and correct the start execution time point of the instruction units remaining inside the window to adapt to the timing range of the window. S55. Unify and standardize the output timing interval of various valve control codes within the instruction unit, eliminate the control signal superposition conflict caused by the synchronous output of multiple instructions, verify that the values ​​of all adjustment parameters do not exceed the adjustment limits allowed by the hydraulic system hardware, and encapsulate all instruction units after standardization to generate the final directly executable power clamp control instruction data.

8. The power caliper control method with adaptive torque control function according to claim 7, characterized in that, S51 includes: Read the complete time sequence entries inside the abnormal response execution sequence data, extract the first abnormal marker code and the corresponding abnormal level time sequence attribute information of the sequence, and generate target abnormal location data; identify the abnormal type of the target abnormal location data, match it with the valve group adjustment rule system corresponding to various faults built into the system, and generate a dedicated adjustment logic dataset. The dedicated control logic dataset is decomposed and parsed to separate the core parameters such as the opening change range, valve core movement direction, and continuous working duration that are adapted to the operating conditions of the main pressure regulating valve, and generate the main pressure regulating valve control parameter group. The dedicated control logic dataset is split into dimensions to extract the flow control range, dynamic compensation duration, and steady-state holding parameters of the proportional flow valve, generating a proportional flow valve control parameter set. The dedicated control logic dataset is finely decomposed into parameters, separating the station switching direction, switching action duration, and state holding period parameters of the directional valve to generate the directional valve control parameter group; the complete parameter groups corresponding to the three types of valves are integrated to complete the summary of multi-valve group coordinated control parameters and generate integrated data of global valve group control parameters.

9. The power caliper control method with adaptive torque control function according to claim 1, characterized in that, The S6 includes: S61. Deploy distributed temperature sensing nodes around the clamp body to synchronously collect three sets of temperature values: hydraulic oil tank temperature, equipment shell temperature, and ambient air temperature. Calculate the weighted average of the three sets of values ​​to obtain the equivalent ambient temperature of the power clamp's overall operation. Embed the final directly executable power clamp control command data into the data with the same timestamp. S62. Retrieve the hydraulic oil viscosity correction coefficient and seal elastic decay correction coefficient corresponding to different temperature ranges, match the corresponding correction coefficient according to the equivalent ambient temperature, and synchronously scale and correct the inflection point peak slope parameter of the preset basic pressure output curve inside the control command. S63. Merge and store all node values ​​of the pressure curve after scaling correction with the original valve adjustment control code torque constraint parameters, label the equivalent ambient temperature value and correction coefficient value corresponding to each record, retain the complete parameter record of the temperature correction process, and generate refined pressure control data with temperature compensation. S64. Extract all instantaneous pressure change nodes within the refined pressure control data with temperature compensation, calculate the pressure change rate between adjacent nodes, compare it with the upper limit of the allowable smooth change rate of the hydraulic cylinder, and identify pressure abrupt change nodes that exceed the rate limit. S65. The identified pressure mutation nodes are split into multi-level transition intermediate pressure nodes, the time span of pressure change is lengthened, the pressure difference between any adjacent time nodes is limited to the threshold corresponding to the steady change rate, the complete set of control parameters is integrated and rewritten, and the final execution control data without rigid impact is generated.

10. A power caliper control system with adaptive torque control function, characterized in that, The system includes: One or more processors; Memory, used to store one or more programs; Wherein, when the one or more programs are executed by the one or more processors, the one or more processors implement the method of any one of claims 1 to 9.

Images