A multi-condition in-situ testing and intelligent coating integrated machine for slip rings and its usage method

Abstract

This invention provides an integrated machine for in-situ testing and intelligent coating of slip rings under multiple working conditions, and its usage method, belonging to the field of precision mechanics and friction testing technology. The integrated machine integrates power drive, quantitative dispensing, servo motor adjustment, quick-change fixture, coating module, sensing and measurement module, and information acquisition and processing module. In-situ switching between coating, testing, and cleaning stations is achieved through servo motor rotation. The quick-change fixture is compatible with various slip rings and brushes. The sensing module simultaneously collects force, heat, vibration, acoustic emission, and electrical signals. The information processing module incorporates a neural network with physical information constraints, which can identify six working conditions in real time: break-in, stable wear, insufficient lubrication, particulate contamination, arc erosion, and failure warning. It predicts the remaining life of the coating and autonomously optimizes test parameters and generates a health index based on deep reinforcement learning. This invention achieves full-process in-situ automation and intelligent decision-making closed-loop for slip ring testing, with strong versatility, rich data dimensions, and significantly improved testing efficiency and intelligence level.

Description

Technical Field

[0001] This invention relates to the field of precision mechanics and friction testing technology, and in particular to an integrated machine for multi-condition in-situ testing and intelligent coating of slip rings, and its usage method. Background Technology

[0002] Conductive slip rings are key electromechanical components that ensure stable transmission of electrical energy and signals between rotating and stationary parts. Their reliability mainly depends on the wear state and electrical contact performance of the friction pair formed by the brush bristles and the ring channel under long-term current-carrying operation. Therefore, the screening and evaluation of high-performance lubricants is crucial for extending slip ring life. However, current performance evaluation methods for slip ring lubricants generally suffer from fragmented testing processes. Traditional methods require repeated disassembly and assembly of the workpiece between coating equipment, friction testing machines, and cleaning stations, which is not only inefficient but also introduces positioning errors that are difficult to eliminate due to repeated clamping, resulting in test results that cannot accurately reflect the performance of the lubricant in the initial uniform coating state. At the same time, existing coating processes are outdated and uncontrollable. The slip ring channel space is narrow and the surface is complex. Manual brush coating or drop application methods suffer from poor coating uniformity, inaccurate control of lubricant dosage, and low repeatability. There is a lack of integrated and precise control methods for coating speed, pressure, and uniformity.

[0003] Furthermore, existing equipment and fixtures lack versatility. Commercially available slip ring polishers or friction testing machines are typically designed only for specific sizes or single-model slip rings, failing to meet the rapid, non-destructive clamping and testing needs of various types of slip rings, such as columnar, disc, cap-type, and split slip rings, as well as different forms of brushes, including filaments, brush bundles, metal sheets, and brush blocks. Regarding data monitoring, traditional equipment has limited monitoring dimensions, typically measuring only friction or basic contact resistance. It cannot simultaneously acquire key parameters reflecting early failure signs, such as dynamic contact resistance fluctuations (electrical noise), abnormal vibrations, acoustic emission signals, and contact point temperature rise. This severely restricts in-depth research and forward-looking health diagnosis of the failure mechanism of slip rings—a thermo-mechanical-electric multi-field coupled system. In terms of data processing and feedback control, most existing equipment can only perform simple signal recording and post-test manual analysis, lacking the ability to use multi-physics monitoring data for real-time equipment action control. It cannot automatically adjust coating parameters, contact pressure, or cleaning processes based on dynamic responses during testing, resulting in insufficient repeatability and adaptability of test results. At the same time, the existing equipment has a low degree of automation, the cleaning process relies on manual labor, and it cannot be integrated into the automated testing process.

[0004] Existing technologies, such as the multifunctional vacuum friction and wear test device and its usage method for conductive slip rings disclosed in Chinese invention patent application CN118310911A, although the device can simulate a vacuum environment and test the friction coefficient and contact resistance of the base, still have significant defects such as a crude lubricant introduction method, lack of automated in-situ cleaning function, lack of integration of multi-dimensional sensors such as vibration and acoustic emission into the data acquisition system, reliance on manual data post-processing and inability to achieve closed-loop control, insufficient fixture versatility, and inability to achieve fully automated operation from coating, testing to cleaning. Summary of the Invention

[0005] The purpose of this invention is to provide an integrated machine for in-situ testing and intelligent coating of slip rings under multiple operating conditions, as well as its usage method. This invention addresses the problems in existing technologies, such as discrete testing processes requiring repeated disassembly and assembly, coarse coating processes with uncontrollable dosage, poor fixture versatility, single-dimensional monitoring data, lack of a fully automated in-situ operation method covering the entire process from coating to testing to cleaning, and the inability to achieve intelligent closed-loop control of process parameters based on multi-physics field data. This invention enables efficient, accurate, and comprehensive evaluation of slip ring lubricant performance and friction pair operating conditions.

[0006] To achieve the above objectives, the present invention provides an integrated machine for multi-condition in-situ testing and intelligent coating of slip rings, comprising: frame; The power drive module is mounted on the frame; The quantitative dispensing module is installed on the power drive module; The upper test piece servo adjustment module is installed on the power drive module; The upper specimen quick-change fixture is installed on the upper specimen servo adjustment module and is used to replaceably hold the coating module or brush. The coating module is used to coat the slip ring track with liquid; The specimen quick-change fixture is installed on the power drive module to hold the slip ring rotor; The sensing and measurement module is used to acquire multi-physics field signals in real time during the testing process; The information acquisition and processing module is connected to the sensing and measurement module and is used to synchronously acquire, process and analyze multi-physics field signals.

[0007] Preferably, the power drive module is used to provide linear motion along the X, Y, and Z directions and rotational motion about the C-axis.

[0008] Preferably, the quantitative dispensing module includes an upper specimen connecting plate, a liquid storage chamber, a micro-injection pump, and pipeline valves; the quantitative dispensing module is mounted on the linear motion component of the power drive module via the upper specimen connecting plate; the input end of the micro-injection pump is connected to the liquid storage chamber, and the output end is connected to the dispensing module via pipeline.

[0009] Preferably, the upper specimen servo adjustment module includes a micro digital servo and a rotating block driven by the servo. The rotating block is connected to the output shaft of the micro digital servo. The upper specimen quick-change fixture is mounted on the rotating block, and the rotating block can rotate from 0° to 180°.

[0010] Preferably, the coating module includes a coating head holder, a coating head fixing plate, and a flexible coating head. The coating head holder is held by a quick-change fixture for the upper specimen. The coating head fixing plate is installed on the coating head holder, and the flexible coating head is fixed on the coating head fixing plate. The coating head holder and the coating head fixing plate are provided with flow channels that communicate with the flexible coating head.

[0011] Preferably, the quick-change fixture for the lower specimen includes a precision three-jaw chuck, and an expansion mandrel and a resilient bushing that are interchangeably mounted on the chuck; the expansion mandrel is used to hold a slip ring positioned by its inner hole, and the resilient bushing is used to hold a slip ring positioned by its outer circle.

[0012] Preferably, the sensing and measurement module includes: A three-dimensional force sensor is installed on the flip block of the upper specimen servo adjustment module; An accelerometer is mounted on the flip block; Temperature sensor and acoustic emission sensor are installed on the upper specimen quick-change fixture near the brush. An auxiliary slip ring, whose rotor is connected to the quick-change fixture for the lower specimen; Circuit test bracket, used to fix test circuits; The electrical signal acquisition chain adopts a four-wire Kelvin connection method and includes a precision constant current source, a high-precision DC voltmeter, a high-pass filter, a low-noise amplifier, and a high-speed data acquisition card.

[0013] The present invention also provides a method for using the above-mentioned slip ring multi-condition in-situ testing and intelligent coating integrated machine, comprising the following steps: Step S1: Install the slip ring rotor: Install the slip ring rotor onto the lower specimen quick-change fixture; Step S2: Prepare the coating: Install the coating module on the upper specimen quick-change fixture, adjust the upper specimen servo motor adjustment module to the coating station and angle, and control the power drive module to make the coating head of the coating module contact the slip ring track and reach the preset pressure. Step S3, Start homogenization coating: Start the rotation shaft of the quantitative drip coating module and the power drive module to quantitatively coat the lubricant onto the slip ring track; Step S4: Prepare for current-carrying friction test: Replace the coating module with a brush, adjust the upper specimen servo adjustment module to the test position and angle, control the power drive module to make the brush contact the slip ring track and reach the preset brush pressure, and connect the current-carrying test circuit. Step S5: Start the current-carrying friction and wear test: Apply the set current, start the rotating shaft of the power drive module to drive the slip ring to rotate, and at the same time start the sensing and measurement module to collect multi-physics field data synchronously. Step S6: Prepare for cleaning: Replace the brush with a coating module, adjust it to the cleaning position and angle, and make the coating head contact the ring channel; Step S7, Start Cleaning: Start the metering and dispensing module, pump cleaning solution into the slip ring channel, and rotate the slip ring to perform in-situ cleaning. Step S8: The information acquisition and processing module collects multi-physics field signals in real time throughout the entire process from steps S1 to S7, and performs working condition identification, remaining service life prediction, and autonomous optimization of process parameters.

[0014] Preferably, the information acquisition and processing module includes: The data fusion preprocessing unit is used to synchronize and align multi-physics field signals and extract features to form a multi-channel time-series data matrix. Physically constrained neural network processors include: The working condition recognition network is used to output the current working condition category of the slip ring-brush friction pair; The remaining useful life prediction network is used to predict the remaining effective life of the slip ring ring coating. Its loss function incorporates constraints from tribological laws. A process parameter autonomous optimization decision network, based on deep reinforcement learning, is used to output recommended combinations of process parameters. A comprehensive health status assessment index generation unit is used to calculate the health index of the slip ring ring coating based on the working condition category and the remaining effective life prediction value. The self-learning and model update unit is used to incrementally learn the neural network processor using newly added test data.

[0015] Preferably, in step S5, the normal contact force and tangential friction force are monitored by the three-dimensional force sensor in the sensing and measurement module, the vibration signal is measured by the accelerometer, the temperature rise and stress wave signal are monitored by the temperature sensor and the acoustic emission sensor, and the dynamic contact resistance and electrical noise signal are simultaneously measured by the four-wire Kelvin connection method.

[0016] Preferably, in steps S3, S5 and S7, the delivery dosage and rate of lubricant or cleaning fluid, the rotational speed of the rotating shaft, the contact pressure and angle, and the magnitude of the test current can all be independently programmed and controlled.

[0017] Therefore, the present invention employs the above-mentioned integrated machine for multi-condition in-situ testing and intelligent coating of slip rings and its usage method, and the beneficial technical effects are as follows: (1) By using servo motor flipping and quick-change fixture design, the coating, testing and cleaning processes are integrated into the same station for continuous completion, eliminating workpiece transfer errors; the modular lower specimen fixture (three-jaw chuck with expansion mandrel / elastic bushing) and configurable upper specimen jaws are compatible with various types and sizes of slip rings and brushes, enabling fast, non-destructive and high-throughput testing.

[0018] (2) A multi-dimensional sensing system integrating force, heat, vibration, acoustic emission and electrical signals, simultaneously capturing tribological, electrical contact, acoustic and thermal signals; a built-in physical information constrained neural network, which can identify six working conditions in real time: running-in, stable wear, insufficient lubrication, particulate contamination, arc erosion and failure warning, and generate a health index H, transforming high-dimensional data into intuitive indicators, providing objective quantitative basis for failure mechanism research.

[0019] (3) The Arcard wear law and Holm electrical contact model are embedded into the neural network loss function to achieve physically consistent remaining lifetime prediction under small sample conditions; based on deep reinforcement learning, parameters such as rotation speed, coating flow rate, and test current are dynamically adjusted, shifting from passive execution to active optimization, thus extending the effective life of the coating; the incremental learning mechanism enables the model to continuously evolve over a long period of time and adapt to specific test scenarios. At the same time, the quantitative dispensing module works in conjunction with the multi-axis motion closed loop to precisely program and control process parameters such as flow rate, speed, pressure, and time, ensuring the repeatability of the lubrication and cleaning process. Attached Figure Description

[0020] Figure 1 This is a three-dimensional structural diagram of the integrated machine of the present invention performing a coating operation on a slip ring with an inner hole for positioning; Figure 2 This is a three-dimensional structural diagram of the integrated machine of the present invention performing a coating operation on a slip ring positioned by an outer circle; Figure 3 for Figure 1 A schematic diagram of the specific structure of the application module shown in the figure; Figure 4 for Figure 2 A schematic diagram of the specific structure of the application module shown in the figure; Figure 5 for Figure 3 The diagram shows the structure of the slip ring clamping module (quick-change fixture for lower specimen), where (a) is the whole and (b) is a part; Figure 6 for Figure 4 The diagram shows the structure of the slip ring clamping module, where (a) is the whole and (b) is a part; Figure 7 This is a three-dimensional structural diagram of the integrated machine of the present invention for conducting current-carrying friction tests on slip rings positioned by their inner holes; Figure 8This is a three-dimensional structural diagram of the integrated machine of the present invention for conducting current-carrying friction tests on a slip ring positioned on an outer circle; Figure 9 The diagram shows the structure of the quick-change fixture for the upper specimen, which clamps the flexible coating head, brush bristles, brush bundle, brush block and metal sheet respectively. (a) clamps the flexible coating head, (b) clamps the brush bristles, (c) clamps the brush bundle, (d) clamps the brush block and (e) clamps the metal sheet. Figure 10 This is a schematic diagram of the four-wire Kelvin connection circuit used in the current-carrying friction test of this invention. Figure 11 This is a system block diagram of the information acquisition and processing module of the present invention, wherein (a) is the overall block diagram and (b) is the flowchart of the information processing and analysis part; Figure 12 This is a flowchart illustrating the usage method of the all-in-one machine of the present invention.

[0021] Figure Labels 1-1 Vertical support; 1-2 Horizontal base; 2-1 Z-axis linear guide; 2-2 Y-axis linear guide; 2-3 X-axis linear guide; 2-4 Rotary C-axis platform; 3-1 Liquid storage chamber; 3-2 Micro-injection pump; 3-3 Upper specimen connecting plate; 4-1 Miniature digital servo motor; 4-2 Tilting block; 5 Upper specimen quick-change fixture; 6. Coating module; 6-1 Coating head holder; 6-2 Coating head fixing plate; 6-3 Flexible coating head; 7. Quick-change fixture for specimen loading; 7-1. Precision three-jaw chuck; 7-2. Expansion mandrel; 7-3. Elastic bushing; 8-1. Brush filaments; 8-2. Brush block; 8-3. Metal sheet; 8-4. Brush bundle; 9-1. Inner hole positioning slip ring; 9-2. Outer circle positioning slip ring; 10-1. Three-dimensional force sensor; 10-2. Accelerometer; 10-3. Temperature sensor; 10-4. Acoustic emission sensor; 10-5. Auxiliary slip ring; 10-6. Circuit test bracket. Detailed Implementation

[0022] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0023] Unless otherwise defined, the technical or scientific terms used in this invention shall have the ordinary meaning as understood by one of ordinary skill in the art to which this invention pertains.

[0024] Example 1 See Figures 1 to 12This embodiment provides an integrated machine for multi-condition in-situ testing and intelligent coating of slip rings, along with its usage method. This equipment achieves full automation of the slip ring testing process through deep integration of mechanical, electrical, and software components. The quick-change fixture 7 for the lower specimen, using a precision three-jaw chuck 7-1 in conjunction with a replaceable expansion mandrel 7-2 and elastic bushing 7-3, enables universal clamping of different types of slip rings. By changing the expansion mandrel 7-2 to different specifications, it can stably clamp through-hole slip rings with inner diameters ranging from 5mm to 65mm; by changing the elastic bushing 7-3 to different specifications, it can stably clamp solid or cap-type slip rings with outer diameters ranging from 12mm to 65mm. The slip ring height does not exceed 120mm. For multi-track slip rings, the power drive module controls the upper specimen to move along the slip ring axis via precision linear motion, sequentially coating or testing each track. The number and spacing of the tracks are automatically adapted by the control program and do not constitute a limitation on the equipment structure.

[0025] Overall structure: like Figure 1 , Figure 2 As shown, the integrated machine includes a frame, a power drive module, a quantitative dispensing module, an upper specimen servo adjustment module, an upper specimen quick-change fixture 5, a coating module 6, a lower specimen quick-change fixture 7, a sensing and measurement module, and an information acquisition and processing module.

[0026] The frame consists of a vertical support 1-1 and a horizontal base 1-2, providing a stable installation foundation for each module.

[0027] The power drive module is mounted on the frame and is responsible for all precision movements. Specifically (see...) Figure 1 , Figure 2 The Z-axis linear guide 2-1 is installed in the groove of the support; the Y-axis linear guide 2-2 is installed on the Z-axis displacement platform; the X-axis linear guide 2-3 is installed on the base 1-2; and the rotary C-axis platform 2-4 is installed on the X-axis displacement platform. This design of three linear axes and one rotary axis ensures that the paint head or brush can be positioned arbitrarily in space and aligned with the ring track at any angle on the slip ring.

[0028] The linear axis of the power drive module is driven by a servo motor in conjunction with a ball screw and precision linear guides, while the rotary axis is driven by a direct-drive torque motor or a servo motor in conjunction with a harmonic reducer to ensure sufficient thrust, rigidity, and high-precision position and speed control. Depending on cost and accuracy requirements, the linear axis can also be driven by a linear motor or a stepper motor in conjunction with a ball screw. The power drive module provides linear motion in the X, Y, and Z directions and rotational motion around the C-axis. To achieve precise alignment and stable contact between the coating head or brush and the slip ring track, the positioning accuracy of the linear axis is better than ±10μm, and the repeatability is better than ±5μm; the positioning accuracy of the rotary axis is better than ±0.2°, and the repeatability is better than ±0.1°. This accuracy range ensures that during the coating and testing process of multi-track slip rings, positional deviations will not significantly affect the contact pressure and test results, and can be adapted and adjusted according to specific application requirements.

[0029] The quantitative dispensing module is mounted on the Y-axis displacement stage via the upper specimen connecting plate 3-3. For example... Figure 3 , Figure 4 As shown, it includes a liquid storage chamber 3-1, a micro-injection pump 3-2, and pipeline valves. The micro-injection pump 3-2 can be programmed to precisely control the dosage and rate of liquid pumped from the liquid storage chamber 3-1, and the liquid is delivered to the application module 6 through channels in the connecting plate.

[0030] The miniature digital servo 4-1 of the servo adjustment module for the upper specimen is also mounted on the Y-axis displacement stage. (Refer to...) Figure 1 , Figure 2 The servo motor drives the flip block 4-2 to precisely rotate from 0° to 180°. A three-dimensional force sensor 10-1 is mounted on the surface of the flip block 4-2 to provide real-time feedback of the contact force. By flipping the flip block 4-2, in-situ rapid switching and contact angle adjustment between the three workstations of coating, testing, and cleaning can be achieved.

[0031] The quick-change fixture 5 for the upper specimen is mounted on the three-dimensional force sensor 10-1 and moves together with the flipping block 4-2. For example... Figure 9 As shown, the upper specimen quick-change clamp 5 uses a motor-driven precision clamp. The clamping force is adjusted by controlling the motor output torque, and the clamping force range can be set from 0.5N to 50N. It can also be programmed to adjust according to the material and size requirements of different brushes or coating modules 6. The clamp jaws can be changed manually or automatically (e.g., pneumatic or electromagnetic). The clamp jaws are made of materials that combine insulation and wear resistance, such as ceramics, polyetheretherketone, polytetrafluoroethylene, or metals with an insulating surface, to prevent short circuits during current-carrying tests and ensure long-term stability. By changing different clamp jaws (standard flat jaws for brush blocks or metal sheets, and V-groove jaws for brush filaments, brush bundles, or coating heads), various types of brushes and coating modules 6 can be quickly, stably, and without damage.

[0032] The coating module 6 is held by the upper specimen quick-change fixture 5. For example... Figure 3 , Figure 4 As shown, it includes a coating head holder 6-1, a coating head fixing plate 6-2, and a flexible coating head 6-3. The flexible coating head 6-3 is made of a porous elastic material or fiber material with resilience and liquid adsorption and release capabilities, including one or more of silicone, polyurethane foam, non-woven fabric, brush, and microporous foam plastic; to withstand the chemical corrosion of different types of lubricants and cleaning fluids, the flexible coating head material is selected from oil- and solvent-resistant nitrile rubber foam, fluororubber foam, or polytetrafluoroethylene fiber felt. The coating head holder 6-1 is held by the upper specimen quick-change clamp 5, the coating head fixing plate 6-2 is installed on the coating head holder 6-1, and the flexible coating head 6-3 is fixed on the coating head fixing plate 6-2. The coating head holder 6-1 and the coating head fixing plate 6-2 have flow channels inside, which are connected to the flexible coating head 6-3, for receiving the liquid delivered by the quantitative dispensing module and achieving uniform coating on the slip ring track.

[0033] The quick-change fixture 7 for the lower specimen is mounted on the rotating C-axis platform 2-4 of the power drive module to hold the slip ring rotor 9 to be tested. Figure 5 , Figure 6 As shown, it adopts a modular design: the core is a precision three-jaw chuck 7-1, equipped with a quick-change expansion spindle 7-2 and a flexible bushing 7-3. Specifically, for Figure 1 , Figure 5 , Figure 7 The inner hole positioning slip ring 9-1 shown (such as a column slip ring) is fitted with an expansion mandrel 7-2 of the corresponding size. When the three-jaw chuck tightens, the expansion mandrel expands outward, thus gripping the inner hole of the slip ring. For Figure 2 , Figure 6 , Figure 8 The outer circular positioning slip ring 9-2 shown (such as a disc slip ring) is fitted with an elastic bushing 7-3 of the corresponding inner diameter. The slip ring is placed within it, and when the three-jaw chuck tightens, the elastic bushing contracts evenly to grip the outer circle of the slip ring. All clamping components, including the jaws of the three-jaw chuck 7-1, the expansion mandrel 7-2, and the elastic bushing 7-3, are made of materials with good insulation properties and sufficient mechanical strength to prevent leakage or short circuits during current-carrying tests. The insulating material is selected from one or more of polyetheretherketone, ceramic, polyoxymethylene, nylon, polytetrafluoroethylene, or polyimide, depending on the clamping force and wear resistance requirements, or equivalent replacements are used. In this embodiment, nylon material is used.

[0034] The sensing and measurement modules are distributed and integrated within the device to acquire force, heat, vibration, acoustic emission, and electrical signals in real time during the testing process. Specifically, they include (see [link to documentation]). Figure 1 , Figure 2 , Figure 7 , Figure 8 , Figure 10 ): The three-dimensional force sensor 10-1 is mounted on the flip block 4-2 and is used to monitor the normal contact force and tangential friction force in real time.

[0035] Accelerometer 10-2, mounted on flip block 4-2, is used to measure mechanical vibration signals during the test process.

[0036] Temperature sensor 10-3 and acoustic emission sensor 10-4 are installed on the upper specimen quick-change fixture 5 near the brush, and are used to monitor the temperature rise at the contact point and capture stress wave signals generated by micro-deformation and cracking of the material, respectively.

[0037] The auxiliary slip ring 10-5, whose rotor is connected to the quick-change fixture 7 for the lower specimen, is used to extract dynamic electrical signals.

[0038] Circuit test bracket 10-6 is used to fix the test circuit.

[0039] A sophisticated electrical signal acquisition chain, employing a four-wire Kelvin connection method, includes a precision constant current source, a high-precision DC voltmeter, a high-pass filter, a low-noise amplifier, and a high-speed data acquisition card, used for synchronously measuring dynamic contact resistance and extracting electrical noise signals.

[0040] The information acquisition and processing module connects to all sensors (see...) Figure 11 This module is used for synchronous acquisition, intelligent processing, and autonomous decision analysis of multi-physics field signals. It includes: a data fusion preprocessing unit, a neural network processor with physical information constraints, a comprehensive health status assessment index generation unit, and a self-learning and model update unit.

[0041] (a) Data fusion preprocessing unit.

[0042] Multiple signals, including three-dimensional force, temperature, vibration, acoustic emission, and dynamic contact resistance, are synchronously acquired via a multi-channel synchronous data acquisition card at a sampling frequency of 1kHz to 1MHz. The acquired raw data undergoes preprocessing including trend removal, noise reduction (using wavelet transform or bandpass filter), and outlier removal. Then, synchronous alignment is performed to form a data structure with dimensions of [missing information]. A multi-channel time-series data matrix, in which Number of sensor channels (in this embodiment) These correspond to normal force, tangential friction, vibration, temperature, acoustic emission, and dynamic contact resistance, respectively. The sampling time is specified. The preprocessed data matrix is ​​then fed into a physically constrained neural network processor for in-depth analysis.

[0043] (ii) Neural network processors constrained by physical information.

[0044] The processor adopts a three-level intelligent decision-making architecture of "operating condition identification - remaining life prediction - process parameter optimization", as detailed below: (1) Operating condition identification network.

[0045] The working condition recognition network consists of a convolutional neural network (CNN) and a long short-term memory neural network (LSTM) connected in series. The CNN part comprises three convolutional layers and two pooling layers stacked together: the first convolutional layer uses 32... Convolutional kernels extract spatial features within a local temporal window; the second convolutional layer uses 64 kernels. The third convolutional layer uses 128 convolutional kernels. Convolutional kernels; each convolutional layer is followed by a batch normalization layer and a ReLU activation function; pooling layers use... Max pooling reduces feature dimensionality and enhances translation invariance. Spatial features extracted by the CNN are flattened and input into the LSTM layer. The LSTM layer consists of two LSTM layers with 128 and 64 neurons in the hidden layers, respectively, used to learn the evolution of operating conditions over a long time scale. The network output layer uses the Softmax activation function to output a probability vector of the current operating condition category of the slip ring-brush friction pair. The operating condition categories include: break-in state, stable wear state, insufficient lubrication state, particulate contamination state, arc erosion state, and failure warning state. The recognition results are displayed in real time on the state identification graph of the human-computer interaction interface, and the state evolution trajectory is shown in timeline form.

[0046] (2) Remaining useful life prediction network.

[0047] The remaining service life prediction network is a deep neural network that incorporates prior tribological physics knowledge. Its input consists of deep feature vectors extracted from the intermediate layer of the operating condition recognition network and the current cumulative test duration. The output is the predicted remaining effective service life of the slip ring track coating under the current operating conditions (expressed in revolutions or hours). This network adopts a multilayer perceptron structure, containing three hidden layers (with 256, 128, and 64 neurons respectively), each followed by a Dropout layer (dropout rate of 0.3) to prevent overfitting.

[0048] Embedding physical constraints: Arcard's law of wear—wear volume With normal load Sliding distance Proportional to the material hardness Inversely proportional, that is And the Holm electrical contact resistance model: contact resistance ,in Resistivity The analytical relationship between the contact spot radius and the contact spot radius is used as a constraint term and embedded in the network's loss function. Let the network predict the wear depth variation sequence as follows: The theoretical wear depth variation sequence derived from Archard's law is as follows: Then the physical constraint deviation Total loss function The form is: ; in, To determine the mean squared error between the predicted remaining lifetime and the actual remaining lifetime. The smoothness constraint for the output sequence is the sum of squares of the differences between the predicted values ​​at adjacent time steps. and The balance coefficient (taken in this embodiment) =0.3, =0.1). This physical information constraint design enables the network to maintain good prediction accuracy and physical consistency even with a small number of training samples, avoiding prediction results that violate basic physical laws that may occur in purely data-driven models.

[0049] (3) Autonomous optimization decision network for process parameters.

[0050] The process parameter autonomous optimization decision network is a decision module based on deep reinforcement learning, employing a deep Q-network architecture. Its state space includes: the current state category code (6-dimensional one-hot vector) output by the operating condition recognition network, the lifespan decay rate (lifespan loss per unit time, ranging from 0 to 1) output by the remaining service life prediction network, and the current health index output by the health status comprehensive evaluation index generation unit (see below for details). The motion space includes: the C-axis speed settings of the power drive module (low speed 10–50 rpm, medium speed 50–200 rpm, high speed 200–500 rpm), the coating flow rate settings of the quantitative dispensing module (0.1 mL / min, 0.5 mL / min, 1.0 mL / min, 5.0 mL / min), and the test current settings (0 A, 0.5 A, 1 A, 2 A, 5 A), forming a total of... A combination of discrete actions. Reward function. The design principle is: ; in, The coefficient of variation (standard deviation / mean) of the friction coefficient over the most recent 60 seconds. This is the coefficient of variation of the contact resistance over the last 60 seconds. To predict the amount of remaining lifetime extension relative to the original strategy after taking the current action, For the rated total lifespan, , , The weighting coefficient (taken in this embodiment) =0.3, =0.4, =0.3). An additional negative penalty is applied when arc erosion or failure warning conditions are detected. =-10. The decision network uses an ε-greedy strategy (initial ε=0.9, gradually decaying to 0.05) to select actions, outputting the recommended combination of process parameters that optimizes friction and wear performance, and can select between two working modes: "recommended only" (executed after confirmation by the operator) or "automatic execution".

[0051] (III) Health Status Comprehensive Assessment Indicator Generation Unit.

[0052] This unit is used to identify the state category probability vector output by the operating condition identification network and the remaining lifetime prediction value output by the remaining lifetime prediction network, and to calculate the health index H of the slip ring annular coating in real time. Its expression is: ; in, The rating values ​​for the current operating condition are as follows: break-in condition: 0.2-0.4 (median 0.3); stable wear condition: 0.7-1.0 (median 0.85); insufficient lubrication condition: 0.3-0.5 (median 0.4); particulate contamination condition: 0.2-0.4 (median 0.3); arc erosion condition: 0.1-0.3 (median 0.2); failure warning condition: 0-0.1 (median 0.05). This is the predicted remaining useful life (in revolutions or hours). and The weighting coefficient (taken in this embodiment) =0.5, =0.5). Health Index The value ranges from 0 to 1 and is displayed in real-time on the instrument panel of the device's human-machine interface. When When the value is below a preset threshold (set to 0.3 in this embodiment), the system automatically triggers an audible and visual alarm and sends an intervention request to the process parameter autonomous optimization decision network to suggest or automatically perform parameter adjustments.

[0053] (iv) Self-learning and model update unit.

[0054] This unit is used to incrementally learn the neural network processor constrained by the physical information by using multi-physics field data collected from each complete test, historical operation records of operators (including whether recommended parameters were adopted and manually adjusted parameter values), and the final test results (such as the actual life and failure mode when the slip ring fails) as new samples. Specifically, an experience replay mechanism is adopted: key samples from each test are stored in a recurrent experience pool. When the accumulated new samples reach a set number (e.g., 100), an offline retraining is triggered, and the network parameters are updated using a mini-batch stochastic gradient descent algorithm. Through continuous self-learning, the model parameters of the operating condition identification network, the remaining service life prediction network, and the process parameter autonomous optimization decision network are continuously optimized, enabling the system to gradually adapt to the user's specific test scenarios and lubrication system during long-term use.

[0055] Depending on specific testing requirements, the aforementioned intelligent processing unit can be selectively enabled or disabled. When the intelligent processing function is disabled, the system reverts to the traditional signal acquisition and conventional time-frequency analysis mode to ensure the system's flexibility and compatibility in different application scenarios.

[0056] Test circuit: Current-carrying friction test adopts Figure 10 The four-wire Kelvin connection method shown is used to eliminate the influence of lead resistance on the measurement of minute contact resistance. Figure 10 middle, : The positive terminal of the constant current source current output (Force High) is connected to the current lead-in line of the brush; The constant current source returns to the negative terminal (Force Low), which is connected to the current lead of the auxiliary slip ring rotor circuit; : The positive terminal for voltage measurement (Sense High) is connected to the voltage sensing line on the side of the brush near the contact point; The negative terminal (Sense Low) of the voltage measurement is connected to the voltage sensing line near the slip ring contact point (or the auxiliary slip ring lead). The above... Forming a current loop, They form a voltage measurement circuit, and the two circuits converge at the contact point, which conforms to the basic principle of four-wire Kelvin measurement.

[0057] Current loop: The constant current source outputs a stable DC test current. The path is: constant current source positive terminal → Force+ line → brush → slip ring contact point → rotor of auxiliary slip ring 10-5 → protection resistor → Force- line → constant current source negative terminal.

[0058] Voltage sensing circuit: The Sense+ and Sense- lines of the high-precision DC voltmeter are directly connected in parallel across the two ends of the contact point between the brush and the slip ring, measuring the current. Flow through contact resistance and auxiliary slip ring resistor Total pressure drop .

[0059] Electrical noise extraction circuit: The signal is drawn in parallel from the voltage sensing circuit and sequentially passed through a high-pass filter (to remove the DC component), a low-noise amplifier (to amplify the weak AC signal), and a high-speed data acquisition card to capture the pure contact electrical noise time-domain signal. .

[0060] During data processing, dynamic contact resistance is determined using the formula... The calculation shows that, among which The inherent resistance of the auxiliary slip ring is obtained through separate calibration. It is then used for time-domain (peak-to-peak value, RMS value) and frequency-domain (FFT power spectrum) analysis to evaluate the stability of electrical contacts.

[0061] The auxiliary slip ring 10-5 should have stable and repeatable dynamic contact resistance characteristics, with its own resistance fluctuation significantly less than the change in the contact resistance of the slip ring under test (the fluctuation amplitude should not exceed 1 / 10 of the change in the measured resistance); within the test speed range (0~100rpm), the dynamic contact resistance fluctuation amplitude should be better than ±2% or the absolute value should be better than ±0.5mΩ, and the coefficient of variation (standard deviation / average value) of the contact resistance after continuous rotation for more than 1 hour should be less than 2%; a low-noise precision conductive slip ring (multi-contact wire brush type or mercury type) should be selected and periodically calibrated to meet the above requirements, and its fixed resistance component should be subtracted through calibration during measurement data processing.

[0062] Example 2 Combination Figure 12 The flowchart shows the steps involved in a complete test: Step S1: Install the slip ring rotor: Depending on the type of slip ring 9 (inner hole positioning or outer circle positioning), select the expansion mandrel 7-2 or elastic bushing 7-3 to install on the precision three-jaw chuck 7-1, and then fix the slip ring rotor in place.

[0063] Step S2: Prepare for coating: Install the coating module 6 on the upper specimen quick-change fixture 5. Control the upper specimen servo adjustment module to flip to the coating station and adjust the coating angle. Control the three linear axes of the power drive module to make the flexible coating head 6-3 contact the slip ring track, and control the contact force to the set value through feedback from the three-dimensional force sensor 10-1. As an example, the contact force during coating can be set to 0.5N to 10N, and the specific value depends on the hardness of the flexible coating head and the width of the slip ring track. In this embodiment, it is set to 1N.

[0064] Step S3, Initiate homogenization application: Set the delivery rate, time, and delay parameters of the micro-injection pump 3-2, and start the program. Lubricant is metered and delivered to the applicator head. Simultaneously, the C-axis is started at a set speed to rotate the slip ring. The linear axis is controlled as needed to move the applicator head along the annular path until a uniform lubricating film is formed on the entire annular path. For example, the delivery rate can be set to 0.1 mL / min to 10 mL / min, the C-axis speed can be set to 10 rpm to 100 rpm, and the applicator head's movement speed along the annular path can be set to 1 mm / s to 10 mm / s.

[0065] Step S4: Preparation for Current-Carrying Friction Test: Remove the coating module 6 and install the selected brush (brush filaments 8-1, brush bundle 8-4, brush block 8-2, or metal sheet 8-3) on the upper specimen quick-change fixture 5. Control the servo motor to rotate to the test position and adjust the brush contact angle according to the slip ring track type (0°~90°: 0° for vertical positive pressure of column slip ring track, 45°~90° for disc slip ring track); control the movement of the three linear axes to make the brush contact the track and achieve the set brush pressure (0.2N~10N, determined according to the brush filament material, slip ring track material, and test current). Connect and check the current-carrying test circuit, and align each sensor probe with the brush-ring contact area. The above-mentioned contact force, delivery rate, rotation speed, brush pressure, angle, and other parameters are all exemplary values ​​for programmable control and can be reasonably adjusted according to the specific size, material, and test target of the slip ring under test. The core of this invention is to provide an automated test process for programmable control of the above parameters, rather than being limited to a specific set of parameters.

[0066] Step S5: Start the current-carrying friction and wear test: Set the test current, C-axis speed (simulating the slip ring's operating speed), and test cycle. Start the equipment and simultaneously begin multi-sensor data acquisition. After the test, the C-axis stops rotating, and the linear axis drives the brush to leave the ring track. For example, the test current can be set to 0.1A to 10A (DC or AC), the C-axis speed can be set to 10rpm to 500rpm, and the test cycle can be set to 30 minutes to 100 hours (or a specific number of revolutions, such as 10, depending on the lubricant life requirements). 5 ~10 7 change).

[0067] Step S6: Prepare for cleaning: Remove the brush and reinstall the application module 6. Adjust it to the cleaning position and angle so that the application head contacts the ring channel.

[0068] Step S7, Initiate Cleaning: The micro-injection pump delivers cleaning solution to the applicator head. The C-axis drives the slip ring to rotate, performing in-situ, automatic cleaning of the ring channel. Multiple cleaning cycles can be programmed (e.g., cleaning solution → cleaning solution → deionized water → deionized water). The cleaning solution can be anhydrous ethanol, isopropanol, acetone, deionized water, or a dedicated slip ring cleaner. The selected cleaning solution should be chemically compatible with the flexible applicator head and fixture materials. For example, acetone can cause swelling of ordinary silicone and other rubber applicators; in this case, an acetone-resistant applicator head material (such as fluororubber or polytetrafluoroethylene) should be selected, or a mild cleaning agent such as anhydrous ethanol should be used. Choose a compatible cleaning solution based on the actual materials.

[0069] Step S8, Data Acquisition and Intelligent Analysis: The information acquisition and processing module synchronously records and processes data from various sensors throughout the process, and finally outputs a multi-dimensional analysis report including friction coefficient curve, dynamic contact resistance curve, electrical noise time / frequency domain graph, vibration spectrum, temperature curve, and acoustic emission signal, which is used to comprehensively evaluate the lubricant performance and slip ring condition.

[0070] In steps S3, S5, and S7 above, the delivery dosage and rate of lubricant or cleaning fluid (exemplarily, delivery rate 0.01 mL / min to 20 mL / min, single dose 0.1 mL to 10 mL), C-axis rotation speed (10 rpm to 1000 rpm, depending on coating or testing requirements), contact pressure and angle between the brush and the ring (contact pressure 0.1 N to 50 N, angle 0° to 90°), and magnitude of the test current (DC or AC, 0 to 50 A) can all be independently programmed and controlled.

[0071] Based on the device structure described in Example 1, this section focuses on the practical application of the physical information constraint neural network processor in the information acquisition and processing module. A certain type of through-hole slip ring (inner diameter 25mm, ring material is gold alloy, rated total lifespan 500 hours or 5×10) is used as an example. 6 For example, a current-carrying friction test was conducted using a commercially available brand of conductive grease.

[0072] 1. Initial stage (0-50 hours) - Identification of break-in status.

[0073] After the test starts, the multi-physics field signals collected by the sensing and measurement module are preprocessed and then sent to the operating condition recognition network. The network outputs the following state probability vectors: break-in state 92%, stable wear state 5%, and other states 3%. The system determines that it is currently in the break-in state. Health Index The dashboard displays a yellow warning zone. Based on the current state (break-in period, health level 0.60), the process parameter autonomous optimization decision network recommends the following parameters: low speed (30 rpm), low flow rate (0.1 mL / min), and test current 1 A. The operator adopts this recommendation, and the equipment automatically adjusts the parameters.

[0074] 2. Stable phase (50-350 hours) — Stable wear state.

[0075] After 80 hours of operation, the probability of the working condition identification network outputting a stable wear state gradually increases to over 85%. (Health Index) The system enters a safe, green zone. Based on the stable wear condition, the autonomous optimization decision network for process parameters automatically increases the C-axis speed to medium (150 rpm), maintains the coating flow rate at 0.1 mL / min (only making minor replenishments), and increases the test current to 2A to accelerate the testing process while maintaining contact resistance stability. During this stage, the coefficient of variation of the friction factor stabilizes below 0.05, the coefficient of variation of the contact resistance stabilizes below 0.08, and the health index slowly and linearly decreases.

[0076] 3. Insufficient lubrication warning stage (350-400 hours).

[0077] After 370 hours of operation, the probability of the condition recognition network detecting insufficient lubrication increased to 78%. The identification criteria were: increased high-frequency fluctuations in the tangential friction force signal (the energy in the 2-10Hz band extracted by Fourier transform increased threefold compared to the stable stage), increased short-term fluctuation amplitude of dynamic contact resistance, and accelerated temperature rise rate at the contact point detected by the temperature sensor. The system issued an "insufficient lubrication" status warning, and the health index... The temperature was approaching the alarm threshold of 0.30. The process parameter autonomous optimization decision network immediately recommended intervention: increase the coating flow rate from 0.1 mL / min to 1.0 mL / min, maintain this for 3 minutes, then restore the replenishment flow rate to 0.5 mL / min, while temporarily reducing the rotation speed to 80 rpm to mitigate wear. The operator adopted the recommendation and performed the reapplied grease.

[0078] 4. Recovery phase after touch-up application (400-480 hours).

[0079] Approximately 10 minutes after the recoating, the operating condition identification network output returned to a stable wear state (probability 82%), and the health index rebounded to 0.45. The system then continued to operate for 480 hours with the optimized parameter combination (coating flow rate 0.5 mL / min, rotation speed 120 rpm, test current 2 A).

[0080] 5. Failure warning stage (480-500 hours).

[0081] After running for 480 hours, the remaining lifespan prediction network predicted a remaining lifespan of 22 hours (actual remaining lifespan was 20 hours, prediction error 10%), and the health index... If the current drops below 0.30, an audible and visual alarm is triggered. The system enters a failure warning state and automatically stops the current-carrying test, prompting the operator to conduct an inspection or terminate the test.

[0082] 6. Self-learning and model update.

[0083] After this test, the self-learning and model update unit used the multiphysics data collected throughout the process, the records of operators adopting recommended parameters, and the final actual failure life (502 hours) as new samples to incrementally learn the neural network processor. Through multiple tests, the system's remaining life prediction accuracy gradually improved, and the effectiveness of process parameter recommendations was continuously optimized.

[0084] This embodiment demonstrates that the information acquisition and processing module can realize a complete intelligent closed loop from "state perception" to "intelligent decision-making" and then to "continuous evolution", which significantly improves the intelligence level and testing efficiency of slip ring multi-condition in-situ testing.

[0085] It is understood that the test current, rotation speed, cycle, type of cleaning fluid, delivery rate, contact pressure, angle, and other parameters listed in the above steps are all exemplary values. Those skilled in the art can reasonably adjust them or select other equivalent cleaning fluids according to the specific specifications of the slip ring under test, the type of lubricant, and the test objectives. The core of the protection claimed by this invention lies in the equipment structure and automated testing process, rather than being limited to a specific set of parameters or a specific cleaning fluid.

[0086] Comparative analysis with existing technologies.

[0087] To further illustrate the substantial progress of the technical solution of this invention compared with the prior art CN118060136A (a quantitative coating and polishing device for the surface of a metal slip ring rotor and its coating process) and CN118310911A (a multifunctional vacuum friction and wear test device for conductive slip ring and its usage method), a quantitative comparison is now made from three dimensions: core functions and intelligence level, work efficiency and time indicators, and technical accuracy and output results, as shown in Tables 1 to 3.

[0088] Table 1 Comparison of Core Functions and Level of Intelligence

[0089] Table 2 Comparison of Work Efficiency and Time Indicators

[0090] Table 3 Comparison of Technical Accuracy and Output Results

[0091] As can be seen from the table above, the present invention is significantly superior to the prior art in terms of functional integration, intelligence level, testing efficiency, measurement accuracy, and richness of output information, achieving unexpected technical effects.

[0092] It is worth noting that all contents not described in detail in this invention are existing technologies and are well known to those skilled in the art.

[0093] Therefore, the present invention adopts the above-mentioned slip ring multi-condition in-situ testing and intelligent coating integrated machine and its usage method, realizing the full-process in-situ automated operation of slip ring coating, testing and cleaning. It has excellent versatility and multi-physical field fusion monitoring capabilities, and can perform precise and controllable programming control of lubrication and cleaning processes. It effectively solves the problems of discrete testing process, uncontrollable coating quality, poor fixture versatility, single dimension of monitoring data and low degree of automation in the prior art.

[0094] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for using an integrated machine for multi-condition in-situ testing and intelligent coating of slip rings, characterized in that, The integrated machine for multi-condition in-situ testing and intelligent coating of slip rings includes: frame; The power drive module is mounted on the frame; The quantitative dispensing module is installed on the power drive module; The upper test piece servo adjustment module is installed on the power drive module; The upper specimen quick-change fixture is installed on the upper specimen servo adjustment module and is used to replaceably hold the coating module or brush. The coating module is used to coat the slip ring track with liquid; The specimen quick-change fixture is installed on the power drive module to hold the slip ring rotor; The sensing and measurement module is used to acquire multi-physics field signals in real time during the testing process; The information acquisition and processing module is connected to the sensing and measurement module and is used for synchronous acquisition, processing and analysis of multi-physics field signals; The usage method includes the following steps: Step S1: Install the slip ring rotor onto the lower specimen quick-change fixture; Step S2: Install the coating module on the upper specimen quick-change fixture, adjust the upper specimen servo adjustment module to the coating position and angle, and control the power drive module to make the coating head of the coating module contact the slip ring track and reach the preset pressure. Step S3: Start the rotating shaft of the quantitative drip coating module and the power drive module to quantitatively apply lubricant to the slip ring track; Step S4: Replace the coating module with a brush, adjust the upper specimen servo adjustment module to the test position and angle, control the power drive module to make the brush contact the slip ring track and achieve the preset brush pressure, and connect the current carrying test circuit. Step S5: Apply the set current to start the rotating shaft of the power drive module to drive the slip ring to rotate, and at the same time start the sensing and measurement module to collect multi-physics field data synchronously. Step S6: Replace the brush with a coating module, adjust it to the cleaning position and angle, and make the coating head contact the ring channel; Step S7: Start the quantitative drip coating module, pump cleaning solution into the slip ring channel, and rotate the slip ring to perform in-situ cleaning. Step S8: The information acquisition and processing module collects multi-physics field signals in real time throughout the entire process from steps S1 to S7, and performs working condition identification, remaining service life prediction, and autonomous optimization of process parameters.

2. The method according to claim 1, characterized in that, The information collection and processing module includes: The data fusion preprocessing unit is used to synchronize and align multi-physics field signals and extract features to form a multi-channel time-series data matrix. Physically constrained neural network processors include: The working condition recognition network is used to output the current working condition category of the slip ring-brush friction pair; The remaining useful life prediction network is used to predict the remaining effective life of the slip ring ring coating. Its loss function incorporates constraints from tribological laws. A process parameter autonomous optimization decision network, based on deep reinforcement learning, is used to output recommended combinations of process parameters. The health status comprehensive assessment index generation unit is used to calculate the health index of the slip ring ring coating based on the working condition category and the predicted value of the remaining effective life. The self-learning and model update unit is used to incrementally learn the neural network processor using newly added test data.

3. The method according to claim 2, characterized in that, In step S5, the normal contact force and tangential friction force are monitored by the three-dimensional force sensor in the sensing and measurement module, the vibration signal is measured by the accelerometer, the temperature rise and stress wave signal are monitored by the temperature sensor and the acoustic emission sensor, and the dynamic contact resistance and electrical noise signal are simultaneously measured by the four-wire Kelvin connection method.

4. The method according to claim 1, characterized in that, The power drive module is used to provide linear motion along the X, Y, and Z directions and rotational motion about the C-axis.

5. The method according to claim 1, characterized in that, The quantitative dispensing module includes an upper specimen connection plate, a liquid storage chamber, a micro-injection pump, and pipeline valves. The quantitative dispensing module is mounted on the linear motion component of the power drive module via the upper specimen connection plate. The input end of the micro-injection pump is connected to the liquid storage chamber, and the output end is connected to the dispensing module via pipeline.

6. The method according to claim 1, characterized in that, The upper specimen servo adjustment module includes a micro digital servo and a rotating block driven by the servo. The rotating block is connected to the output shaft of the micro digital servo. The upper specimen quick-change fixture is mounted on the rotating block, which can rotate from 0° to 180°.

7. The method according to claim 1, characterized in that, The coating module includes a coating head holder, a coating head fixing plate, and a flexible coating head. The coating head holder is held by a quick-change fixture for the upper specimen. The coating head fixing plate is installed on the coating head holder, and the flexible coating head is fixed on the coating head fixing plate. The coating head holder and the coating head fixing plate are provided with flow channels that communicate with the flexible coating head.

8. The method according to claim 1, characterized in that, The specimen quick-change fixture includes a precision three-jaw chuck, and an expansion mandrel and a resilient bushing that are interchangeably mounted on the chuck; the expansion mandrel is used to hold a slip ring positioned by its inner hole, and the resilient bushing is used to hold a slip ring positioned by its outer diameter.

9. The method according to claim 1, characterized in that, The sensing and measurement module includes: A three-dimensional force sensor is installed on the flip block of the upper specimen servo adjustment module; An accelerometer is mounted on the flip block; Temperature sensor and acoustic emission sensor are installed on the upper specimen quick-change fixture near the brush. An auxiliary slip ring, whose rotor is connected to the quick-change fixture for the lower specimen; Circuit test bracket, used to fix test circuits; The electrical signal acquisition chain adopts a four-wire Kelvin connection method and includes a precision constant current source, a high-precision DC voltmeter, a high-pass filter, a low-noise amplifier, and a high-speed data acquisition card.

Images