Metal matrix composite reinforcement layer mechanical property intelligent detection device and method

Abstract

The application discloses a kind of metal matrix composite reinforcement layer mechanical performance intelligent detection device and method, including control system, mechanical loading device and with the main circuit loading device of control system connection;Mechanical loading device includes motion control platform, hard brush insulation fixed clamp and insulation sample placement platform, hard brush insulation fixed clamp is set on motion control platform, hard brush insulation fixed clamp is equipped with the hard brush with the pressure contact of sample reinforcement layer, and conductive rod is built-in with the pressure contact of sample metal matrix in insulation sample placement platform;Main circuit electric loading device is connected with hard brush and conductive rod respectively, and infrared thermal imager and laser vision sensor are equipped on the side of sample, and high-frequency vibration accelerometer is equipped on hard brush insulation fixed clamp.The application realizes the collaborative on-line detection of current impact resistance and wear resistance;Realize the dynamic damage on-line monitoring of reinforcement layer from perfect to failure whole process.

Description

Technical Field

[0001] This invention relates to the field of materials testing technology, specifically to an intelligent testing device and method for the mechanical properties of a metal matrix composite reinforced layer. Background Technology

[0002] Metal matrix composite reinforcement layers, compared to single metals or alloys, possess higher specific strength and specific modulus, excellent high-temperature performance, good hardness and wear resistance, and fatigue and creep resistance. Therefore, they are often used as reinforcement layers for critical components and are widely applied in extreme environments such as aerospace, nuclear energy, and defense. For example, the metal matrix composite reinforcement layers on the surfaces of components such as aero-engine blades, rocket nozzle bushings, and electromagnetic launch rails simultaneously withstand the combined effects of high temperatures, mechanical loads, and oxidizing / corrosive media during service. In particular, electromagnetic launch rails are subjected to significant intermittent current surges during continuous launches, leading to arc erosion, material melting and delamination, and wear failure on the rail surface, severely impacting launch accuracy and service life.

[0003] Currently, performance testing of reinforced layers mainly focuses on evaluating properties such as friction and wear, corrosion resistance, and high-temperature resistance under uninterrupted current. Some researchers have used continuously energized pin-disc friction and wear testing machines to study the current-carrying wear behavior of copper-based composite materials, or coupled high-temperature furnaces with mechanical testing machines for thermo-mechanical performance testing. However, these methods cannot simulate the transient thermal shock, arc erosion, and cooling and re-oxidation processes of materials during current interruptions caused by intermittent current impacts (i.e., periodic current switching). In actual electromagnetic launches, each launch lasts only a few milliseconds with extremely high current density (up to tens of thousands of amperes per square centimeter), followed by a current-free intermittent period where the orbital surface temperature fluctuates drastically. The friction and wear mechanism under this intermittent current condition is fundamentally different from that under uninterrupted current conditions. It is more likely to induce fatigue spalling, thermal stress crack propagation, and cumulative arc ablation in the reinforced layer material. Currently, there is no existing testing device capable of simulating the real-time coupling of this intermittent current impact and reciprocating frictional motion.

[0004] Furthermore, existing testing equipment generally lacks online intelligent detection capabilities. Traditional methods mostly employ offline detection techniques, such as assessing the damage level of the reinforcement layer after testing using scanning electron microscopy, energy dispersive spectroscopy, or 3D profilometry. This approach cannot reveal the dynamic damage evolution process during testing, makes it difficult to establish a real-time mapping relationship between the number of current impacts and material performance degradation, and makes it impossible to predict the remaining lifespan of the reinforcement layer online. For components requiring high reliability and predictable lifespan, such as electromagnetic launch tracks, this technological gap severely restricts the engineering application verification of novel reinforcement layers.

[0005] In recent years, although some studies have attempted to integrate sensors into tribological testing machines, such as measuring contact area temperature using thermocouples and collecting vibration signals using accelerometers, these systems mostly monitor single signals and have not been deeply integrated with artificial intelligence algorithms. In particular, for the performance testing of reinforced layers under extreme conditions of electro-thermal-mechanical multi-field coupling, there is still no integrated intelligent testing device that can simultaneously achieve cyclic current impact loading, mechanical friction and wear, multi-sensor data fusion, and online evaluation using deep learning.

[0006] Therefore, developing a device that can couple extreme working conditions of multiple fields of electricity, heat and force, simulate the impact of intermittent current, and integrate intelligent detection technology to realize the monitoring of the entire testing process and the prediction of remaining life has become a technical problem that urgently needs to be solved in this field.

[0007] The existing technology has the following problems: (1) It cannot simulate the multi-field coupling condition under intermittent current impact. Existing equipment can only realize the friction and wear test under uninterrupted current, and cannot simulate the transient thermal shock, arc erosion and intermittent cooling and re-oxidation process caused by the periodic switching of current.

[0008] (2) Lack of testing environment for extreme coupling of multiple electric-thermal-mechanical fields. Most existing testing devices are single or dual field coupling (thermal-mechanical, electric-thermal, etc.), lacking the ability to test three or more extreme working conditions simultaneously coupling electric field, temperature field, and mechanical field (normal load + reciprocating friction).

[0009] (3) The dynamic damage process of the reinforcement layer cannot be detected online. Existing methods rely on offline detection methods, which cannot know the real-time damage evolution during the test process and make it difficult to establish a mapping relationship between the number of current impacts and performance degradation.

[0010] (4) Lack of dedicated testing methods for current surge resistance. Current surge resistance and wear resistance are key evaluation indicators for reinforced layers, but existing equipment has failed to achieve simultaneous, online testing of these properties.

[0011] (5) Lack of intelligent assessment and remaining life prediction capabilities. Even if existing systems integrate sensors, they mostly monitor single signals or require offline analysis of collected data. They are not deeply integrated with artificial intelligence algorithms and cannot achieve online damage assessment and remaining life prediction based on multi-sensor data fusion. Summary of the Invention

[0012] The purpose of this invention is to provide an intelligent testing device and method for the mechanical properties of metal matrix composite reinforcement layers, which can simulate actual working conditions to achieve coordinated online testing of current impact resistance and wear resistance; and realize dynamic online monitoring of the entire process of reinforcement layer from integrity to failure.

[0013] The technical solution adopted in this invention is: A smart testing device for the mechanical properties of a metal matrix composite reinforced layer includes a mechanical loading device, a main circuit electric loading device, and a control system. The mechanical loading device includes a motion control platform, a hard brush insulating fixture, and an insulating sample placement platform. The insulating sample placement platform is provided with multiple insulating sample fixing sliders for clamping the sample placed on the insulating sample placement platform. The sample is a metal matrix with a surface-reinforced reinforcing layer. The motion control platform is arranged opposite to the insulating sample placement platform. The hard brush insulating fixture is connected to the motion control platform. The hard brush is connected to the hard brush and is used to make pressure contact with the reinforcing layer at one end of the sample. The motion control platform drives the hard brush insulating fixture and the hard brush to move back and forth laterally relative to the sample by controlling the transverse drive motor, so that the hard brush and the corresponding contact surface of the sample form back and forth friction. The motion control platform is connected to the control system. The insulating sample placement platform has a built-in conductive rod. One end of the conductive rod passes through the insulating sample placement platform and makes pressure contact with the metal substrate at the other end of the sample. The two ends of the main circuit electric loading device are connected to the hard brush and the conductive rod, respectively. The control system is connected to the main circuit electric loading device. An infrared thermal imager and a laser vision sensor are installed on one side of the sample placement position. A high-frequency vibration accelerometer is installed on the hard brush insulated fixing fixture. The infrared thermal imager, laser vision sensor, and high-frequency vibration accelerometer are connected to the control system.

[0014] Preferably, the main circuit electrical loading device includes a high-power DC power supply, a high-speed electronic switch, a current sensor, and an oscilloscope. One end of the high-power DC power supply is connected to a hard brush through the high-speed electronic switch, and the other end is connected to a conductive rod through the current sensor. The control terminal of the high-power DC power supply is connected to the control system through the oscilloscope.

[0015] Preferably, the hard brush includes a hard brush fixing head, a hard brush connecting rod, and a hard brush friction contact head. The hard brush fixing head is connected to the hard brush friction contact head through the hard brush connecting rod. The main circuit electric loading device is connected to the hard brush fixing head wire. The hard brush friction contact head is used for pressure contact with the metal-based reinforcing layer of the sample.

[0016] Preferably, the motion control platform includes a normal loading mechanism, a lateral driving mechanism, and a motion controller. The motion controller is connected to the normal loading mechanism and the lateral driving mechanism respectively. The hard brush insulation fixing fixture is slidably connected to the normal loading mechanism. The normal loading mechanism is used to apply a normal force perpendicular to the friction contact surface of the sample to the hard brush insulation fixing fixture. The lateral driving mechanism is connected to the hard brush insulation fixing fixture and is used to drive the hard brush insulation fixing fixture and the hard brush to move laterally back and forth. The motion controller is connected to the control system.

[0017] Preferably, the normal loading mechanism includes a normal loading motor and a sliding guide rod. The output end of the normal loading motor is connected to one end of the sliding guide rod, and the other end of the sliding guide rod is slidably connected to the rigid brush insulation fixing fixture. The normal loading motor is connected to the motion controller. The lateral drive mechanism includes a lateral drive motor, a ball screw, and a screw nut. The output end of the lateral drive motor is connected to one end of the ball screw, and the other end of the ball screw is connected to one side of the rigid brush insulation fixing fixture. The screw nut is threaded onto the ball screw and is fixedly arranged. The lateral drive motor is connected to the motion controller.

[0018] Preferably, the rigid brush insulation fixing fixture is provided with a sliding guide rail, and the end of the sliding guide rod is disposed on the sliding guide rail. The sliding guide rod can move back and forth along the sliding guide rail relative to the rigid brush insulation fixing fixture.

[0019] Preferably, the insulating sample placement platform is provided with a slide groove, and the insulating sample fixing slider is disposed on the slide groove. The insulating sample fixing slider can move along the slide groove on the insulating sample placement platform. The insulating sample placement platform is used to place the sample, and the insulating sample fixing sliders on both sides move along the slide groove to clamp the sample.

[0020] Preferably, the control system includes a multi-sensor fusion data acquisition system and an intelligent detection system. The intelligent detection system is connected to the motion control platform, an infrared thermal imager, a laser vision sensor, and a high-frequency vibration accelerometer through the multi-sensor fusion data acquisition system.

[0021] Preferably, the intelligent detection system includes a computer and an alarm, the alarm is connected to the computer, the computer is connected to the high-power DC power supply of the main circuit electric loading device, and the computer is connected to a multi-sensor fusion data acquisition system; The multi-sensor fusion data acquisition system is a multi-modal data acquisition card.

[0022] A detection method using the intelligent detection device for the mechanical properties of the metal matrix composite reinforced layer described above includes the following steps: The test sample is fixed on the insulation sample placement platform, and the test sample is clamped by the insulation sample fixing slider. The hard brush and the conductive rod are in pressure contact with the reinforcing layer at the top and the bottom surface of the metal substrate at the bottom of the sample, respectively. The main circuit electric loading device is electrically connected to the hard brush and the conductive rod to form a closed electric loading circuit. The motion control platform is started. The motion control platform drives the hard brush insulation fixing fixture and hard brush to apply a set normal load to the sample by controlling the normal loading motor, so that the hard brush and the surface of the reinforcement layer of the sample are in pressure contact. The motion control platform drives the hard brush to form a reciprocating cyclic friction with the contact surface of the sample, and / or the main circuit electric loading device impacts the output current of the sample. The process of using a motion control platform to drive a hard brush to form a reciprocating cyclic friction with the sample contact surface refers to the motion control platform driving the hard brush insulation fixing fixture to move the hard brush back and forth laterally at a set frequency, so that the hard brush forms a reciprocating cyclic friction with the sample contact surface under normal load pressure, and the number of cycles of friction is recorded. The impact of the main circuit electric loading device on the sample output current refers to the intermittent or continuous impact of the main circuit electric loading device on the sample output current. The intermittent output current impact is the main circuit electric loading device that outputs a current impact once every N cycles of friction or at a set interval t; N or t can be set before starting the motion control platform. The control system acquires the sample temperature in real time using an infrared thermal imager, and acquires the changes in the height and position of the hard brush and the friction surface image of the sample reinforcement layer in real time using a laser vision sensor. The control system obtains the wear depth of the sample reinforcement layer based on the changes in the height and position of the hard brush, and obtains the ablation area ratio of the sample reinforcement layer based on the friction surface image of the sample reinforcement layer. The system also acquires the friction vibration amplitude between the hard brush and the sample in real time using a high-frequency vibration accelerometer, and plots the friction vibration spectrum based on the changes in the friction vibration amplitude over time, thereby constructing a friction curve that changes over time. If one of the following two failure conditions is detected, the sample is judged to be in failure; otherwise, the hard brush is continuously rubbed against the contact surface of the sample until the sample is judged to be in failure. One of the failure conditions includes the satisfaction of any one of the following four conditions: The ablation area ratio reaches the first preset threshold. The wear depth reaches the second preset threshold, or the thickness of the reinforcement layer reaches the third preset threshold; The friction curve abrupt change factor reaches the fourth preset threshold; The sample temperature exceeds the melting point of the corresponding material. Another failure condition is: failure probability > 0.8, or the predicted number of residual current impact cycles ≤ 10% of the maximum current impact cycles; Failure probability

[0023] Where Z is a weighted linear combination of the input features;

[0024] A This represents the proportion of the arc-eroded area. w A This is the weighting coefficient for the arc ablation area scaling factor. D For wear depth, w D Here, represents the weighting coefficient of the wear depth factor, and T represents the sample temperature. w T This represents the weighting coefficient for the sample temperature factor. f The amplitude of frictional vibration. w f The weighting coefficients for the friction vibration amplitude factor. ΔI The deviation of the current amplitude of the main circuit electric loading device. w ΔI This is the weighting coefficient for the current amplitude deviation factor. R The contact resistance of the sample. R This is the weighting coefficient for the contact resistance factor. b For bias terms; The predicted number of residual current impact cycles is the maximum number of current impact cycles minus the number of current impact cycles already completed. The maximum number of current impact cycles is determined based on the different sample materials.

[0025] The beneficial effects of this invention are: In this invention, the control system applies electric and thermal fields through the main circuit electric loading device and applies mechanical loading through the motion control platform. The mechanical loading includes normal load and reciprocating friction. It is integrated with an infrared thermal imager for online temperature field monitoring, thus constructing a three-field extreme working condition test environment that can simultaneously couple electric field, temperature field and mechanical field. While the current impact is applied, the reciprocating friction and wear test can be performed simultaneously, simulating actual working conditions and realizing the coordinated online detection of current impact resistance and wear resistance. During the test, the temperature distribution, wear mark depth, arc ablation area ratio, vibration spectrum and current magnitude are collected in real time, realizing dynamic online monitoring of the entire process of the reinforcement layer from intact to failure. A real-time mapping relationship between the number of current impacts and performance degradation is established, eliminating the dependence on offline detection methods. Attached Figure Description

[0026] Figure 1This is a schematic diagram of the intelligent detection device for the mechanical properties of the metal matrix composite reinforcement layer in Embodiment 1 of the present invention.

[0027] Figure 2 This is a flowchart of the intelligent detection method for mechanical properties of metal matrix composite reinforced layers in this embodiment of the invention, which is used for sample failure judgment.

[0028] Figure 3 This is a flowchart of the intelligent detection method for the mechanical properties of the metal matrix composite reinforcement layer in this embodiment of the invention.

[0029] In the figure: 1. Insulation sample placement platform; 2. Slide groove; 3. Insulation sample fixing slider; 4. Infrared thermal imager; 5. Laser vision sensor; 6. High-frequency vibration accelerometer; 7. Motion controller; 8. Sliding guide rod; 9. Sliding guide rail; 10. Hard brush insulation fixing clamp; 11. Hard brush; 11. Hard brush fixing head; 111. Hard brush connecting rod; 112. Hard brush friction contact head; 113. Metal base reinforcement layer; 12. High-speed electronic switch; 13. High-power DC power supply; 14. Lateral drive motor; 15. Connecting wire; 16. Multimodal data acquisition card; 17. Alarm; 18. Computer; 19. Oscilloscope; 20. Hall current sensor; 21. Conductive rod; 22. Metal substrate; 23. Normal loading motor; 24. Ball screw; 25. Screw nut; 26. Detailed Implementation

[0030] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0031] In the description of this invention, it should be understood that if terms such as "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," and "counterclockwise" are used to indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, they are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of the stated features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0032] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "joining" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection. They can refer to a mechanical connection or an electrical connection. They can refer to a direct connection or an indirect connection through an intermediate medium, and they can refer to the internal communication of two components or the interaction between two components. For those skilled in the art, the specific meaning of the above terms in this invention can be understood according to the specific circumstances.

[0033] Example 1 A smart testing device for the mechanical properties of a metal matrix composite reinforcement layer, such as Figure 1 As shown, it includes a mechanical loading device, a main circuit electric loading device, and a control system; the mechanical loading device includes a motion control platform, a hard brush insulating fixing clamp 10, and an insulating sample placement platform 1. The insulating sample placement platform 1 is provided with multiple insulating sample fixing sliders 3 for clamping the sample placed on the insulating sample placement platform 1. The sample is a metal substrate 23 with a surface-reinforced bonding layer 12. The motion control platform is positioned above the insulating sample placement platform 1. The rigid brush insulating fixture 10 is connected to the motion controller 7. The rigid brush insulating fixture 10 is connected to a rigid brush 11, which is used to make pressure contact with the reinforcing layer 12 at the top of the sample. The motion control platform drives the rigid brush insulating fixture 10 and the rigid brush 11 to move back and forth laterally relative to the sample, so that the rigid brush 11 and the corresponding contact surface of the sample form back and forth friction. The motion control platform is connected to the control system. The insulating sample placement platform 1 has a built-in conductive rod 22. One end of the conductive rod 22 extends out of the insulating sample placement platform 1 and makes pressure contact with the metal substrate 23 at the bottom of the sample. The input and output ends of the main circuit electric loading device are connected to the hard brush 11 and the conductive rod 22, respectively. The control system is connected to the main circuit electric loading device. An infrared thermal imager 4 and a laser vision sensor 5 are provided on one side of the sample. A high-frequency vibration accelerometer 6 is provided on the hard brush insulating fixing fixture 10. The infrared thermal imager 4, the laser vision sensor 5, and the high-frequency vibration accelerometer 6 are connected to the control system.

[0034] Furthermore, the main circuit electrical loading device includes a high-power DC power supply 14, a high-speed electronic switch 13, a Hall current sensor 21, and an oscilloscope 20. The high-power DC power supply 14 is connected to the hard brush 11 through the high-speed electronic switch 13 and to the conductive rod 22 through the Hall current sensor 21. The control terminal of the high-power DC power supply 14 is connected to the control system through the oscilloscope 20.

[0035] The output terminal of the high-power DC power supply 14 is connected to the input port of the high-speed electronic switch 13 via a wire. The high-speed electronic switch 13 performs precise control of the current on and off. The Hall current sensor 21 is used to acquire the current change during the test in real time to form a complete electrical circuit. The oscilloscope 20 is used to monitor the output waveform.

[0036] Furthermore, the hard brush 11 includes a hard brush fixing head 111, a hard brush connecting rod 112, and a hard brush friction contact head 113. The hard brush fixing head 111 is connected to the hard brush friction contact head 113 through the hard brush connecting rod 112. The output terminal of the high-speed electronic switch 13 of the main circuit electric loading device is connected to the wire of the hard brush fixing head 111. The hard brush friction contact head 113 is used for pressure contact with the metal base reinforcement layer 12 of the sample.

[0037] Example 2 The motion control platform is further defined based on Example 1.

[0038] The motion control platform includes a normal loading mechanism, a lateral drive mechanism, and a motion controller 7. The motion controller 7 is connected to the normal loading mechanism and the lateral drive mechanism respectively. The hard brush insulation fixing fixture 10 is slidably connected to the normal loading mechanism. The normal loading mechanism is used to apply a normal force perpendicular to the friction contact surface of the sample to the hard brush insulation fixing fixture 10. The lateral drive mechanism is connected to the hard brush insulation fixing fixture 10. The lateral drive mechanism is used to drive the hard brush insulation fixing fixture 10 and the hard brush 11 to move laterally back and forth. The motion controller 7 is connected to the control system.

[0039] The normal loading mechanism includes a normal loading motor 24 and a sliding guide rod 8. The output end of the normal loading motor 24 is connected to one end of the sliding guide rod 8, and the other end of the sliding guide rod 8 is slidably connected to the upper end of the hard brush insulating fixing clamp 10. The normal loading motor 24 is connected to the motion controller 7. The lateral drive mechanism includes a lateral drive motor 15, a ball screw 25, and a screw nut 26. The output end of the lateral drive motor 15 is connected to one end of the ball screw 25, and the other end of the ball screw 25 is connected to one side of the rigid brush insulation fixing clamp 10. The screw nut 26 is threaded onto the ball screw 25 and is fixedly arranged. The lateral drive motor 15 is connected to the motion controller 7. The lateral drive motor 15 drives the ball screw 25 to extend and rotate, thereby pushing the rigid brush insulation fixing clamp 10 to move back and forth laterally.

[0040] The contact pressure is provided by the normal loading drive motor 24, and the vertical pressure of the normal loading motor 24 is controlled by the motion controller 7. The hard brush insulation fixing fixture 10 is connected to the transverse drive motor 15 through the ball screw 25. The rotation of the motor drives the screw nut 26 to move back and forth, thereby driving the hard brush insulation fixing fixture 10 and the hard brush 11 to move laterally back and forth relative to the sample, so that the hard brush 11 and the corresponding contact surface of the sample form back and forth friction.

[0041] Furthermore, the rigid brush insulation fixing fixture 10 is provided with a sliding guide rail 9, and the end of the sliding guide rod 8 is disposed on the sliding guide rail 9. The sliding guide rod 8 can move back and forth along the sliding guide rail 9 relative to the rigid brush insulation fixing fixture 10.

[0042] Furthermore, the normal loading drive motor 24 directly outputs normal pressure. The normal loading drive motor 24 applies a vertically downward load to the hard brush insulation fixing fixture 10 and the hard brush 11 via the sliding guide rod 8. Simultaneously, the lateral drive motor 15 drives the hard brush insulation fixing fixture 10 and the hard brush 11 to move laterally back and forth via the ball screw 25, causing friction between the hard brush 11 and the sample. The friction drive frequency can be set via the motion controller 7 connected to the lateral drive motor 15. This is used to realize the movement of the hard brush on the sample surface and test the wear resistance of the sample.

[0043] Furthermore, the transverse drive motor 15 outputs linear reciprocating motion to the motion control platform, where the motion controller 7, combined with position / speed feedback, achieves closed-loop control. According to set parameters (speed, frequency, stroke), the motion control platform pushes the hard brush insulation fixture 10 connected to it to perform linear reciprocating motion along the sliding guide rail 9, applying a normal force. The hard brush 11, fixed within the fixture, moves synchronously, causing its friction contact head to generate sliding friction relative to the sample surface. During this process, the sliding guide rod 8 applies a directional force and slides along the sliding guide rail 9 with the hard brush insulation fixture 10. Purpose: To achieve controllable reciprocating sliding friction between the hard brush (friction pair) and the sample surface, simulating the relative motion between the friction pair and the reinforcing layer under actual working conditions. Simultaneously, in conjunction with cyclic current impact, it achieves coordinated testing of wear resistance and current impact resistance under electro-thermal-mechanical multi-field coupling conditions.

[0044] Furthermore, the insulating sample placement platform 1 is provided with a slide groove 2, and the insulating sample fixing slider 3 is set on the slide groove 2. The insulating sample fixing slider 3 can move along the slide groove 2 on the insulating sample placement platform 1. The insulating sample placement platform 1 is used to place the sample, and the insulating sample fixing sliders 3 on both sides move along the slide groove 2 to clamp the sample.

[0045] Furthermore, the conductive rod 22 is L-shaped.

[0046] Furthermore, the control system includes a multi-sensor fusion data acquisition system and an intelligent detection system. The intelligent detection system is connected to the motion controller 7, infrared thermal imager 4, laser vision sensor 5, and high-frequency vibration accelerometer 6 of the motion control platform through the multi-sensor fusion data acquisition system.

[0047] Furthermore, all sensors communicate with the multimodal data acquisition card to construct time series data. An infrared thermal imager 4 is placed horizontally, focusing on the temperature distribution at the contact area between the reinforcing layer and the hard brush 11. A laser vision sensor is aligned with the contact area to monitor the wear track depth and the proportion of arc ablation area in real time. A high-frequency vibration accelerometer is fixed to the hard brush insulation fixture to collect the friction vibration spectrum.

[0048] Furthermore, the insulating sample fixing slider 3 is arranged circumferentially along the sample placement position on the insulating sample placement platform 1, and the slide groove 2 is arranged radially along the sample placement position.

[0049] When there are 2 insulating sample fixing sliders 3, the insulating sample fixing sliders 3 are arranged on the left and right sides of the sample placement position respectively; when there are 4 insulating sample fixing sliders 3, the insulating sample fixing sliders 3 are arranged on the front, back, left and right sides of the sample placement position respectively.

[0050] Furthermore, the intelligent detection system includes a computer 19 and an alarm 18. The alarm 18 is connected to the computer 19. The computer 19 is connected to the high-power DC power supply 14 of the main circuit electric loading device via an oscilloscope 20. The computer 19 is also connected to a multi-sensor fusion data acquisition system. The multi-sensor fusion data acquisition system is a multi-modal data acquisition card 17. The computer 19 communicates with the multi-modal data acquisition card 17 in real time to acquire data streams.

[0051] A detection method using the intelligent detection device for the mechanical properties of the metal matrix composite reinforcement layer described above, such as... Figure 2-3 As shown, it includes the following steps: The test sample is fixed on the insulation sample placement platform 1, and the test sample is clamped by the insulation sample fixing slider 3. The hard brush 11 and the conductive rod 22 are respectively in pressure contact with the reinforcing layer 12 at the top of the sample and the bottom surface of the metal substrate 23 at the bottom. The main circuit electric loading device is electrically connected to the hard brush 11 and the conductive rod 22 to form a closed electric loading circuit. Calibrate the infrared thermal imager 4, the laser vision sensor 5, and the high-frequency vibration accelerometer 6; The motion control platform is activated, and it applies a set normal load to the sample through the hard brush insulating fixture 10 and the hard brush 11, so that the hard brush 11 makes pressure contact with the surface of the reinforced layer of the sample. The motion control platform drives the hard brush 11 to form a reciprocating cyclic friction with the contact surface of the sample, and / or the main circuit electric loading device impacts the output current of the sample. The reciprocating cyclic friction between the hard brush 11 and the sample contact surface caused by the motion control platform refers to the following: the motion control platform controls the hard brush insulation fixing clamp 10 to move the hard brush 11 back and forth laterally at a set frequency by controlling the transverse drive motor 15, so that the hard brush 11 forms a reciprocating cyclic friction with the sample contact surface under normal load pressure, and the number of cyclic frictions is recorded. The impact of the main circuit electric loading device on the sample output current refers to the intermittent or continuous impact of the main circuit electric loading device on the sample output current. Intermittent current surge output is either once every N cycles of friction or once at a set interval t; N or t can be set before starting the motion control platform. The control system uses an infrared thermal imager 4 to collect sample temperature in real time, and a laser vision sensor 5 to collect images of the height position change of the hard brush 11 and the friction surface of the sample's reinforcing layer in real time. The control system obtains the wear track depth of the reinforcing layer based on the height position change of the hard brush 11, and the ablation area ratio of the reinforcing layer based on the friction surface image. A high-frequency vibration accelerometer 6 collects the friction vibration amplitude between the hard brush 11 and the sample in real time, and plots the friction vibration spectrum along the time axis, thus constructing a friction curve that varies with time; the vibration spectrum is the friction curve. The horizontal axis represents time, and the vertical axis represents the vibration amplitude. When the sample surface is rubbed, the equipment generates vibration signals, and the collected vibration signals form the friction curve that varies with time. If the friction is stable, the vibration spectrum changes smoothly; if the coating is worn through, the vibration spectrum changes abruptly.

[0052] If one of the following two failure conditions is detected, the sample is judged to be failed; otherwise, the hard brush 11 is continuously rubbed against the contact surface of the sample until the sample is judged to be failed. One of the failure conditions includes the satisfaction of any one of the following four conditions: The ablation area ratio reaches the first preset threshold. The wear depth reaches the second preset threshold, or the thickness of the reinforcement layer reaches the third preset threshold; The friction curve abrupt change factor reaches the fourth preset threshold; The sample temperature exceeds the melting point of the corresponding material. Another failure condition is: failure probability > 0.8, or the predicted number of residual current impact cycles ≤ 10% of the maximum current impact cycles; Failure probability

[0053] Where Z is a weighted linear combination of the input features;

[0054] A This represents the proportion of the arc-eroded area. w A This is the weighting coefficient for the arc ablation area scaling factor. D For wear depth, w D This represents the weighting coefficient of the wear depth factor. T The temperature of the sample. w T This represents the weighting coefficient for the sample temperature factor. f The amplitude of frictional vibration. w f The weighting coefficients for the friction vibration amplitude factor. ΔI The deviation of the current amplitude of the main circuit electric loading device. w ΔI This is the weighting coefficient for the current amplitude deviation factor. R The contact resistance of the sample. R This is the weighting coefficient for the contact resistance factor. b These are bias terms (determined after training with the fully connected layer or determined through calibration). The predicted number of residual current impact cycles is the maximum number of current impact cycles minus the number of current impact cycles already completed. The maximum number of current impact cycles is calibrated according to different sample materials under the corresponding test modes.

[0055] The control system automatically terminates the test, triggers an alarm, and generates a test performance degradation report after determining that the test sample has failed.

[0056] Furthermore, the wear depth of the reinforcing layer is obtained. The specific process is as follows: first, the height position of the hard brush 11 is identified by scanning with the laser vision sensor 5, and then the wear depth of the reinforcing layer of the sample is obtained based on the height position of the hard brush 11.

[0057] Infrared sensors monitor the temperature change of the entire sample. If the temperature exceeds the melting point of the test sample, an alarm is triggered to stop the test. During friction and electrical impact, different currents and friction parameters will cause temperature changes in the sample. Recording data can establish the temperature rise of the sample under different parameters. If the temperature exceeds the melting point of the metal sample, it can prevent the metal sample from vaporizing under the current impact and damaging the equipment. If the melting point is not exceeded, the recorded data can reflect the temperature change of the sample under that parameter.

[0058] The control system obtains the ablation area ratio of the sample reinforcement layer based on the friction surface image of the sample reinforcement layer. The specific process is as follows: the laser vision sensor is controlled to scan the ablation area on the sample surface line by line to collect the three-dimensional point cloud data of the ablation area; based on the three-dimensional point cloud data, the three-dimensional irregular curved surface model of the ablation area is reconstructed; the three-dimensional irregular curved surface model is subjected to mesh discretization processing, and the true surface area of ​​the ablation area is calculated based on the area integral algorithm. The calculated value is used as the ablation area to obtain the ablation area ratio.

[0059] Computer 19 has built-in control software and a deployed deep learning model. The control software handles parameter setting, start / stop control, data storage, and alarms. The deep learning model takes temperature, vibration characteristic values, wear depth, and ablation area ratio as inputs and outputs the remaining cycle count and failure probability.

[0060] Furthermore, the computer 19 of the control system has a built-in deep learning model, which is an LSTM model, but can also be replaced with other deep learning models.

[0061] The LSTM deep learning model takes as input temperature signals from an infrared thermal imager, vibration signals from a high-frequency vibrometer, and surface ablation area ratio and wear track depth data from a laser vision sensor. It assesses the current damage level using data from multiple sensors. Temperature signals: Temperature changes in the contact area are caused by both Joule heating and frictional heat. The temperature rise rate reflects changes in the thermal conductivity of the reinforcing layer. When microcracks or delamination appear within the reinforcing layer, the interfacial thermal resistance increases, and the temperature rise rate becomes abnormal. The LSTM model determines whether structural degradation has occurred within the reinforcing layer by capturing the "inflection point of the temperature rise rate trend." Vibration signals: Frictional vibration is generated by changes in the surface roughness and wear behavior. As wear intensifies, the spectral amplitude characteristics change. The LSTM model assesses the severity of wear by the degree of change in spectral amplitude. Wear track depth: Directly reflects the volumetric wear of the reinforcing layer. A linear growth region of wear track depth represents a uniform wear stage; when the wear track depth curve shows "exponential accelerated growth," it indicates that the reinforcing layer is close to penetrating failure. LSTM models determine whether accelerated failure has begun by identifying abrupt changes in the second derivative of the "wear depth-cycle count" curve. The ablation area ratio reflects the cumulative effect of arc erosion. A "step-like increase" in the ablation area typically corresponds to sudden damage caused by a high-current impact. LSTM models assess the severity of sudden damage events by identifying the "step characteristics of the ablation area."

[0062] The output of the LSTM deep learning model assesses the remaining lifetime of the specimen and the probability of failure of the reinforcing layer under the current state, based on material property degradation characteristics and damage degree. Remaining lifetime (regression task): The model learns the mapping relationship between "sensor feature evolution trajectory" and "remaining cycle count" in historical full-lifetime data. It performs temporal pattern matching between the feature vector of the current sliding window and the feature vectors of different degradation stages in the historical database, and combines this with the regression output to provide a point estimate of the remaining lifetime. If the degradation trajectory of the current temperature sequence combined with vibration features has the highest similarity to the trajectory at cycle k in the database, then the remaining lifetime at that moment is estimated to be at cycle k.

[0063] Failure Risk Probability (Classification Task): The model compares the current feature vector with two categories of samples from historical data: "failed within k cycles" and "did not fail within k cycles." It outputs the conditional probability of failure occurring within the next k cycles under the current state. An alert is triggered when this probability > 0.8. The physical meaning of this probability is: under the same or similar sensor feature states, more than 80% of the samples in the historical data failed within k cycles.

[0064] The high-power DC power supply has an output power range of 0~120 kW, an output voltage range of 0~150 V, and an output current range of 0~12000 A, providing the required electrical load for testing.

[0065] Furthermore, the high-speed electronic switch has a switching frequency of 0.1~100 Hz, which is used to precisely control the switching sequence of the current, thereby simulating the intermittent current in actual working conditions.

[0066] Furthermore, the Hall current sensor has a measurement range of 0~3000 A.

[0067] Furthermore, the drive motor is used to control the loading pressure, while simultaneously achieving precise control of the reciprocating motion frequency, speed, and position. Its loading pressure range is 1-500 N, with a control accuracy of ±1%, and a frequency range of 0.1-50 Hz.

[0068] Furthermore, the motion control platform is used to control the reciprocating speed of the hard brush, which has a reciprocating speed range of 0.001~0.5 m / s and is continuously adjustable.

[0069] Furthermore, the multimodal data acquisition card can acquire signals from multiple sensors in real time, with ≥8 channels and a maximum sampling rate of 1MS / s, establishing a continuous time series.

[0070] A detection method using the intelligent detection device for the mechanical properties of the metal matrix composite reinforced layer described above includes the following steps: S1: Sample Installation and Parameter Setting. Fix the sample to be tested on the insulating sample placement platform 1 and clamp it using the insulating sample fixing slider 3 to ensure the sample remains stationary during the test. Connect the main circuit electric loading device in sequence to form a closed current loop: Start the motion control platform and apply a normal load through the normal loading motor 24 to make the hard brush 11 contact the surface of the reinforcement layer. Set the loading pressure to the target value within the range of 1~500 N. Calibrate and align the infrared thermal imager 4, laser vision sensor 5, and high-frequency vibration accelerometer 6, and establish communication between all sensors and the multimodal data acquisition card 17 and computer 19. Set the electrical parameters (voltage, current, switching frequency, total number of cycles), motion parameters (reciprocating speed, stroke length), damage threshold (ablation area ratio, wear depth, wear curve abrupt change factor), and model output interval (output once every N cycles) in the control software. Select the current impact mode and / or reciprocating friction mode according to the test requirements. The current impact mode includes intermittent current impact mode and continuous current impact mode. This equipment offers several modes to suit different test materials and corresponding simulated operating conditions. The first is a simple current impact mode, testing the sample's resistance to current impact. The second is a simple wear performance test mode, where no current is applied, only the material's wear resistance is tested. The third is an intermittent current impact superimposed wear test mode, providing both intermittent current impact and friction testing. The fourth is a continuous current superimposed wear test mode, where the current is continuously present, testing the material's wear resistance under continuous current.

[0071] S2: Start the test. After confirming that the parameters are correct, start the test device through the control software. During the test, the high-speed electronic switch 13 automatically switches on and off according to the set frequency and duty cycle, generating intermittent current surges. The motion control platform drives the hard brush insulating fixing fixture 10 and the hard brush 11 to reciprocate and slide relative to the surface of the stationary sample in the transverse direction. The transverse drive motor 15 maintains a constant output in real time.

[0072] S3: Multi-sensor synchronous data acquisition. Throughout the test, the infrared thermal imager 4 records the temperature distribution of the contact area in real time, the laser vision sensor 5 dynamically monitors the depth of the wear marks and the proportion of the arc ablation area, the high-frequency vibration accelerometer 6 collects the friction vibration spectrum, and the Hall current sensor 21 collects the main circuit current waveform. The multi-modal data acquisition card 17 aligns all sensor data according to a unified time reference, forming a multi-channel time series data stream, and transmits it to the computer 19.

[0073] S4: The intelligent evaluation model processes the incoming data stream in real time. After receiving the raw data stream, computer 19 performs preprocessing such as denoising, normalization, and feature extraction. The pre-trained LSTM model deployed in computer 19 reads the preprocessed data from the most recent current impacts in a sliding window manner. The inputs include temperature sequence, vibration characteristic values, surface morphology characteristics, and current parameters. The outputs are the remaining lifespan of the sample and the probability of failure of the reinforcement layer under the current state. After every N preset current impact cycles, the control software outputs an evaluation result and generates a performance degradation trend graph. At the same time, the control software monitors sensor data and model output in real time: when the arc ablation area ratio, wear mark depth, or wear curve change rate exceeds the set threshold, or the model outputs a failure risk probability greater than 0.8, the system automatically terminates the test, cuts off the current loop and motion control, triggers the alarm, and saves all data.

[0074] S5: Generate a test performance degradation report. The system automatically generates a test performance degradation report regardless of whether the preset number of cycles is reached or the test is terminated prematurely due to sample failure. The report includes sample information, test parameters, raw data from each sensor, intelligent evaluation results (failure probability, performance degradation trend graph), failure determination records, and test conclusions.

[0075] The performance degradation trend graph shows the changes in the wear mark depth and ablation area of ​​the sample material over time.

[0076] Four test modes: Mode 1: Simple current impact mode, used to test the current impact resistance of the sample.

[0077] Applicable materials and working conditions: Reinforcing layer + copper base. The reinforcing layer material can be titanium-based composite material (melting point ~1660℃), nickel-based composite material (melting point ~1450℃), etc. It is used to simulate the working condition of electromagnetic railgun statically bearing a large current impact without sliding friction (initial firing moment), as well as the instantaneous current impact of high voltage electric contacts and electrical discharge machining electrodes.

[0078] Process parameter setting range: current: 2000-5000 A, current waveform: intermittent impact, 10-50 Hz, voltage: 30-50 V, normal load: 50-300 N, failure characteristics: arc ablation area ratio ≥ 50% of the overall surface area, sample temperature ≥ the melting temperature of the corresponding material, LSTM predicted failure probability > 0.8, predicted number of electric impacts ≤ 10% of the cycled area.

[0079] Mode 2: Simple wear performance test mode, which tests the wear resistance of the reinforced layer at room temperature.

[0080] Applicable materials and working conditions: Reinforcing layer + iron base, the reinforcing layer material can be titanium-based composite material, iron-based composite material, etc., which can be used to simulate the wear of rail transit lines.

[0081] Process parameter setting range: normal load: 10-500 N, reciprocating speed: 0.001-0.5 m / s, failure characteristics: wear mark depth ≥ reinforcement layer thickness, abrupt change in friction curve, LSTM predicted failure probability > 0.8.

[0082] Mode 3: Intermittent current impact superimposed wear test mode, which detects the performance of the sample under the action of current and load.

[0083] Applicable materials and working conditions: Reinforcing layer + copper base. The reinforcing layer material can be titanium-based composite material (melting point ~1660℃), nickel-based composite material (melting point ~1450℃), etc. It is used to simulate the real environment of electromagnetic railgun firing and the performance changes of the sample under multi-field coupling.

[0084] Process parameter setting range: Current: 2000-5000 A, Current waveform: intermittent impact, 10-50 Hz, Voltage: 30-50 V, Normal load: 50-300 N, Reciprocating speed: 0.01-0.3 m / s, Failure characteristics: Arc ablation area ≥ 50% of the overall surface area, Wear track depth ≥ Reinforcement layer thickness, Abrupt change in friction curve, Sample temperature ≥ Melting temperature of the corresponding material, LSTM predicted failure probability > 0.8, Predicted number of electrical impacts ≤ 10% of the cycled area.

[0085] Mode 4: Continuous current superimposed wear test mode, which detects the wear resistance of the sample under continuous current.

[0086] Applicable materials and working conditions: Reinforced layer + copper base. The reinforcement layer material can be iron-based composite material (melting point ~1660℃), nickel-based composite material (melting point ~1450℃), etc., to simulate the working conditions of electrical contact components such as high-voltage switch contacts, pantograph slide plates, and microelectronic device contacts that are constantly energized and have relative sliding.

[0087] Process parameter setting range: current: 100-1000 A, voltage: 30-100 V, normal load: 50-300 N, reciprocating speed: 0.01-0.5 m / s, failure characteristics: arc ablation area ratio ≥ 50% of the overall surface area, wear mark depth ≥ reinforcement layer thickness, abrupt change in the friction curve, sample temperature ≥ melting temperature of the corresponding material, LSTM predicted failure probability > 0.8.

[0088] Under the criterion that the failure probability output by the LSTM deep learning model in the control system is >0.8, the weight allocation is determined according to the physical contribution of each input parameter to the failure of the reinforcement layer. The total weight is 1, and the weights of each parameter can be selected according to different selection modes.

[0089] Table 1. Factors influencing LSTM model prediction and weight selection in control systems.

[0090] In actual testing, LSTM reads the most recent input feature tensors (temperature sequence, vibration feature value, surface morphology feature and current parameter) in a sliding window manner, extracts the temporal features through the LSTM hidden layer, and generates a high-dimensional feature vector. This vector encodes the multimodal temporal information of the current degradation stage.

[0091] The feature vectors output by the LSTM hidden layer are dimensionality reduced by the fully connected layer, and then the weighted features are linearly combined and mapped to probability values ​​through the logistic regression function.

[0092] When the failure probability output by the LSTM model P When the probability is greater than 0.8, the system triggers an early warning. The failure probability is calculated using a logistic regression function.

[0093] in, Z It is a weighted linear combination of input features.

[0094]

[0095] b The bias term (determined after training with the fully connected layer) is where the input parameters are all normalized values, all in the range of 0 to 1.

[0096] The weight coefficients of each input parameter are determined based on the physical contribution, and the allocation principle varies depending on the selected mode.

[0097] Mode 1: Weighting under simple current impact: w A =0.50, w T =0.30, w ΔI =0.10, w R =0.10. The direct damage to the reinforced layer surface by the current surge, and the proportion of ablation area directly reflects the severity of arc erosion, which is the most critical criterion for failure. In this process, temperature is used as the second weighted data, reflecting the peak value of transient thermal shock, and excessively high temperature triggers an early warning.

[0098] Mode 2: Weighting under standard wear performance testing: w D =0.60, w f =0.30, w T=0.10. The depth of wear track under no-current conditions is the most direct indicator of failure, reflecting the volumetric wear of the material. At the same time, the friction curve reflects the changes in the state of the friction interface. When wear debris behavior and surface crack initiation occur, the amplitude will change significantly.

[0099] Mode 3: Weighting under intermittent current impact + friction and wear: w A =0.40, w D =0.30, w T =0.10, w f =0.10, w ΔI =0.05, w R =0.05. In the coupled damage of current impact and frictional wear, the ablation area is still the primary factor, but its weight decreases compared to mode 1. Friction and current coupling accelerate wear, and wear track depth becomes the second core indicator.

[0100] Mode 4: Weighting under continuous current + tribological wear test: w T =0.25, w D =0.25, w A =0.20, w R =0.15, w f =0.10, w ΔI =0.05. Under continuous current, Joule heat accumulates continuously, and the temperature rise is more significant than under intermittent conditions. Continuous current heating softens the material and accelerates wear; the weight of wear track depth is equal to that of temperature. Secondly, while the arc ablation effect is relatively weak under continuous current, contact resistance heat can still cause localized damage, so it is used as a second weighted data point. Simultaneously, interface degradation is highly sensitive under continuous current, and the rate of change in resistance is also an important criterion under this mode.

[0101] The weighting coefficients mentioned above are for reference only. During actual testing, they should be dynamically adjusted based on the test materials and operating conditions. The sum of all weighting coefficients should be 1. P A smart warning is triggered when the value exceeds 0.8. Each input parameter triggers an independent sensor termination criterion when it exceeds its corresponding physical threshold, without relying on weight calculations.

[0102] In summary, this invention provides an intelligent testing device and method for the mechanical properties of a metal matrix composite reinforcement layer. It constructs a test environment under extreme conditions of coupled electro-thermal-mechanical multi-fields, utilizes high-energy current to act on the matrix material containing the reinforcement layer, and achieves control over the current impact time, interval time, and number of impacts. It is also equipped with a loading device and an online monitoring device to realize loading and reciprocating motion under the action of current impact.

[0103] 1. This invention integrates the main circuit electric loading (electric field + Joule thermal field), mechanical loading (normal load + reciprocating friction) and online temperature field monitoring into one, and constructs a three-field extreme working condition test environment that can simultaneously couple electric field, temperature field and mechanical field. It overcomes the defects of existing devices that are mostly single or dual field coupling, and makes the test conditions closer to the actual service conditions of electromagnetic launch tracks, railway pantographs and other equipment.

[0104] 2. This invention employs multi-sensor fusion technology to synchronously collect temperature distribution, wear depth, arc ablation area ratio, vibration spectrum, and current magnitude in real time during the testing process. This enables dynamic online monitoring of the entire process of the reinforcement layer from intact to failure, and establishes a real-time mapping relationship between the number of current impacts and performance degradation, thus eliminating the reliance on offline detection methods.

[0105] 3. This invention can simultaneously perform reciprocating friction and wear tests while conducting current impact tests, achieving coordinated online detection of current impact resistance and wear resistance. By connecting the reinforcing layer and the metal substrate in series, it ensures that the current passes through both the reinforcing layer and the metal substrate. By setting damage thresholds such as the arc ablation area ratio and wear track depth, the ablation resistance and wear resistance of the reinforcing layer under intermittent current impact can be quantitatively evaluated, providing a dedicated testing tool for the engineering application of reinforcing layers in key components such as electromagnetic launch tracks.

[0106] 4. This invention embeds a pre-trained Long Short-Term Memory (LSTM) network model, using multi-sensor fusion data streams as input. Every N iterations, it outputs the remaining lifetime and the probability of reinforcement layer failure, and generates a performance degradation trend graph. This intelligent evaluation system achieves online damage assessment and remaining lifetime prediction based on multi-sensor data fusion, overcoming the limitations of existing systems that only monitor a single signal or require offline analysis, significantly improving the intelligence level of testing and its engineering practical value. Furthermore, by integrating a Long Short-Term Memory (LSTM) deep learning model, the device can predict the wear resistance lifetime of the reinforcement layer of the test sample based on online monitoring data.

[0107] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0108] It should be understood that those skilled in the art can make improvements or modifications based on the above description, and all such improvements and modifications should fall within the protection scope of the appended claims.

Claims

1. An intelligent testing device for the mechanical properties of a metal matrix composite reinforcement layer, characterized in that: It includes a mechanical loading device, a main circuit electric loading device and a control system; the mechanical loading device includes a motion control platform, a hard brush insulation fixing clamp (10) and an insulation sample placement platform (1), the insulation sample placement platform (1) is provided with a plurality of insulation sample fixing sliders (3) for clamping the sample placed on the insulation sample placement platform (1), and the sample is a metal substrate (23) with a surface-strengthened bonding layer (12). The motion control platform is arranged opposite to the insulating sample placement platform (1). The hard brush insulating fixing clamp (10) is connected to the motion control platform. The hard brush insulating fixing clamp (10) is connected to a hard brush (11). The hard brush (11) is used to make pressure contact with the reinforcing layer (12) at one end of the sample. The motion control platform is connected to the control system. The insulating sample placement platform (1) has a built-in conductive rod (22). One end of the conductive rod (22) passes through the insulating sample placement platform (1) and makes pressure contact with the metal substrate (23) at the other end of the sample. The two ends of the main circuit electric loading device are connected to the hard brush (11) and the conductive rod (22) respectively. The control system is connected to the main circuit electric loading device. An infrared thermal imager (4) and a laser vision sensor (5) are provided on one side of the sample placement position. A high-frequency vibration accelerometer (6) is provided on the hard brush insulating fixing fixture (10). The infrared thermal imager (4), the laser vision sensor (5), and the high-frequency vibration accelerometer (6) are connected to the control system.

2. The intelligent testing device for the mechanical properties of the metal matrix composite reinforced layer as described in claim 1, characterized in that: The main circuit electric loading device includes a high-power DC power supply (14), a high-speed electronic switch (13), a current sensor, and an oscilloscope (20). One end of the high-power DC power supply (14) is connected to a hard brush (11) through the high-speed electronic switch (13), and the other end is connected to a conductive rod (22) through the current sensor. The control terminal of the high-power DC power supply (14) is connected to the control system through the oscilloscope (20).

3. The intelligent testing device for the mechanical properties of the metal matrix composite reinforced layer as described in claim 1, characterized in that: The hard brush (11) includes a hard brush fixing head (111), a hard brush connecting rod (112), and a hard brush friction contact head (113). The hard brush fixing head (111) is connected to the hard brush friction contact head (113) through the hard brush connecting rod (112). The main circuit electric loading device is connected to the hard brush fixing head (111) wire. The hard brush friction contact head (113) is used to make pressure contact with the metal base reinforcement layer (12) of the sample.

4. The intelligent testing device for the mechanical properties of the metal matrix composite reinforced layer as described in claim 1, characterized in that: The motion control platform includes a normal loading mechanism, a lateral drive mechanism, and a motion controller (7). The motion controller (7) is connected to the normal loading mechanism and the lateral drive mechanism respectively. The hard brush insulation fixing fixture (10) is slidably connected to the normal loading mechanism. The normal loading mechanism is used to apply a normal force perpendicular to the friction contact surface of the sample to the hard brush insulation fixing fixture (10). The lateral drive mechanism is connected to the hard brush insulation fixing fixture (10). The lateral drive mechanism is used to drive the hard brush insulation fixing fixture (10) and the hard brush (11) to move back and forth laterally. The motion controller (7) is connected to the control system.

5. The intelligent testing device for the mechanical properties of the metal matrix composite reinforced layer as described in claim 4, characterized in that: The normal loading mechanism includes a normal loading motor (24) and a sliding guide rod (8). The output end of the normal loading motor (24) is connected to one end of the sliding guide rod (8), and the other end of the sliding guide rod (8) is slidably connected to the hard brush insulation fixing clamp (10). The normal loading motor (24) is connected to the motion controller (7). The lateral drive mechanism includes a lateral drive motor (15), a ball screw (25), and a screw nut (26). The output end of the lateral drive motor (15) is connected to one end of the ball screw (25), and the other end of the ball screw (25) is connected to one side of the hard brush insulation fixing clamp (10). The screw nut (26) is threaded onto the ball screw (25). The screw nut (26) is fixedly arranged. The lateral drive motor (15) is connected to the motion controller (7).

6. The intelligent testing device for the mechanical properties of the metal matrix composite reinforced layer as described in claim 5, characterized in that: The rigid brush insulation fixing fixture (10) is provided with a sliding guide rail (9), and the end of the sliding guide rod (8) is set on the sliding guide rail (9). The sliding guide rod (8) can move back and forth along the sliding guide rail (9) relative to the rigid brush insulation fixing fixture (10).

7. The intelligent testing device for the mechanical properties of the metal matrix composite reinforced layer as described in claim 1, characterized in that: The insulating sample placement platform (1) is provided with a slide groove (2), and the insulating sample fixing slider (3) is set on the slide groove (2). The insulating sample fixing slider (3) can move along the slide groove (2) on the insulating sample placement platform (1). The insulating sample placement platform (1) is used to place the sample, and the insulating sample fixing sliders (3) on both sides move along the slide groove (2) to clamp the sample.

8. The intelligent testing device for the mechanical properties of the metal matrix composite reinforced layer as described in claim 1, characterized in that: The control system includes a multi-sensor fusion data acquisition system and an intelligent detection system. The intelligent detection system is connected to the motion controller (7), infrared thermal imager (4), laser vision sensor (5), and high-frequency vibration accelerometer (6) through the multi-sensor fusion data acquisition system.

9. The intelligent testing device for the mechanical properties of the metal matrix composite reinforced layer as described in claim 8, characterized in that: The intelligent detection system includes a computer (19) and an alarm (18). The alarm (18) is connected to the computer (19). The computer (19) is connected to the high-power DC power supply (14) of the main circuit electric loading device. The computer is connected to the multi-sensor fusion data acquisition system. The multi-sensor fusion data acquisition system is a multi-modal data acquisition card (17).

10. A detection method using the intelligent detection device for the mechanical properties of the metal matrix composite reinforcement layer as described in claim 1, characterized in that: Includes the following steps: The test sample is fixed on the insulation sample placement platform (1), and the test sample is clamped by the insulation sample fixing slider (3). The hard brush (11) and the conductive rod (22) are respectively in pressure contact with the reinforcing layer (12) at the top of the sample and the bottom surface of the metal substrate (23) at the bottom. The main circuit electric loading device is electrically connected to the hard brush (11) and the conductive rod (22) to form a closed electric loading circuit. The motion control platform is started. The motion control platform applies a set normal load to the sample through the hard brush insulation fixing clamp (10) and the hard brush (11), so that the hard brush (11) is in pressure contact with the surface of the reinforcing layer of the sample. The hard brush (11) is driven by the motion control platform to form a reciprocating cyclic friction with the contact surface of the sample, and / or the output current of the sample is impacted by the main circuit electric loading device. The reciprocating cyclic friction between the hard brush (11) and the sample contact surface caused by the motion control platform refers to the motion control platform driving the hard brush insulation fixing fixture (10) to move the hard brush (11) back and forth laterally according to the set frequency, so that the hard brush (11) forms a reciprocating cyclic friction with the sample contact surface under normal load pressure, and the number of cyclic frictions is recorded. The impact of the main circuit electric loading device on the sample output current refers to the intermittent or continuous impact of the main circuit electric loading device on the sample output current. Intermittent current surge output is either once every N cycles of friction or once at a set interval t; N or t can be set before starting the motion control platform. The control system collects the sample temperature in real time through an infrared thermal imager (4), and collects the changes in the height position of the hard brush (11) and the friction surface image of the sample reinforcement layer in real time through a laser vision sensor (5). The control system obtains the wear depth of the sample reinforcement layer based on the changes in the height position of the hard brush (11), and obtains the ablation area ratio of the sample reinforcement layer based on the friction surface image of the sample reinforcement layer. The system collects the friction vibration amplitude between the hard brush (11) and the sample in real time through a high-frequency vibration accelerometer (6), and plots the friction vibration spectrum based on the changes in the friction vibration amplitude over time, thereby constructing a friction curve that changes with time. If one of the following two failure conditions is detected, the sample is judged to be in failure; otherwise, the hard brush (11) continues to rub against the contact surface of the sample until the sample is judged to be in failure. The control system terminates the test, triggers an alarm, and generates a sample performance degradation report after determining that the sample has failed. One of the failure conditions includes the satisfaction of any one of the following four conditions: The ablation area ratio reaches the first preset threshold. The wear depth reaches the second preset threshold, or the thickness of the reinforcement layer reaches the third preset threshold; The friction curve abrupt change factor reaches the fourth preset threshold; The sample temperature exceeds the melting point of the corresponding material. Another failure condition is: failure probability > 0.8, or the predicted number of residual current impact cycles ≤ 10% of the maximum current impact cycles; Failure probability Where Z is a weighted linear combination of the input features; A This represents the proportion of the arc-eroded area. w A This is the weighting coefficient for the arc ablation area scaling factor. D For wear depth, w D Here, represents the weighting coefficient of the wear depth factor, and T represents the sample temperature. w T This represents the weighting coefficient for the sample temperature factor. f The amplitude of frictional vibration. w f The weighting coefficients for the friction vibration amplitude factor. ΔI The deviation of the current amplitude of the main circuit electric loading device. w ΔI This is the weighting coefficient for the current amplitude deviation factor. R The contact resistance of the sample. R This is the weighting coefficient for the contact resistance factor. b For bias terms; The predicted number of residual current impact cycles is the maximum number of current impact cycles minus the number of current impact cycles already completed. The maximum number of current impact cycles is determined based on the different sample materials.

Images