A super-small torque wrench calibration method and system based on micro-vibration energy spectrum inversion

Abstract

This application discloses a method and system for calibrating an ultra-small torque wrench based on micro-vibration energy spectrum inversion. The method includes: acquiring the micro-vibration signal, transient energy fluctuation information, and local contact stiffness change information of the contact interface between the wrench output end and the test piece during target torque loading; constructing the friction fingerprint energy spectrum under the current contact state; constructing a friction state sequence and establishing the state transition relationship between each friction state to generate a friction state evolution path; determining the stick-slip transition trend and friction state migration characteristics of the contact interface; constructing a nonlinear mapping relationship between friction state parameters and output torque to invert the actual output torque during the current target torque loading process; and generating the calibration result of the ultra-small torque wrench based on the actual output torque. This invention achieves dynamic and accurate calibration during ultra-small torque loading, improving calibration accuracy and robustness.

Description

Technical Field

[0001] This application relates to the field of precision metrology and instrument calibration technology, and in particular to a calibration method and system for ultra-small torque wrenches based on micro-vibration energy spectrum inversion. Background Technology

[0002] Ultra-low torque wrenches are widely used in precision electronic assembly, micro-machining, medical device assembly, MEMS device mounting, and micro-connection structure fastening. In these applications, torque output is typically in the micro-Newton-meter range or extremely low torque range. Traditional torque calibration methods are easily affected by factors such as structural micro-gap, nonlinear contact friction, micro-slippage, and changes in local contact stiffness, leading to a decrease in calibration stability and accuracy.

[0003] Most existing ultra-low torque calibration methods employ strain detection, force conversion, or static compensation, with their core still based on static mechanical parameters. However, during ultra-low torque loading, the interface between the wrench output and the workpiece under test exhibits significant unsteady frictional behavior, including adhesion, micro-slippage, periodic impact, and contact rebound. These behaviors cause changes in the distribution of local micro-vibration energy and result in complex spectral drift characteristics.

[0004] Traditional techniques typically treat vibration signals only as auxiliary noise processing targets, without analyzing the evolution of the friction state at the contact interface or establishing a correlation between friction state transition and actual output torque. Therefore, existing techniques struggle to accurately reflect the dynamic friction behavior during ultra-low torque loading, leading to drift errors, poor repeatability, and insufficient dynamic stability in calibration results.

[0005] Therefore, how to invert the evolution of friction state during ultra-small torque loading based on the micro-vibration energy change characteristics of the contact interface, and further realize the dynamic calibration of the real output torque, has become a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0006] This application provides a calibration method and system for ultra-small torque wrenches based on micro-vibration energy spectrum inversion, which realizes dynamic and accurate calibration during ultra-small torque loading process, and improves calibration accuracy and robustness.

[0007] This application provides the following solution:

[0008] According to the first aspect, a calibration method for an ultra-small torque wrench based on micro-vibration energy spectrum inversion is provided. The method includes: acquiring micro-vibration signals, transient energy fluctuation information, and local contact stiffness change information of the contact interface between the wrench output end and the test piece during target torque loading; constructing a friction fingerprint energy spectrum under the current contact state based on the micro-vibration signals, the transient energy fluctuation information, and the local contact stiffness change information; identifying the friction states corresponding to each moment during target torque loading based on the friction fingerprint energy spectrum, constructing a friction state sequence, establishing state transition relationships between each friction state, and generating a friction state evolution path; detecting local energy abrupt change intervals formed by the transition from a viscous state to a micro-slip state according to the friction state evolution path, determining the stick-slip transition trend and friction state migration characteristics of the contact interface; constructing a nonlinear mapping relationship between friction state parameters and output torque based on the stick-slip transition trend, the friction state migration characteristics, and the friction fingerprint energy spectrum, and inverting the actual output torque during the current target torque loading process; and generating the calibration result of the ultra-small torque wrench based on the actual output torque.

[0009] According to one achievable method in this application embodiment, the acquisition of micro-vibration signals, transient energy fluctuation information, and local contact stiffness change information of the contact interface between the wrench output end and the test piece during the target torque loading process includes: deploying micro-vibration sensing units at the wrench output end, the contact edge area, and the contact area of ​​the test piece; collecting transient vibration signals, local high-frequency impact signals, and contact stiffness fluctuation signals during the target torque loading process; and performing synchronous time-frequency processing on the transient vibration signals, local high-frequency impact signals, and contact stiffness fluctuation signals to generate micro-vibration signals, transient energy fluctuation information, and local contact stiffness change information.

[0010] According to one achievable method in this application embodiment, constructing the triboelectric fingerprint energy spectrum under the current contact state based on the micro-vibration signal, the transient energy fluctuation information, and the local contact stiffness change information includes: performing multi-band energy decomposition on the micro-vibration signal to generate local energy distribution results for multiple frequency bands; calculating the short-time energy mutation rate and spectral centroid drift of each frequency band based on the transient energy fluctuation information; calculating the high-frequency disturbance intensity and frequency band energy dispersion of each frequency band based on the local contact stiffness change information; and constructing the triboelectric fingerprint energy spectrum based on the local energy distribution results, the short-time energy mutation rate, the spectral centroid drift, the high-frequency disturbance intensity, and the frequency band energy dispersion.

[0011] According to one achievable method in this application embodiment, the step of identifying the friction state corresponding to each moment during the target torque loading process based on the friction fingerprint energy spectrum and constructing a friction state sequence includes: obtaining a preset friction state set, which at least includes a viscous state, a micro-slip state, a periodic impact state, and a steady-state friction state; extracting the energy proportion of a characteristic frequency band, the energy fluctuation gradient, and the discrete distribution characteristics of the spectrum based on the friction fingerprint energy spectrum; constructing a friction state constraint vector corresponding to each moment based on the energy proportion of the characteristic frequency band, the energy fluctuation gradient, and the discrete distribution characteristics of the spectrum; calculating the degree of state coupling between each friction state constraint vector and each preset friction state; determining the preset friction state with the highest degree of state coupling as the friction state at the corresponding moment, and generating a friction state sequence in chronological order.

[0012] According to one achievable method in an embodiment of this application, establishing the state transition relationship between each friction state and generating the friction state evolution path includes: obtaining the friction state constraint vectors corresponding to adjacent moments in the friction state sequence; constructing state transition edges between each friction state based on the energy distribution offset, spectral structure change, and disturbance growth trend between the friction state constraint vectors at adjacent moments; generating a friction state transition network based on the migration direction, migration intensity, and duration of each state transition edge; identifying high-frequency migration paths and cyclic migration paths in the friction state transition network; and generating the friction state evolution path based on the temporal correlation of the high-frequency migration paths, the cyclic migration paths, and each state transition edge.

[0013] According to one achievable method in this application embodiment, the step of detecting local energy abrupt change intervals formed by the transition from a viscous state to a microslip state based on the friction state evolution path, and determining the stick-slip transition trend and friction state migration characteristics of the contact interface, includes: detecting state abrupt change nodes formed by the transition from a viscous state to a microslip state in the friction state evolution path; extracting the local energy abrupt change intervals corresponding to the state abrupt change nodes; calculating the spectral centroid shift, high-frequency disturbance growth rate, and transient impact amplification in the local energy abrupt change intervals; determining the stick-slip transition trend based on the spectral centroid shift, the high-frequency disturbance growth rate, and the transient impact amplification; extracting the migration intensity, migration period, and state dwell time of each state transition edge based on the friction state transition network; and generating friction state migration characteristics based on the migration intensity, the migration period, and the state dwell time.

[0014] According to one achievable method in this application embodiment, the step of constructing a nonlinear mapping relationship between friction state parameters and output torque based on the stick-slip transition trend, the friction state transition characteristics, and the friction fingerprint energy spectrum, and inverting the true output torque during the current target torque loading process, includes: constructing a nonlinear mapping model between friction state parameters, spectral energy parameters, and output torque based on a combination of physical mechanisms and data-driven approaches; determining a friction abrupt change correction amount based on the stick-slip transition trend; determining a state transition compensation amount based on the friction state transition characteristics; and inverting the true output torque based on the friction fingerprint energy spectrum, the friction abrupt change correction amount, and the state transition compensation amount.

[0015] According to one achievable embodiment of this application, the method further includes: calculating the spectral energy entropy, the frequency band discrete entropy, and the perturbation chaos degree based on the triboelectric fingerprint energy spectrum; generating a triboelectric complexity parameter based on the spectral energy entropy, the frequency band discrete entropy, and the perturbation chaos degree; identifying the unsteady friction interval of the contact interface based on the triboelectric complexity parameter; and correcting the inversion result of the actual output torque based on the unsteady friction interval.

[0016] According to one achievable method in an embodiment of this application, generating the calibration result of the ultra-small torque wrench based on the actual output torque includes: calculating the dynamic torque deviation between the actual output torque and the target loading torque; extracting torque drift trajectories under different friction states based on the dynamic torque deviation, the friction state migration characteristics, and the friction fingerprint energy spectrum; generating drift correlation relationships between different friction states based on the drift direction change and drift rate change between each torque drift trajectory; performing dynamic drift correction on the output torque of the ultra-small torque wrench during continuous loading based on the drift correlation relationships; and generating a comprehensive calibration result of the ultra-small torque wrench that includes static calibration values ​​and dynamic drift characteristics based on the corrected output torque.

[0017] According to the second aspect, a calibration system for an ultra-small torque wrench based on micro-vibration energy spectrum inversion is provided. The system includes: a signal acquisition module for acquiring micro-vibration signals, transient energy fluctuation information, and local contact stiffness change information of the contact interface between the wrench output end and the workpiece under test during target torque loading; a friction fingerprint energy spectrum construction module for constructing a friction fingerprint energy spectrum under the current contact state based on the micro-vibration signals, the transient energy fluctuation information, and the local contact stiffness change information; and a friction state identification and evolution module for identifying the friction state at each moment during target torque loading based on the friction fingerprint energy spectrum, and constructing... The system generates a friction state sequence and establishes the state transition relationship between each friction state to generate a friction state evolution path. A stick-slip transition detection module detects local energy abrupt change intervals formed by the transition from a viscous state to a micro-slip state based on the friction state evolution path, determining the stick-slip transition trend and friction state migration characteristics of the contact interface. A torque inversion module constructs a nonlinear mapping relationship between friction state parameters and output torque based on the stick-slip transition trend, the friction state migration characteristics, and the friction fingerprint energy spectrum, inverting the actual output torque during the current target torque loading process. A calibration result generation module generates calibration results for the ultra-small torque wrench based on the actual output torque.

[0018] According to the specific embodiments provided in this application, the following technical effects are disclosed:

[0019] This invention constructs a uniquely identifiable friction fingerprint energy spectrum by collecting micro-vibration signals, transient energy fluctuations, and local contact stiffness changes at the contact interface during ultra-low torque loading. This enables precise identification of the friction state at each moment and generates a friction state evolution path that includes stick-slip transition trends and state transition characteristics. Based on this, a nonlinear mapping relationship between friction state parameters and output torque is established, effectively compensating for uncertainties such as nonlinear interface friction, stick-slip abrupt changes, and dynamic drift. The true output torque is then obtained through inversion, ultimately generating highly reliable calibration results. Compared to traditional static calibration methods, this method significantly improves measurement accuracy and signal-to-noise ratio in ultra-low torque scenarios, achieving dynamic, non-contact / weak-contact calibration, reducing the impact of contact disturbances, and exhibiting higher robustness and environmental adaptability. It is particularly suitable for micro-domains with extremely high torque control requirements, such as medical implants and precision instruments.

[0020] Of course, any product implementing this application does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 A flowchart illustrating the calibration method for an ultra-small torque wrench based on micro-vibration energy spectrum inversion provided in this application embodiment;

[0023] Figure 2 This is a schematic diagram showing the layout of the micro-vibration sensing unit provided in the embodiments of this application;

[0024] Figure 3 A structural block diagram of the ultra-small torque wrench calibration system based on micro-vibration energy spectrum inversion provided in this application embodiment;

[0025] Figure 4 A schematic block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0026] 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. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0027] The terminology used in the embodiments of this invention is for the purpose of describing particular embodiments only and is not intended to limit the invention. The singular forms “a,” “the,” and “the” as used in the embodiments of this invention and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise.

[0028] It should be understood that the term "and / or" used in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0029] Depending on the context, the word "if" as used here can be interpreted as "when," "when," "in response to determination," or "in response to detection." Similarly, depending on the context, the phrase "if determination" or "if detection (of the stated condition or event)" can be interpreted as "when determination," "in response to determination," "when detection (of the stated condition or event)," or "in response to detection (of the stated condition or event)."

[0030] Figure 1 A flowchart illustrating the ultra-small torque wrench calibration method based on micro-vibration energy spectrum inversion provided in this application embodiment. Figure 1 As shown, the method may include the following steps:

[0031] Step 101: Obtain the micro-vibration signal, transient energy fluctuation information, and local contact stiffness change information of the contact interface between the output end of the ultra-small torque wrench and the test piece during the target torque loading process.

[0032] Step 102: Based on the micro-vibration signal, the transient energy fluctuation information, and the local contact stiffness change information, construct the triboelectric fingerprint energy spectrum under the current contact state;

[0033] Step 103: Based on the triboelectric fingerprint energy spectrum, identify the friction state corresponding to each moment during the target torque loading process, construct a friction state sequence, establish the state transition relationship between each friction state, and generate a friction state evolution path;

[0034] Step 104: Detect the local energy abrupt change range formed by the transition from viscous state to microslip state according to the friction state evolution path, and determine the stick-slip transition trend and friction state migration characteristics of the contact interface.

[0035] Step 105: Based on the stick-slip transition trend, the friction state transition characteristics, and the friction fingerprint energy spectrum, construct a nonlinear mapping relationship between friction state parameters and output torque, and invert the actual output torque during the current target torque loading process;

[0036] Step 106: Generate the calibration results of the ultra-small torque wrench based on the actual output torque.

[0037] As can be seen from the above process, this invention constructs a uniquely identifiable friction fingerprint energy spectrum by collecting micro-vibration signals, transient energy fluctuation information, and local contact stiffness changes at the contact interface during ultra-low torque loading. This enables accurate identification of the friction state at each moment and generates a friction state evolution path that includes stick-slip transition trends and state transition characteristics. Based on this, a nonlinear mapping relationship between friction state parameters and output torque is constructed, effectively compensating for uncertainties such as nonlinear interface friction, stick-slip abrupt changes, and dynamic drift. The true output torque is then obtained through inversion, ultimately generating highly reliable calibration results. Compared with traditional static calibration methods, this method significantly improves measurement accuracy and signal-to-noise ratio in ultra-low torque scenarios, achieves dynamic, non-contact / weak-contact calibration, reduces the impact of contact disturbances, and exhibits higher robustness and environmental adaptability. It is particularly suitable for micro-domains with extremely high torque control requirements, such as medical implants and precision instruments.

[0038] The following describes in detail each step of the above process and the effects that can be further produced, with reference to the embodiments.

[0039] First, the above step 101, namely "acquiring the micro-vibration signal, transient energy fluctuation information, and local contact stiffness change information of the contact interface between the wrench output end and the test piece during the target torque loading process of the ultra-small torque wrench", will be described in detail with reference to the embodiments.

[0040] This step is the core step in signal acquisition in the entire calibration method. Its purpose is to comprehensively and synchronously capture the subtle physical changes that occur at the contact interface during the dynamic process of ultra-small torque loading, so as to provide a high-fidelity, multi-dimensional data foundation for the subsequent construction of the triboelectric fingerprint energy spectrum.

[0041] The target torque refers to the ideal torque value that is preset and expected to be output by the torque wrench during calibration or use. During actual loading, due to factors such as friction, contact conditions, and mechanical clearance, the actual output torque often deviates from the set target torque. This invention calculates the actual output torque through micro-vibration energy spectrum inversion, then compares it with the target torque to calculate the dynamic deviation, ultimately generating an accurate calibration result.

[0042] In practice, the first step is to precisely deploy micro-vibration sensing units at key locations. Specifically, sensors are installed directly at the output end of the wrench to capture the vibration transmitted by the wrench itself, sensors are placed in the contact edge area to monitor stress concentration and shear fluctuations at the interface edge, and sensors are also placed in the contact area of ​​the workpiece to record the response on the workpiece side. Figure 2This diagram illustrates the placement of the micro-vibration sensing unit according to an embodiment of this application. The diagram shows the specific sensor installation locations at the output end of the ultra-small torque wrench, the contact edge area, and the contact area of ​​the measured component. This multi-point, distributed placement method enables spatial complementarity, avoids blind spots caused by single-location measurements, and ensures a comprehensive reflection of vibration propagation and energy distribution across the entire contact interface.

[0043] Next, signal acquisition is performed simultaneously during the target torque loading process. At this time, the wrench applies a set small torque to the workpiece, and the sensor records transient vibration signals in real time: the overall weak vibration waveform, local high-frequency impact signals (brief high-frequency pulses caused by small protrusions or frictional abrupt changes at the contact interface), and contact stiffness fluctuation signals (dynamic fluctuations in local stiffness caused by changes in pressure and friction). These raw signals together constitute a complete "dynamic profile" of the contact interface at the moment of loading, capable of capturing microscopic nonlinear behaviors that are difficult to detect with traditional static measurements, such as minute stick-slip precursors or energy dissipation processes.

[0044] Finally, synchronous time-frequency processing was performed on the three types of raw signals acquired. This processing typically employs methods such as short-time Fourier transform or wavelet transform to simultaneously analyze both time and frequency dimensions. Synchronous processing precisely aligns different types of signals along the time axis, avoiding feature misalignment caused by sampling delays; simultaneously, it separates information from different frequency bands in the frequency domain, ultimately generating micro-vibration signals, transient energy fluctuation information, and local contact stiffness variation information that can be directly used for subsequent analysis. This processed information retains both the transient characteristics in time and highlights the energy distribution features in frequency, laying a solid foundation for constructing a high-resolution tribofinite energy spectrum.

[0045] The following describes in detail step 102, namely, "constructing the triboelectric fingerprint energy spectrum under the current contact state based on the micro-vibration signal, the transient energy fluctuation information, and the local contact stiffness change information," with reference to the embodiments.

[0046] This step transforms the acquired multi-dimensional signals into a highly unique and discriminative energy spectrum feature vector, enabling precise "fingerprinting" identification of interfacial friction behavior under different contact states, providing a reliable feature basis for subsequent friction state judgment and torque inversion.

[0047] First, multi-band energy decomposition is performed on the micro-vibration signal. Specifically, the micro-vibration signal is divided into several continuous or overlapping frequency bands, such as 0-2kHz, 2-8kHz, and 8-20kHz, using filter banks or wavelet packet decomposition. The local energy distribution is calculated within each frequency band, and the result is expressed by the following formula:

[0048] For the The local energy distribution results of each frequency band It can be represented as:

[0049]

[0050] in This represents the amplitude of the signal in the time-frequency domain. This step decomposes complex micro-vibration signals into energy contributions of different frequency components, highlighting the vibration characteristics of the contact interface at different scales.

[0051] Next, based on transient energy fluctuation information, the short-time energy mutation rate and spectral centroid shift for each frequency band are calculated. The short-time energy mutation rate reflects the degree of drastic energy change within a short time window, and its calculation formula is usually as follows:

[0052]

[0053] in For a short time window, This represents the statistical window length. The spectral centroid shift characterizes the offset of the energy distribution centroid, and the formula is:

[0054]

[0055]

[0056] These two parameters together capture the dynamic characteristics of transient impacts and rapid energy migration, and are an important basis for identifying early signs of stick-slip.

[0057] Then, based on the information on local contact stiffness variations, the high-frequency disturbance intensity and band energy dispersion of each frequency band are calculated. The high-frequency disturbance intensity quantifies the degree of enhancement of high-frequency components caused by stiffness fluctuations and can be expressed as:

[0058]

[0059] in This represents the change in local contact stiffness. Bandwidth energy dispersion reflects the uniformity of energy distribution within the band, and is commonly calculated using standard deviation or entropy.

[0060]

[0061] These parameters fully demonstrate the modulating effect of dynamic changes in the stiffness of the contact interface on vibration energy.

[0062] Finally, the aforementioned local energy distribution results, short-term energy mutation rate, spectral centroid shift, high-frequency perturbation intensity, and frequency band energy dispersion are normalized and then spliced ​​or weighted according to fixed dimensions to jointly constitute the triboelectric fingerprint energy spectrum under the current contact state. This spectrum not only includes the static energy distribution but also incorporates dynamic evolution indicators, forming a multidimensional fingerprint. This allows the contact interface under different friction states to have highly distinguishable spectral patterns, providing unique and stable feature inputs for subsequent state identification.

[0063] Through this series of precise decomposition and quantification calculations, this technology significantly improves the sensitivity and specificity of friction characteristics, effectively addressing the challenges of weak signals and complex noise under ultra-low torque, and laying a solid foundation for the high-precision inversion of the entire calibration method.

[0064] The following describes in detail step 103, namely, "based on the triboelectric fingerprint energy spectrum, identify the friction state corresponding to each moment during the target torque loading process, construct a friction state sequence, establish the state transition relationship between each friction state, and generate a friction state evolution path," with reference to the embodiments.

[0065] This step is the key link in the entire calibration method from the triboelectric fingerprint energy spectrum to the dynamic modeling of the triboelectric state. Its core lies in transforming the continuous spectral features into a discrete triboelectric state sequence and further constructing the evolution relationship between the states, thereby accurately depicting the dynamic triboelectric behavior evolution process of the contact interface during torque loading.

[0066] As torque is gradually applied to the wrench, the contact interface between the wrench output end and the workpiece being measured does not maintain the same friction mode. Instead, it continuously changes with the increase of torque, the change of time, the interface pressure, and relative micro-motion. This state of "current friction behavior" is the friction state corresponding to each moment.

[0067] First, a pre-defined set of friction states is obtained. This set includes at least four typical states: viscous state, where the interface is relatively stationary and energy is predominantly low-frequency; micro-slip state, where the interface begins to exhibit minute relative sliding and high-frequency components increase; periodic impact state, where the interface experiences regular micro-impacts and energy fluctuates periodically; and steady-state friction state, where interface friction is relatively stable and energy distribution is uniform. These pre-defined states constitute the basic classification framework for friction behavior, providing a reference template for subsequent identification.

[0068] Next, key features are extracted based on the triboelectric fingerprint energy spectrum, including the energy proportion of characteristic frequency bands, energy fluctuation gradient, and spectral discrete distribution characteristics. The energy proportion of characteristic frequency bands reflects the contribution of each frequency band to the total energy, and its calculation formula is as follows:

[0069]

[0070] in For the first Local energy of each frequency band Let be the total number of frequency bands. The energy fluctuation gradient describes the rate of change of energy over time and can be expressed as:

[0071]

[0072] The discrete distribution characteristics of the spectrum quantify the degree of energy dispersion along the frequency axis, and are commonly calculated using the standard deviation.

[0073]

[0074] These features together constitute a multi-dimensional quantitative description of the current contact state.

[0075] Then, friction state constraint vectors for each time step are constructed based on the extracted features. This constraint vector is a multi-dimensional feature vector, which can be denoted as:

[0076]

[0077] The vector dimension is consistent with the number of extracted features, fully representing the spectral mode at that moment.

[0078] Subsequently, the degree of coupling between each friction state constraint vector and the preset friction state is calculated. The degree of coupling can be measured using cosine similarity or weighted Euclidean distance; for example, the cosine similarity formula is:

[0079]

[0080] in For the first A template vector representing a preset friction state. The higher the coupling degree, the better the current spectral features match the preset state.

[0081] Finally, the preset friction state with the highest degree of state coupling is determined as the friction state at the corresponding moment, and they are connected sequentially in chronological order to generate a complete friction state sequence. This sequence intuitively reflects the trajectory of friction state changes over time during the loading process.

[0082] Based on the aforementioned friction state sequence, state transition relationships between each friction state are further established, and friction state evolution paths are generated. Specifically, firstly, friction state constraint vectors corresponding to adjacent time points are obtained. Then, state transition edges are constructed based on the energy distribution offset, spectral structure change, and perturbation growth trend between adjacent vectors. The energy distribution offset can be calculated as:

[0083]

[0084] The magnitude of spectral structure changes and the trend of perturbation growth can be quantified by changes in the spectral correlation coefficient or the high-frequency energy growth rate. These parameters collectively define the transition strength and direction between two adjacent states.

[0085] Next, a frictional state transition network is generated based on the migration direction, migration intensity, and duration of each transition edge. This network can be viewed as a directed weighted graph, where nodes represent frictional states and edges represent transition relationships.

[0086] In this network, high-frequency migration paths and cyclic migration paths are identified, forming closed-loop transfer patterns, such as the cyclical movement of viscosity-microslippage-viscosity. Finally, by integrating the high-frequency migration paths, cyclic migration paths, and the temporal correlations of each transfer edge, a complete friction state evolution path is generated. This evolution path not only records the order of state changes but also quantifies the intensity and trend of each transfer, providing a dynamic evolutionary basis for subsequent detection of stick-slip transitions and torque inversion.

[0087] For example, the specific implementation of generating a complete friction state evolution path includes the following steps:

[0088] First, path mining is performed on the frictional state transition network to extract high-frequency migration paths and cyclic migration paths. High-frequency migration paths are determined by statistically analyzing the frequency of each transition edge in the network throughout the entire loading time series. A frequency threshold is set, for example, if the number of occurrences exceeds 15% of the total number of transitions. Continuous transition sequences that meet this condition are defined as high-frequency migration paths, such as the frequently occurring transition pattern "viscous state → micro-slip state → periodic impact state". Cyclic migration paths are identified using a loop detection algorithm in graph theory to identify transition sequences with closed loops in the network, such as the recurring cycle of "viscous state → micro-slip state → viscous state", and the occurrence count and average period of each cycle are recorded.

[0089] Secondly, extract the temporal correlation relationships of each transition edge. For each transition edge in the network, record its occurrence time, duration, and information of its preceding and following adjacent transition edges in the time series, construct a temporal correlation matrix or temporal adjacency list, and quantify the sequential constraints and time interval statistical characteristics between different transition edges, such as the average transition waiting time and the time window distribution of transitions.

[0090] Finally, the complete friction state evolution path is generated by integrating the above three types of information. In specific implementation, the following fusion strategy is adopted: using time sequence as the basic framework, starting from the initial friction state at the beginning of loading; prioritizing the advancement of the evolution backbone along high-frequency migration paths to ensure the path conforms to the most common friction evolution patterns; when encountering branches or low-frequency transitions, path correction or completion is performed by referring to the statistical patterns of cyclic migration paths, such as maintaining the cyclic pattern within high-frequency cyclic intervals; and time alignment and smoothing are performed using the temporal correlations of each transition edge to ensure that the generated path is continuous and consistent on the time axis, for example, the time interval between adjacent transitions must satisfy a statistical distribution.

[0091] The final output friction state evolution path can be represented as a directed sequence with time labels:

[0092]

[0093] in For the first A friction state; For the corresponding transition edge, it includes attributes such as migration strength and period; For a moment.

[0094] This implementation method retains the typical evolution pattern that occurs frequently, while incorporating cyclic characteristics and strict temporal constraints. This makes the generated friction state evolution path both statistically representative and consistent with the time development law of the actual loading process, providing an accurate and complete dynamic evolution basis for subsequent stick-slip transition detection and torque inversion.

[0095] The following describes in detail step 104, namely, "detecting the local energy mutation interval formed by the transition from viscous state to microslip state according to the friction state evolution path, and determining the stick-slip transition trend and friction state migration characteristics of the contact interface," with reference to the embodiments.

[0096] The main purpose of this step is to accurately capture the critical moment of the transition from viscous state to microslip state in the friction state evolution path, and to quantify the energy mutation characteristics and overall migration law in this transition process, thereby providing an important nonlinear compensation basis for torque inversion.

[0097] First, state abrupt change nodes, representing the transition from a viscous state to a microslip state, are detected within the generated frictional state evolution path. These nodes are locations where specific state transitions occur between adjacent time points in the path, typically manifesting as a sudden shift from a viscous state dominated by low-frequency stable energy to a microslip state with increased high-frequency components. By traversing the time series of the evolution path, all locations satisfying the condition of "viscous state in the previous time step and microslip state in the next time step" are identified as state abrupt change nodes. This detection process can accurately pinpoint the critical moment when interfacial friction transitions from adhesion to the initiation of microslip.

[0098] Next, the local energy mutation interval corresponding to each state mutation node is extracted. This interval is centered on the mutation node and extends forward and backward by a certain time window (e.g., ±). This is a segment of energy change. Within this range, the release or redistribution of interfacial energy is most dramatic, making it the core data segment for capturing stick-slip phenomena.

[0099] Then, the local energy mutation interval is quantitatively calculated, mainly including three key indicators: spectral centroid shift, high-frequency disturbance growth rate, and transient impact amplification. The formula for calculating the spectral centroid shift is:

[0100]

[0101] in Calculate the spectral centroid before and after the mutation. This value reflects the overall migration direction and magnitude of energy distribution along the frequency axis.

[0102] The formula for calculating the high-frequency disturbance growth rate is:

[0103]

[0104] in This indicates the intensity of high-frequency disturbances. This index quantifies the degree of enhancement of high-frequency vibration components during the transition process and is directly related to the severity of sudden drops in interface stiffness and microslippage.

[0105] The formula for calculating the transient impact amplification is:

[0106]

[0107] in The peak energy value within the range. This is the baseline average energy. This parameter characterizes the intensity of the transient shock at the moment of abrupt change.

[0108] Based on the three quantitative indicators calculated above, the stick-slip transition trend is comprehensively judged. For example, when the spectral center of gravity shifts significantly to higher frequencies, the rate of increase of high-frequency disturbances exceeds a set threshold, and the transient impact increase is significant, a strong stick-slip transition trend can be identified; otherwise, a mild transition is identified. This trend provides physical mechanism support for subsequent torque correction.

[0109] Simultaneously, the migration intensity, migration period, and state dwell time of each state transition edge are extracted from the friction state transition network. The migration intensity can be quantified by the weighted values ​​on the transition edges, the migration period is the average time interval between adjacent transitions of the same type, and the state dwell time is the duration for which a certain state is continuously maintained in the path. Finally, a complete friction state transition feature is generated based on these parameters, including migration frequency, stability, and periodicity. This feature comprehensively describes the frictional dynamic evolution mode of the contact interface throughout the entire loading process.

[0110] Through the above detection, extraction, calculation, and feature generation processes, this technology transforms the abstract state evolution path into quantifiable stick-slip trends and migration characteristics, effectively revealing the nonlinear nature of interface friction under ultra-low torque loading. It provides key dynamic compensation information for constructing a high-precision torque inversion model, significantly improving the accuracy and reliability of the calibration method.

[0111] The following describes in detail step 105, namely, "based on the stick-slip transition trend, the friction state migration characteristics, and the friction fingerprint energy spectrum, constructing a nonlinear mapping relationship between friction state parameters and output torque, and inverting the actual output torque during the current target torque loading process," with reference to the embodiments.

[0112] This step organically integrates the previously obtained triboelectric fingerprint energy spectrum, stick-slip transition trend, and triboelectric state migration characteristics. By constructing a nonlinear mapping model, it achieves accurate inversion from complex interface friction dynamics to the actual output torque, effectively overcoming the measurement deviations caused by nonlinear friction, uncertainty, and dynamic disturbances during ultra-small torque loading.

[0113] First, a nonlinear mapping model between frictional state parameters, spectral energy parameters, and output torque is constructed, combining physical mechanisms with data-driven approaches. This model integrates classical tribological theories (such as the Coulomb friction model and stick-slip dynamics equations) with data-driven methods such as machine learning or neural networks, enabling it to simultaneously consider prior knowledge of physical laws and adaptability to actual experimental data. Its basic form can be expressed as:

[0114]

[0115] in To provide the actual output torque, This is a vector of friction state parameters, including state type, residence time, etc. The spectral energy parameter vector is derived from the tribofinite energy spectrum. These are the trainable parameters for the model. The model can be implemented using physically constrained neural networks or support vector regression, ensuring both mechanistic interpretability and strong generalization ability for complex working conditions.

[0116] Next, the correction amount for frictional abrupt change is determined based on the stick-slip transition trend. The stick-slip transition is the most significant source of nonlinear error under ultra-low torque; when a strong stick-slip transition is detected, the interface experiences transient energy release and a sudden torque drop. Therefore, the correction amount is calculated based on indicators such as spectral centroid shift, high-frequency disturbance growth rate, and transient impact amplification. The formula can be expressed as:

[0117]

[0118] in These are the weighting coefficients calibrated experimentally. This correction is used to compensate for abrupt changes in the inversion results, avoiding systematic errors caused by neglecting stick-slip in traditional methods.

[0119] Then, the state migration compensation amount is determined based on the friction state migration characteristics. These characteristics include migration intensity, migration period, and state residence time, which reflect the overall evolution of the friction state. The compensation amount calculation formula can be designed as follows:

[0120]

[0121] in For the first The strength of class transfer, For the migration cycle, For the length of stay, This is the weight. This compensation amount is used to correct the cumulative torque drift caused by repeated state transitions, making the inversion results more stable.

[0122] Finally, the triboelectric fingerprint energy spectrum, triboelectric abrupt change correction, and state transition compensation are input into the constructed nonlinear mapping model. The actual output torque during the current target torque loading process is obtained through forward computation or iterative optimization inversion. The complete inversion expression can be summarized as follows:

[0123]

[0124] Through this series of steps, the technology achieves a precise mapping from multi-source friction dynamics to actual torque, significantly improving the accuracy and reliability of ultra-small torque calibration.

[0125] Preferably, the present invention further includes: calculating the spectral energy entropy, the frequency band discrete entropy, and the perturbation chaos degree based on the triboelectric fingerprint energy spectrum; generating a triboelectric complexity parameter based on the spectral energy entropy, the frequency band discrete entropy, and the perturbation chaos degree; identifying the unsteady friction interval of the contact interface based on the triboelectric complexity parameter; and correcting the inversion result of the actual output torque based on the unsteady friction interval.

[0126] Specifically, firstly, the spectral energy entropy, frequency band discrete entropy, and perturbation chaos degree are calculated based on the triboelectric fingerprint energy spectrum. The spectral energy entropy measures the uniformity of energy distribution across different frequency bands, and its calculation formula is as follows:

[0127]

[0128] in For the first Normalized energy percentage of each frequency band This represents the total number of frequency bands. The larger this value, the more chaotic the energy distribution and the more irregular the friction process.

[0129] The frequency band discrete entropy further quantifies the degree of dispersion of energy distribution within each frequency band, and the calculation formula is as follows:

[0130]

[0131] in For the first The normalized sub-energy distribution within a frequency band. This index can capture subtle inhomogeneities within a frequency band.

[0132] The degree of disturbance chaos is used to describe the nonlinear chaotic characteristics of vibration disturbances, and can be calculated using approximate entropy or fuzzy entropy:

[0133]

[0134] in To reconstruct phase space where the distance is less than a threshold The vector ratio. The higher the value, the more chaotic the disturbance and the more unstable the interface friction.

[0135] Next, a friction complexity parameter is generated based on the above three indicators. This parameter comprehensively characterizes the complexity of the friction process through a weighted fusion method, and its calculation formula can be expressed as:

[0136]

[0137] in These are the weighting coefficients optimized through experiments. A higher friction complexity parameter value indicates a more complex frictional behavior at the current contact interface, making prediction more difficult.

[0138] Then, the unsteady friction intervals of the contact interface are identified based on the friction complexity parameter. A reasonable threshold is set; when the complexity parameter continuously exceeds this threshold, the corresponding time period is determined to be an unsteady friction interval. These intervals typically correspond to moments of rapid stick-slip transition, interface contamination, activation of micro-defects, or enhanced external disturbances, and are the regions where torque measurement errors are most easily amplified.

[0139] Finally, the inversion results of the actual output torque are corrected based on the identified unsteady friction range. Within the unsteady range, a compensation factor is introduced to smooth or correct the torque value. The correction method is as follows:

[0140]

[0141] in This step corrects the strength coefficient. It effectively suppresses abnormal torque drift in the unsteady region, resulting in a more stable and reliable final calibration result.

[0142] Through this series of complexity analyses and corrections based on information entropy and chaos theory, this technology significantly enhances the adaptability of the calibration method to complex working conditions and reduces system errors caused by unsteady interface behavior during ultra-small torque loading.

[0143] The following describes step 106, namely "generating the calibration result of the ultra-small torque wrench based on the actual output torque", in detail with reference to the embodiments.

[0144] This step transforms the actual output torque obtained from the inversion into practical and comprehensive calibration results, which include both static calibration values ​​and dynamic drift characteristics, thereby providing a reliable calibration basis and performance evaluation for the practical use of ultra-small torque wrenches.

[0145] First, calculate the dynamic torque deviation between the actual output torque and the target loading torque. This deviation reflects the real-time difference between the actual wrench output and the set target during loading, and is a fundamental indicator for evaluating calibration accuracy. The calculation formula is as follows:

[0146]

[0147] in To obtain the true output torque through inversion, The target torque is set. This dynamic deviation sequence can comprehensively record the error change patterns throughout the entire loading process.

[0148] Next, based on dynamic torque deviation, friction state transition characteristics, and triboelectric fingerprint energy spectrum, torque drift trajectories under different friction states are extracted. Within the time period of each friction state (e.g., viscous state, micro-slip state, etc.), the change in torque deviation over time is plotted as a trajectory curve. These trajectories not only contain the magnitude of the deviation but also record the direction and rate of deviation evolution with the state, thus forming a drift pattern unique to each state.

[0149] Specifically, the friction state label corresponding to each moment is obtained from the previously generated friction state sequence, and the corresponding friction state migration features are extracted, including migration intensity, migration period and state residence time, as well as the complete feature vector of the friction fingerprint energy spectrum at that moment.

[0150] Then, the time series is segmented according to friction states. For each preset friction state, all continuous or discrete time segments belonging to that state are identified. For example, all time points corresponding to the viscous state. Within this segment, extract the dynamic torque deviation subsequence. The corresponding triboelectric fingerprint energy spectrum sequence and state transition feature parameters.

[0151] Next, a torque drift trajectory is constructed within the time segment of each state. Specifically, time is used as the horizontal axis and dynamic torque deviation as the vertical axis, while key dimensions from the triboelectric fingerprint energy spectrum, such as the spectral centroid or high-frequency energy percentage, are integrated as auxiliary coordinates or color / coarseness encoding to form a multi-dimensional trajectory curve. For each friction state... Its torque drift trajectory can be expressed as:

[0152]

[0153] in These are representative features extracted from the energy spectrum of tribofinite fingerprints, such as the combination of short-time energy mutation rate and spectral centroid shift.

[0154] During trajectory construction, frictional state transition features are also used for smoothing and weighting. For time points near transition edges with high state transition intensity, their weight in the trajectory is appropriately increased to highlight the drift characteristics of the transition phase. Simultaneously, piecewise linear fitting or spline interpolation is performed on the trajectory to obtain a smooth drift path, and the statistical characteristics of the trajectory, such as average drift rate, maximum drift amplitude, and directional change trend, are calculated.

[0155] Then, based on the changes in drift direction and drift rate between each torque drift trajectory, drift correlations between different friction states are generated. The change in drift direction can be calculated using the angle between trajectory vectors, and the change in drift rate can be expressed as the difference in slope between adjacent trajectory states. The correlation can be quantified using correlation coefficients or transition matrices; for example, the drift correlation strength formula is:

[0156]

[0157] in and The first and the The drift trajectories under various friction conditions were analyzed. This correlation reveals the mutual influence of torque drift under different friction conditions, providing an important reference for subsequent corrections.

[0158] Based on the aforementioned drift correlation, dynamic drift correction is performed on the output torque of the ultra-small torque wrench during continuous loading. This correction eliminates the cumulative error caused by state transition, making the torque output more stable under continuous working conditions. The correction formula can be expressed as:

[0159]

[0160] in This represents the drift component corresponding to the state. This step significantly improves the applicability of the calibration results under actual dynamic operating conditions.

[0161] Finally, a comprehensive calibration result for the ultra-small torque wrench, including static calibration values ​​and dynamic drift characteristics, is generated based on the corrected output torque. The static calibration values ​​are typically taken as the average deviation or correction value during the steady-state phase of the loading process, while the dynamic drift characteristics are presented in the form of drift trajectory, correlation, and maximum drift amount. This comprehensive result can be used for single calibration or to provide data support for real-time compensation during continuous use.

[0162] Specifically, the time period corresponding to the steady-state friction state (or the stable interval with small energy fluctuations throughout the entire loading process) is extracted from the corrected torque sequence. The average deviation between the actual output torque and the target torque within this interval is calculated and used as the static calibration value. The calculation formula is as follows:

[0163]

[0164] It can also calculate the static repeatability error and the maximum static deviation, thereby obtaining a stable and reliable static calibration parameter.

[0165] Secondly, dynamic drift characteristics are extracted. Based on the corrected torque sequence and the torque drift trajectories generated previously under various friction states, the following key dynamic indicators are statistically analyzed: Maximum dynamic drift amount: Dynamic drift rate: the average slope of the drift trajectory under various states; drift correlation amplitude between different friction states, derived from the drift correlation matrix; overall drift fluctuation range; maximum transient error within the unsteady-state interval. These indicators together constitute the dynamic drift characteristic vector, comprehensively describing the unstable behavior of the wrench during continuous loading. Finally, the static calibration values ​​and dynamic drift characteristics are integrated to generate a comprehensive calibration result for the ultra-small torque wrench.

[0166] To further illustrate the technical effects of this application, a specific implementation method and the test results of this implementation method are given below.

[0167] This embodiment provides a typical implementation of a calibration method for an ultra-small torque wrench based on micro-vibration energy spectrum inversion. A precision micro-torque wrench with a range of 0.01~1.0 mN·m is used as the calibration object, and the test piece is a 5mm diameter stainless steel micro-screw. High-sensitivity piezoelectric micro-vibration sensing units are deployed at the wrench output end, the contact edge area, and the contact area of ​​the test piece, with a sensitivity of 50 mV / g and a sampling frequency of 50 kHz. The target torque is set to 0.25 mN·m, and a continuous loading experiment is conducted at a loading rate of 0.05 mN·m / s.

[0168] During loading, micro-vibration signals, transient energy fluctuations, and local contact stiffness changes are acquired in real time. A triboelectric fingerprint energy spectrum is constructed through multi-band energy decomposition (divided into six frequency bands: 0-2 kHz, 2-8 kHz, and 8-25 kHz). Subsequently, viscous, micro-slip, periodic impact, and steady-state friction states are identified, generating a friction state sequence and evolution path. After detecting three distinct transition points from viscous to micro-slip, the spectral centroid shift, high-frequency disturbance growth rate, and transient impact amplification are calculated. A nonlinear mapping model based on a combination of physical mechanisms and a BP neural network is constructed to invert and obtain the true output torque. Finally, after dynamic drift correction, a comprehensive calibration result including static calibration values ​​and dynamic drift characteristics is generated.

[0169] Twenty repeated calibration experiments were conducted within the range of 0.05–0.5 mN·m. The average static calibration deviation of the method of this invention was ±0.009 mN·m, and the standard deviation of repeatability was 0.006 mN·m. Compared with the traditional static strain gauge calibration method, the maximum error was reduced from ±0.038 mN·m to ±0.011 mN·m, with an accuracy improvement of approximately 71% and an average signal-to-noise ratio improvement of 12.6 dB. Under continuous loading conditions, the torque fluctuation range after dynamic drift correction was reduced from ±0.027 mN·m to ±0.008 mN·m, demonstrating a significant correction effect in the unsteady-state range. Experiments show that this method has high accuracy, robustness, and practicality in ultra-low torque scenarios, and is particularly suitable for the fields of minimally invasive medical devices and MEMS precision assembly.

[0170] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0171] According to another embodiment, an ultra-small torque wrench calibration system based on micro-vibration energy spectrum inversion is provided. Figure 3 A schematic block diagram of an ultra-small torque wrench calibration system based on micro-vibration energy spectrum inversion according to one embodiment is shown. Figure 3 As shown, the device 300 includes:

[0172] The signal acquisition module 301 is used to acquire the micro-vibration signal, transient energy fluctuation information of the contact interface between the output end of the ultra-small torque wrench and the test piece during the target torque loading process.

[0173] The tribofinite fingerprint energy spectrum construction module 302 is used to construct the tribofinite fingerprint energy spectrum under the current contact state based on the micro-vibration signal, the transient energy fluctuation information, and the local contact stiffness change information.

[0174] The friction state identification and evolution module 303 is used to identify the friction state corresponding to each moment during the target torque loading process based on the friction fingerprint energy spectrum, construct a friction state sequence, establish the state transition relationship between each friction state, and generate a friction state evolution path.

[0175] The stick-slip transition detection module 304 is used to detect the local energy abrupt change range formed by the transition from the viscous state to the micro-slip state according to the friction state evolution path, and to determine the stick-slip transition trend and friction state migration characteristics of the contact interface.

[0176] The torque inversion module 305 is used to construct a nonlinear mapping relationship between friction state parameters and output torque based on the stick-slip transition trend, the friction state migration characteristics, and the friction fingerprint energy spectrum, and to invert the actual output torque during the current target torque loading process.

[0177] The calibration result generation module 306 is used to generate calibration results for the ultra-small torque wrench based on the actual output torque.

[0178] As an feasible approach, the signal acquisition module 301 acquires the micro-vibration signal, transient energy fluctuation information, and local contact stiffness change information of the contact interface between the wrench output end and the workpiece under test during the target torque loading process. This includes: deploying micro-vibration sensing units at the wrench output end, the contact edge area, and the contact area of ​​the workpiece under test; acquiring transient vibration signals, local high-frequency impact signals, and contact stiffness fluctuation signals during the target torque loading process; and performing synchronous time-frequency processing on the transient vibration signals, local high-frequency impact signals, and contact stiffness fluctuation signals to generate micro-vibration signals, transient energy fluctuation information, and local contact stiffness change information.

[0179] As an implementable method, the tribofinite fingerprint energy spectrum construction module 302 constructs the tribofinite fingerprint energy spectrum under the current contact state based on the micro-vibration signal, the transient energy fluctuation information, and the local contact stiffness change information. This includes: performing multi-band energy decomposition on the micro-vibration signal to generate local energy distribution results for multiple frequency bands; calculating the short-time energy mutation rate and spectral centroid shift of each frequency band based on the transient energy fluctuation information; calculating the high-frequency disturbance intensity and frequency band energy dispersion of each frequency band based on the local contact stiffness change information; and constructing the tribofinite fingerprint energy spectrum based on the local energy distribution results, the short-time energy mutation rate, the spectral centroid shift, the high-frequency disturbance intensity, and the frequency band energy dispersion.

[0180] As an implementable approach, the friction state identification and evolution module 303 identifies the friction state corresponding to each moment during the target torque loading process based on the friction fingerprint energy spectrum, and constructs a friction state sequence, including: obtaining a preset friction state set, which at least includes viscous state, micro-slip state, periodic impact state, and steady-state friction state; extracting the energy proportion of characteristic frequency bands, energy fluctuation gradient, and spectral discrete distribution features based on the friction fingerprint energy spectrum; constructing friction state constraint vectors corresponding to each moment according to the energy proportion of characteristic frequency bands, the energy fluctuation gradient, and the spectral discrete distribution features; calculating the degree of state coupling between each friction state constraint vector and each preset friction state; determining the preset friction state with the highest degree of state coupling as the friction state at the corresponding moment, and generating a friction state sequence in chronological order.

[0181] As an implementable approach, the friction state identification and evolution module 303 establishes state transition relationships between various friction states and generates friction state evolution paths, including: obtaining friction state constraint vectors corresponding to adjacent moments in the friction state sequence; constructing state transition edges between various friction states based on the energy distribution offset, spectral structure change, and disturbance growth trend between adjacent friction state constraint vectors; generating a friction state transition network based on the migration direction, migration intensity, and duration of each state transition edge; identifying high-frequency migration paths and cyclic migration paths in the friction state transition network; and generating friction state evolution paths based on the temporal correlation relationships of the high-frequency migration paths, the cyclic migration paths, and each state transition edge.

[0182] As an implementable approach, the stick-slip transition detection module 304 detects local energy abrupt change intervals formed by the transition from a viscous state to a microslip state according to the friction state evolution path, and determines the stick-slip transition trend and friction state migration characteristics of the contact interface. This includes: detecting state abrupt change nodes formed by the transition from a viscous state to a microslip state in the friction state evolution path; extracting the local energy abrupt change intervals corresponding to the state abrupt change nodes; calculating the spectral centroid shift, high-frequency disturbance growth rate, and transient impact amplification in the local energy abrupt change intervals; determining the stick-slip transition trend based on the spectral centroid shift, the high-frequency disturbance growth rate, and the transient impact amplification; extracting the migration intensity, migration period, and state dwell time of each state transition edge based on the friction state transition network; and generating friction state migration characteristics based on the migration intensity, the migration period, and the state dwell time.

[0183] As an implementable approach, the torque inversion module 305 constructs a nonlinear mapping relationship between friction state parameters and output torque based on the stick-slip transition trend, the friction state transition characteristics, and the friction fingerprint energy spectrum, and inverts the true output torque during the current target torque loading process. This includes: constructing a nonlinear mapping model between friction state parameters, spectral energy parameters, and output torque based on a combination of physical mechanisms and data-driven approaches; determining the friction abrupt change correction amount based on the stick-slip transition trend; determining the state transition compensation amount based on the friction state transition characteristics; and inverting the true output torque based on the friction fingerprint energy spectrum, the friction abrupt change correction amount, and the state transition compensation amount.

[0184] As an implementable approach, the torque inversion module 305 further includes: calculating the spectral energy entropy, frequency band discrete entropy, and perturbation chaos degree based on the triboelectric fingerprint energy spectrum; generating a triboelectric complexity parameter based on the spectral energy entropy, the frequency band discrete entropy, and the perturbation chaos degree; identifying the unsteady friction interval of the contact interface based on the triboelectric complexity parameter; and correcting the inversion result of the true output torque based on the unsteady friction interval.

[0185] As an implementable approach, the calibration result generation module 306 generates calibration results for the ultra-small torque wrench based on the actual output torque, including: calculating the dynamic torque deviation between the actual output torque and the target loading torque; extracting torque drift trajectories under different friction states based on the dynamic torque deviation, the friction state migration characteristics, and the friction fingerprint energy spectrum; generating drift correlations between different friction states based on the drift direction change and drift rate change between each torque drift trajectory; performing dynamic drift correction on the output torque of the ultra-small torque wrench during continuous loading based on the drift correlations; and generating a comprehensive calibration result for the ultra-small torque wrench that includes static calibration values ​​and dynamic drift characteristics based on the corrected output torque.

[0186] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, for system or system embodiments, since they are basically similar to method embodiments, the description is relatively simple, and relevant parts can be referred to the descriptions in the method embodiments. The systems and system 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 creative effort.

[0187] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation entry points are provided for users to choose to authorize or refuse.

[0188] In addition, embodiments of this application also provide a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the method described in any of the foregoing method embodiments.

[0189] And an electronic device comprising: one or more processors; and a memory associated with the one or more processors, the memory being used to store program instructions that, when read and executed by the one or more processors, perform the steps of the method described in any of the foregoing method embodiments.

[0190] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the method described in any of the foregoing method embodiments.

[0191] in, Figure 4An exemplary architecture of an electronic device is shown, which may include a processor 410, a video display adapter 411, a disk drive 412, an input / output interface 413, a network interface 414, and a memory 420. The processor 410, video display adapter 411, disk drive 412, input / output interface 413, network interface 414, and memory 420 can communicate with each other via a communication bus 430.

[0192] The processor 410 can be implemented using a general-purpose CPU, microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits to execute relevant programs and implement the technical solution provided in this application.

[0193] The memory 420 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 420 can store the operating system 421 for controlling the operation of the electronic device 400, and the basic input / output system BIOS 422 for controlling the low-level operations of the electronic device 400. Additionally, it can store a web browser 423, a data storage management system 424, and an ultra-small torque wrench calibration system 425 based on micro-vibration energy spectrum inversion, etc. The aforementioned ultra-small torque wrench calibration system 425 based on micro-vibration energy spectrum inversion can be the application program that specifically implements the aforementioned steps in this embodiment. In summary, when implementing the technical solution provided in this application through software or firmware, the relevant program code is stored in the memory 420 and executed by the processor 410.

[0194] Input / output interface 413 is used to connect input / output modules to realize information input and output. Input / output modules can be configured as components in the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Input devices may include keyboards, mice, touch screens, microphones, various sensors, etc., and output devices may include displays, speakers, vibrators, indicator lights, etc.

[0195] Network interface 414 is used to connect a communication module (not shown in the figure) to enable communication and interaction between this device and other devices. The communication module can communicate via wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0196] Bus 430 includes a pathway for transmitting information between various components of the device, such as processor 410, video display adapter 411, disk drive 412, input / output interface 413, network interface 414, and memory 420.

[0197] It should be noted that although the above-described device only shows the processor 410, video display adapter 411, disk drive 412, input / output interface 413, network interface 414, memory 420, bus 430, etc., in specific implementations, the device may also include other components necessary for normal operation. Furthermore, those skilled in the art will understand that the above-described device may only include the components necessary for implementing the solution of this application, and does not necessarily include all the components shown in the figures.

[0198] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a computer program product. This computer program product can be stored in a 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, or network device, etc.) to execute the methods described in various embodiments or some parts of the embodiments of this application.

[0199] The technical solutions provided in this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A calibration method for an ultra-small torque wrench based on micro-vibration energy spectrum inversion, characterized in that, The method includes: The micro-vibration signal, transient energy fluctuation information of the contact interface between the wrench output end and the test piece, and local contact stiffness change information of the ultra-small torque wrench during the target torque loading process are obtained. Based on the micro-vibration signal, the transient energy fluctuation information, and the local contact stiffness change information, a tribofinite energy spectrum of the current contact state is constructed. Based on the triboelectric fingerprint energy spectrum, the friction state corresponding to each moment during the target torque loading process is identified, a friction state sequence is constructed, and the state transition relationship between each friction state is established to generate a friction state evolution path. Based on the friction state evolution path, detect the local energy mutation interval formed by the transition from viscous state to microslip state, and determine the stick-slip transition trend and friction state migration characteristics of the contact interface. Based on the stick-slip transition trend, the friction state transition characteristics, and the friction fingerprint energy spectrum, a nonlinear mapping relationship between friction state parameters and output torque is constructed to invert the actual output torque during the current target torque loading process. The calibration results for the ultra-small torque wrench are generated based on the actual output torque.

2. The method according to claim 1, characterized in that, The acquisition of micro-vibration signals, transient energy fluctuation information, and local contact stiffness changes at the interface between the wrench output end and the workpiece under test during the target torque loading process includes: Micro-vibration sensing units are installed at the output end of the wrench, the contact edge area, and the contact area of ​​the workpiece under test. Acquire transient vibration signals, local high-frequency impact signals, and contact stiffness fluctuation signals during the target torque loading process; Synchronous time-frequency processing is performed on the transient vibration signal, local high-frequency impact signal, and contact stiffness fluctuation signal to generate micro-vibration signal, transient energy fluctuation information, and local contact stiffness change information.

3. The method according to claim 1, characterized in that, The construction of the triboelectric fingerprint energy spectrum under the current contact state based on the micro-vibration signal, the transient energy fluctuation information, and the local contact stiffness change information includes: Multi-band energy decomposition is performed on the micro-vibration signal to generate local energy distribution results in multiple frequency bands; Based on the transient energy fluctuation information, the short-time energy mutation rate and spectral centroid shift of each frequency band are calculated; The high-frequency disturbance intensity and frequency band energy dispersion of each frequency band are calculated based on the local contact stiffness change information. The triboelectric fingerprint energy spectrum is constructed based on the local energy distribution results, the short-time energy mutation rate, the spectral centroid shift, the high-frequency perturbation intensity, and the frequency band energy dispersion.

4. The method according to claim 1, characterized in that, The step of identifying the friction state at each moment during the target torque loading process based on the triboelectric fingerprint energy spectrum and constructing a friction state sequence includes: Obtain a preset set of friction states, which includes at least a viscous state, a micro-slip state, a periodic impact state, and a steady-state friction state. Based on the tribofinite fingerprint energy spectrum, the characteristic frequency band energy ratio, energy fluctuation gradient, and spectral discrete distribution characteristics are extracted. Based on the energy proportion of the characteristic frequency band, the energy fluctuation gradient, and the discrete distribution characteristics of the spectrum, a friction state constraint vector corresponding to each moment is constructed. Calculate the degree of state coupling between each friction state constraint vector and each preset friction state; The preset friction state with the highest degree of state coupling is determined as the friction state at the corresponding moment, and a friction state sequence is generated in chronological order.

5. The method according to claim 4, characterized in that, The process of establishing state transition relationships between various friction states and generating friction state evolution paths includes: Obtain the friction state constraint vectors corresponding to adjacent time points in the friction state sequence; Based on the energy distribution offset, spectral structure change and disturbance growth trend between friction state constraint vectors at adjacent time points, state transition edges between each friction state are constructed. A frictional state transition network is generated based on the migration direction, migration intensity, and duration of each state transition edge. Identify high-frequency migration paths and cyclic migration paths in the frictional state transition network; Based on the high-frequency migration path, the cyclic migration path, and the temporal correlation of each state transition edge, a friction state evolution path is generated.

6. The method according to claim 5, characterized in that, The step of detecting the local energy abrupt change interval formed by the transition from the viscous state to the microslip state based on the friction state evolution path, and determining the stick-slip transition trend and friction state migration characteristics of the contact interface, includes: Detecting abrupt state transition nodes in the friction state evolution path, where the state changes from a viscous state to a microslip state; Extract the local energy mutation intervals corresponding to the state mutation nodes; Calculate the spectral centroid shift, high-frequency disturbance growth rate, and transient impact amplification in the local energy mutation interval; The stick-slip transition trend is determined based on the spectral centroid shift, the high-frequency disturbance growth rate, and the transient impact amplification. Based on the aforementioned frictional state transition network, the migration intensity, migration period, and state dwell time of each state transition edge are extracted. Frictional state migration features are generated based on the migration intensity, the migration period, and the state residence time.

7. The method according to claim 1, characterized in that, The process of constructing a nonlinear mapping relationship between friction state parameters and output torque based on the stick-slip transition trend, the friction state transition characteristics, and the friction fingerprint energy spectrum, and inverting the actual output torque during the current target torque loading process, includes: A nonlinear mapping model between friction state parameters, spectral energy parameters, and output torque is constructed based on a combination of physical mechanisms and data-driven approaches. The amount of frictional abrupt change correction is determined based on the described stick-slip transition trend; The state migration compensation amount is determined based on the aforementioned friction state migration characteristics. The true output torque is inverted based on the triboelectric fingerprint energy spectrum, the triboelectric mutation correction amount, and the state transition compensation amount.

8. The method according to claim 1, characterized in that, The method further includes: Spectral energy entropy, frequency band discrete entropy, and perturbation chaos degree are calculated based on the tribo fingerprint energy spectrum. The frictional complexity parameter is generated based on the spectral energy entropy, the frequency band discrete entropy, and the perturbation chaos degree. The non-steady-state friction range of the contact interface is identified based on the aforementioned friction complexity parameter; The inversion result of the true output torque is corrected based on the aforementioned unsteady friction range.

9. The method according to claim 1, characterized in that, The process of generating the calibration result for the ultra-small torque wrench based on the actual output torque includes: Calculate the dynamic torque deviation between the actual output torque and the target loaded torque; Based on the dynamic torque deviation, the friction state migration characteristics, and the friction fingerprint energy spectrum, the torque drift trajectory under different friction states is extracted; Based on the changes in drift direction and drift rate between each torque drift trajectory, drift correlations between different friction states are generated. Based on the aforementioned drift correlation, dynamic drift correction is performed on the output torque of the ultra-small torque wrench during continuous loading. A comprehensive calibration result for the ultra-small torque wrench, including static calibration values ​​and dynamic drift characteristics, is generated based on the corrected output torque.

10. A calibration system for an ultra-small torque wrench based on micro-vibration energy spectrum inversion, characterized in that, The system includes: The signal acquisition module is used to acquire the micro-vibration signal, transient energy fluctuation information of the contact interface between the wrench output end and the test piece during the target torque loading process of the ultra-small torque wrench. The tribofinite fingerprint energy spectrum construction module is used to construct the tribofinite fingerprint energy spectrum under the current contact state based on the micro-vibration signal, the transient energy fluctuation information, and the local contact stiffness change information. The friction state identification and evolution module is used to identify the friction state corresponding to each moment during the target torque loading process based on the friction fingerprint energy spectrum, construct a friction state sequence, establish the state transition relationship between each friction state, and generate a friction state evolution path. The stick-slip transition detection module is used to detect the local energy abrupt change range formed by the transition from the viscous state to the micro-slip state according to the friction state evolution path, and to determine the stick-slip transition trend and friction state migration characteristics of the contact interface. The torque inversion module is used to construct a nonlinear mapping relationship between friction state parameters and output torque based on the stick-slip transition trend, the friction state transition characteristics and the friction fingerprint energy spectrum, and to invert the actual output torque during the current target torque loading process. The calibration result generation module is used to generate calibration results for the ultra-small torque wrench based on the actual output torque.

Images