Integrated experimental IGBT (Insulated Gate Bipolar Translator) device press-fitting clamp and use method

Abstract

The invention relates to an integrated experimental IGBT device press fitting clamp and a use method, the clamp comprises a lower pressing plate, a first clamping pair, a second clamping pair, an upper pressing plate, an IGBT device and a pressure applying assembly, the lower pressing plate and the upper pressing plate are fixedly connected through a double-thread screw to form a rigid stress frame; a first clamping pair is fixedly arranged above the lower pressing plate; a pressure applying assembly is fixedly arranged on the upper pressing plate; a second clamping pair is arranged on the four double-thread screws in an up-down sliding mode. And a pressure sensor is fixedly arranged on the second clamping pair. The heating table and the pressure sensor are arranged in the press-fitting clamp, quantitative analysis in IGBT device process verification, limiting working conditions and failure analysis experiments is achieved, and errors of experimental result analysis are reduced. The trapezoidal conduction block uniformly diffuses the point clamping force of the ball stud to the surface of the device, so that uniform stress is ensured, and damage to the device caused by local stress concentration is avoided; and the spherical contact structure of the ball stud can automatically compensate the angle deviation.

Description

Technical Field

[0001] This application relates to the field of press-fit fixture technology, specifically to an integrated experimental IGBT device press-fit fixture and its usage method. Background Technology

[0002] Insulated Gate Bipolar Transistors (IGBTs), as a new type of high-power semiconductor device, have been widely used in high-voltage direct current transmission systems, power electronic conversion equipment, and large motor drives. To ensure excellent thermal and electrical conductivity, IGBTs require significant clamping force during assembly to reliably fix and stably electrically connect core components such as the chip, electrodes, and heat sink. To achieve this, the industry commonly employs dedicated press-fit structures. A typical press-fit structure consists of a support base, upper and lower pressure plates, guide pillars, a screw or hydraulic loading unit, locating pins, insulating pads, copper busbars, heat sinks, ball head bolts, and trapezoidal pads. During press-fitting, the support base supports the device, while the lower pressure plate applies axial force driven by the screw or hydraulic device, ensuring reliable contact between the IGBT device and the heat sink. Locating pins ensure precise device positioning, insulating pads provide electrical isolation, copper busbars ensure current conduction, heat sinks enhance heat dissipation, and disc springs, ball head bolts, and trapezoidal pads compensate for assembly angle errors or uneven local stress.

[0003] While current IGBT press-fit structures can meet the press-fit requirements under normal operating conditions, they have the following shortcomings: 1. They cannot directly and accurately display the pressure borne by the press-fit IGBT: Experiments require verification of the static and dynamic parameters of the press-fit IGBT under different pressures. Existing press-fit structures can only apply pressure to the press-fit IGBT through a torque wrench or hydraulic loading unit. Therefore, the pressure borne by the IGBT can only be indirectly represented by the pressure gauge of the torque wrench or hydraulic loading unit, and cannot directly and accurately display the pressure borne by the press-fit IGBT. 2. They cannot heat the press-fit IGBT: In special application scenarios such as process verification, extreme condition experiments, and failure analysis, it is necessary to uniformly heat the press-fit IGBT. Existing press-fit structures do not have this function and cannot achieve heating of the press-fit IGBT. 3. They cannot perform quantitative analysis in special application scenarios such as verifying IGBT extreme conditions and failure analysis: Existing IGBT device fixtures cannot record the pressure and IGBT chip temperature under extreme conditions and failure, which causes certain errors in the analysis of experimental results.

[0004] In summary, the existing IGBT press-fit structure has functional limitations and cannot meet the requirements of directly and accurately displaying the pressure borne by the press-fit IGBT and heating the press-fit IGBT. Therefore, it urgently needs improvement.

[0005] Therefore, this invention proposes an integrated experimental IGBT device press-fit fixture and its usage method. Summary of the Invention

[0006] To overcome the shortcomings of the prior art, this application provides an integrated experimental IGBT device press-fit fixture, specifically adopting the following technical solution.

[0007] An integrated experimental IGBT device press-fit fixture includes a lower pressure plate, a first clamping pair, a second clamping pair, an upper pressure plate, an IGBT device, a pressure sensor, and a pressure application assembly.

[0008] The lower and upper pressure plates are fixedly connected by four symmetrically distributed double-headed screws to form a rigid load-bearing frame.

[0009] The first clamping assembly is detachably fixed above the lower pressure plate.

[0010] The upper pressure plate is detachably fixed with a pressure application component.

[0011] A second clamping assembly is slidably mounted on each of the four double-ended screws. A pressure sensor is fixedly mounted on the second clamping assembly.

[0012] In this technical solution, the IGBT device to be tested is placed on top of a first mounting assembly, and a second mounting assembly is placed on top of the IGBT device. Pressure is applied to the second mounting assembly by a pressure-applying component, thereby applying pressure to the IGBT device for experimental testing. The pressure on the IGBT device is detected by a pressure sensor.

[0013] This invention provides a method for using an integrated experimental IGBT device press-fit fixture, comprising the following steps.

[0014] S1. Using a replaceable positioning sleeve adapted to the IGBT device model to be tested, place the IGBT device above the first copper busbar, ensuring precise contact between the IGBT device electrodes and the conductive area of ​​the copper busbar, avoiding the installation location of the temperature sensor. The adaptive calibration module calculates the calibration amount based on the ratio of the rated voltage and rated current of the current experimental IGBT device to the average of common IGBT models, combined with preset calibration coefficients. The calibration amount is then used to adjust the basic weights of input variables such as pressure, temperature, and image features in the model.

[0015] S2. The PLC control module controls four electric push rods to extend and retract synchronously, driving the second clamping pair to move downward along the double-headed screw. The displacement of the support cylinder of the second clamping pair is monitored in real time using a laser rangefinder. When the lower end face of the second copper busbar is precisely aligned with the upper end face of the IGBT device, the electric push rod stops working. Then, the moving rod of the electric push rod continues to move downward by a preset stroke to reserve space for subsequent pressure loading. Next, the high-voltage insulation monitoring module is activated to detect the insulation resistance of the first and second insulation blocks. If the insulation resistance is ≤ 80% of the preset threshold, an audible and visual alarm is triggered and the experiment is paused to check for insulation faults. If the insulation resistance is ≥ the preset threshold, the process proceeds to the next step.

[0016] S3. Set the target temperatures of the first and second heating stages according to the experimental requirements, and enable the dual over-temperature protection of the temperature controller; start the first and second heating stages, and collect the surface temperature of the IGBT device in real time through the surface-mount PT100 temperature sensor. Use the multi-point acquisition fusion algorithm to capture the surface temperature field distribution of the device and feed it back to the temperature controller to form a closed-loop control; after the temperature stabilizes at the target value, maintain preheating for ten to fifteen minutes to ensure that the overall temperature of the device is uniform.

[0017] S4. Input the target pressure value, loading speed, and holding time into the PLC control module, and set the secondary pressure protection threshold. Start the servo motor, which, after torque amplification by the planetary reducer, drives the hexagonal head drive rod to move the ball head bolt downwards at a uniform speed along the guide post threaded hole. The universal joint compensates for coaxiality deviation, and the guide sleeve ensures transmission stability. The ball head bolt drives the trapezoidal transmission block downwards through the ball joint, spreading the point load along the conical surface into a uniform surface load, compressing the second insulating block. The pressure is sequentially transmitted to the pressure sensor, the second clamping pair, and the IGBT device to achieve uniform pressing. The pressure sensor collects pressure data in real time and feeds it back to the PLC control module. When the actual pressure reaches the target value, the PLC control module instructs the motor to stop rotating, and at the same time, the electromagnetic brake device locks the drive rod to prevent pressure rebound.

[0018] S5. Synchronously acquire pressure data, temperature data, and high-definition images. Combined with IGBT device performance parameters monitored by external equipment, the data acquisition instrument achieves millisecond-level synchronization and binds a unified timestamp for storage. This constructs an integrated analysis system for multi-dimensional data fusion, dynamic correlation analysis, and quantitative evaluation, enabling comprehensive assessment and anomaly identification. Pressure data undergoes Kalman filtering preprocessing to remove abnormal data points. Anomaly identification includes pressure and temperature exceeding limits, device appearance abnormalities, or performance parameter deviations. If an anomaly is detected, an emergency response is immediately triggered, stopping loading and heating and initiating a servo motor to reverse and release pressure, ensuring experimental safety.

[0019] S6. During the preset pressure holding time, the electromagnetic brake remains locked, the pressure sensor monitors the pressure stability in real time, and the data acquisition instrument continuously records the pressure-time and temperature-time fluctuation curves. After the pressure holding period ends, the pressure-time curve, temperature-time curve, device surface image sequence, and other data are exported for subsequent performance analysis.

[0020] S7. After the experiment, the PLC control module instructs the servo motor to reverse, driving the ball head bolt upward to release the pressure load; the electromagnetic brake is unlocked; the first and second heating stages are turned off, and the residual heat is quickly discharged using the first and second heat sinks; after the device temperature drops to room temperature, the electric push rod is controlled to move upward, driving the second clamping pair away from the device; the IGBT device after the experiment is taken out, completing the press-fit experiment process.

[0021] As an optional embodiment of the present invention, the first clamping assembly includes a first insulating block, a first heat sink, a first heating platform, a first insulating heat-conducting block, and a first copper busbar.

[0022] The first insulating block is detachably fixed above the lower pressure plate. The first heat sink is detachably fixed above the first insulating block. The first heating platform is detachably fixed above the first heat sink. The first insulating heat-conducting block is detachably fixed above the first heating platform. The first copper busbar is detachably fixed above the first insulating heat-conducting block.

[0023] As an optional embodiment of the present invention, the second clamping assembly includes a second insulating block, a second heat dissipation block, a second heating platform, a second insulating heat-conducting block, and a second copper busbar.

[0024] A second insulating heat-conducting block is detachably fixed above the second copper busbar. A second heating platform is detachably fixed above the second insulating heat-conducting block. A second heat sink is detachably fixed above the second heating platform. A pressure sensor is detachably fixed above the second heat sink. A second insulating block is detachably fixed above the pressure sensor.

[0025] A support cylinder is fixedly installed on the outside of the second insulating block, the second heat dissipation block, the second heating platform, the second insulating heat-conducting block, and the second copper busbar, and the second copper busbar is fixedly installed inside the support cylinder.

[0026] The support cylinder is slidably mounted on four double-ended screws. Four electric push rods are fixedly mounted on the upper pressure plate. The movable rods of the electric push rods slide in conjunction with the support cylinder, and a support plate is detachably fixed to the lower end of the movable rod. A laser rangefinder is fixedly mounted on the support cylinder. A shock-absorbing rubber pad is fixedly mounted above the support plate.

[0027] As an optional embodiment of the present invention, the pressure application assembly includes a trapezoidal transmission block, a ball head bolt, a guide post, and a pad sleeve.

[0028] The guide post is detachably fixed to the upper pressure plate. A lower limit block is fixedly installed at the lower end of the guide post. A butterfly spring is installed above the lower limit block, and a pad sleeve is installed above the butterfly spring. The pad sleeve and the butterfly spring are sleeved on the outside of the guide post. An annular groove is opened on the guide post. A positioning block is detachably fixed to the upper pressure plate. The guide post is fixed by the positioning block and the annular groove.

[0029] The lower limit block is detachably and fixedly connected to the upper pressure plate. The outer diameter of the lower limit block is larger than the outer diameter of the pad sleeve and the disc spring. The guide post is located directly above the trapezoidal conduction block. A circular recess is provided at the center of the trapezoidal conduction block.

[0030] The guide post is threaded with a ball-end bolt, which protrudes from the center hole of the guide post from top to bottom. A trapezoidal conductive block is mounted on the lower end of the ball-end bolt via a ball joint. The lower surface of the trapezoidal conductive block is a large-area plane, while the upper end is a small-area plane. The ball joint diffuses the point load generated by the ball-end bolt into a uniformly distributed surface pressure.

[0031] The technical solution of this application has achieved the following beneficial effects.

[0032] 1. Arrange a heating stage and pressure sensor in the press-fit fixture to achieve quantitative analysis in IGBT device process verification, extreme condition and failure analysis experiments, and reduce the error of experimental result analysis.

[0033] 2. The trapezoidal conductive block evenly distributes the clamping force of the ball head bolt to the surface of the device, ensuring uniform force distribution and avoiding damage to the device caused by local stress concentration; the spherical contact structure of the ball head bolt can automatically compensate for angular deviations.

[0034] 3. A dynamic correlation model of pressure-temperature-performance-appearance is constructed based on the random forest algorithm. Combined with the adaptive correction module, the model weights can be automatically adjusted according to the rated voltage and current parameters of the IGBT device, adapting to the characteristic differences of different models of devices, and reducing the model prediction error.

[0035] 4. Adding a pressure sensor between the insulating block and the heat sink allows for real-time monitoring and recording of the pressure applied to the IGBT device, improving the data repeatability of repeated experiments. Adding a temperature-controlled heating stage to both ends of the IGBT device ensures that the IGBT device is uniformly heated during IGBT device process verification, extreme condition and failure analysis experiments, ensuring a constant experimental temperature and improving the rigor of the experiment. Attached Figure Description

[0036] Figure 1 This is an overall schematic diagram of the experimental IGBT device press-fit fixture integrated in the embodiments of this application.

[0037] Figure 2This is a schematic diagram of the pressure application component of the experimental IGBT device press-fit fixture integrated in the embodiments of this application.

[0038] Figure 3 This is a schematic diagram of the positioning pins of the experimental IGBT device press-fit fixture integrated in the embodiments of this application.

[0039] Figure 4 This is an assembly diagram of the experimental IGBT device press-fit fixture integrated in the embodiments of this application.

[0040] Reference numerals: 100, lower pressure plate; 110, positioning hole; 120, first threaded hole; 200, first clamping pair; 210, first insulating block; 220, first heat sink; 230, first heating platform; 240, first insulating heat-conducting block; 250, first copper busbar; 300, second clamping pair; 310, second insulating block; 320, second heat sink; 330, second heating platform; 340, second insulating heat-conducting block; 350, second copper busbar; 400, Upper pressure plate; 410, Second threaded hole; 420, Nut; 500, IGBT device; 600, Pressure sensor; 700, Pressure application assembly; 710, Trapezoidal transmission block; 711, Circular recess; 720, Ball head bolt; 730, Guide post; 731, Center hole; 732, Lower limit block; 740, Butterfly spring; 750, Pad sleeve; 800, Double-ended screw; 810, Lower end; 820, Upper end; 900, Positioning pin. Detailed Implementation

[0041] The present application will now be further described with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the present application and should not be construed as limiting the scope of protection of the present application. It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of the present application.

[0042] like Figure 1 As shown, the present invention discloses an integrated experimental IGBT device press-fit fixture, comprising: a lower pressure plate 100, a first clamping pair 200, a second clamping pair 300, an upper pressure plate 400, an IGBT device 500, a pressure sensor 600, and a pressure application assembly 700.

[0043] The lower pressure plate 100 and the upper pressure plate 400 are fixedly connected by four symmetrically distributed double-ended screws 800, forming a rigid load-bearing frame. The lower pressure plate 100 and the upper pressure plate 400 are made of No. 45 steel and have undergone quenching and tempering treatment (hardness HRC28-32) to ensure rigidity and wear resistance. The double-ended screws 800 are made of 40Cr steel and have a galvanized surface for rust prevention.

[0044] A first clamping assembly 200 is detachably fixed above the lower pressure plate 100.

[0045] The upper pressure plate 400 is detachably fixed with a pressure application component 700.

[0046] A second clamping assembly 300 is slidably mounted on each of the four double-ended screws 800. A pressure sensor 600 is fixedly mounted on the second clamping assembly 300.

[0047] In this technical solution, the IGBT device 500 to be tested is placed above the first clamping assembly 200, and the second clamping assembly 300 is placed above the IGBT device 500. Pressure is applied to the second clamping assembly 300 by the pressure application component 700, thereby applying pressure to the IGBT device 500 for experimental testing. The pressure on the IGBT device 500 is detected by the pressure sensor 600.

[0048] This invention provides a method for using an integrated experimental IGBT device press-fit fixture, comprising the following steps.

[0049] S1. Place the IGBT device 500 to be tested above the first copper busbar 250, aligning it with the positioning pin, ensuring that the electrodes of the IGBT device 500 are precisely in contact with the conductive area of ​​the copper busbar, avoiding the installation position of the temperature sensor.

[0050] S2. The PLC control module controls the synchronous extension and retraction of four electric push rods, driving the second clamping pair 300 to move downward along the double-headed screw 800; the laser rangefinder monitors the displacement of the support cylinder of the second clamping pair 300 in real time. When the lower end face of the second copper busbar 350 is precisely in contact with the upper end face of the IGBT device 500 (the displacement reaches the preset value, adjusted according to the thickness of the IGBT device 500), the electric push rod stops working; then the moving rod of the electric push rod continues to move downward by the preset stroke, reserving space for subsequent pressure loading.

[0051] S3. Set the target temperatures of the first heating stage 230 and the second heating stage 330 according to the experimental requirements. Start the first heating stage 230 and the second heating stage 330, and collect the surface temperature of the IGBT device 500 in real time through the surface-mount PT100 temperature sensor, and feed it back to the temperature controller to form a closed-loop control; after the temperature stabilizes at the target value (fluctuation ≤ ±2℃), maintain preheating for ten to fifteen minutes to ensure that the overall temperature of the device is uniform.

[0052] S4. Input the target pressure value, loading speed, and holding time into the PLC control module; start the servo motor to drive the hexagonal head drive rod to move the ball head bolt 720 downwards at a uniform speed along the threaded hole of the guide post 730. The ball head bolt 720 drives the trapezoidal transmission block 710 downwards through the ball joint, spreading the point load into a uniform surface load along the conical surface, pressing the second insulating block 310; the pressure is sequentially transmitted to the pressure sensor, the second clamping pair 300, and the IGBT device 500, achieving uniform pressing. The pressure sensor collects pressure data in real time and feeds it back to the PLC control module. When the actual pressure reaches the target value, the PLC control module instructs the motor to stop rotating, and at the same time, the electromagnetic brake device locks the drive rod to prevent pressure rebound.

[0053] S5. Simultaneously acquire pressure data, temperature data, and high-definition images. Combined with IGBT device performance data monitored by external equipment, construct an integrated analysis system for multi-dimensional data fusion, dynamic correlation analysis, and quantitative evaluation for comprehensive assessment and anomaly identification. If pressure or temperature exceeds limits or device appearance abnormalities (cracks, deformation) are detected, an emergency response is immediately triggered to stop loading and heating, ensuring experimental safety.

[0054] S6. During the preset pressure holding time, the electromagnetic brake remains locked, the pressure sensor monitors the pressure stability in real time, and the data acquisition instrument continuously records the pressure and temperature fluctuation curves.

[0055] After the pressure holding period is completed, data such as pressure-time curves, temperature-time curves, and device surface image sequences are exported for subsequent performance analysis.

[0056] S7. After the experiment, the PLC control module instructs the servo motor to reverse, driving the ball head bolt upwards to release the pressure load; the electromagnetic brake is then unlocked. The first heating stage 230 and the second heating stage 330 are shut off, and the first heat sink 220 and the second heat sink 320 quickly dissipate residual heat. After the device temperature drops to room temperature, the electric push rod is controlled to move upwards, driving the second clamping pair away from the device. The IGBT device is then removed from the experiment, completing the press-fit experiment process.

[0057] Example 1.

[0058] This embodiment 1 discloses an integrated experimental IGBT device press-fit fixture, referring to... Figure 1 and Figure 4 The first clamping assembly 200 includes a first insulating block 210, a first heat sink 220, a first heating platform 230, a first insulating heat-conducting block 240, and a first copper busbar 250.

[0059] The first insulating block 210 is detachably fixed above the lower pressure plate 100 to provide electrical isolation and prevent short circuits during the pressing process. A first heat sink 220 is detachably fixed above the first insulating block 210, a first heating platform 230 is detachably fixed above the first heat sink 220, a first insulating heat-conducting block 240 is detachably fixed above the first heating platform 230, and a first copper busbar 250 is detachably fixed above the first insulating heat-conducting block 240.

[0060] The first insulating block 210 is made of aluminum nitride ceramic with a thermal conductivity ≥180W / (mK) and a breakdown voltage ≥15kV / mm; it achieves electrical isolation between the lower pressure plate 100 and subsequent conductive and heating components, preventing short circuits during high-voltage experiments. The first heat sink 220 is made of 6061 aluminum alloy with an anodized surface treatment to enhance wear resistance; it quickly dissipates residual heat after the first heating platform 230 stops heating, preventing overheating damage to the IGBT device 500.

[0061] The first heating stage 230 uses a customized ceramic heating element with a heating power of 500W. Its surface is coated with thermal grease to enhance heat transfer efficiency. An aviation connector (model XLR-3P) leads to the power cord, connecting to an external temperature controller (model RKC REX-C100) for closed-loop temperature control. The first insulating heat-conducting block 240 is made of alumina ceramic and coated with high-temperature insulating thermal grease to ensure uniform heat transfer to the first copper busbar 250. The first copper busbar 250 is made of T2 copper with a silver-plated surface (thickness ≥2μm) to reduce contact resistance. Symmetrically welded copper terminals on both sides are used to connect experimental test circuits.

[0062] In this technical solution, the IGBT device 500 is heated by the first heating stage 230. The first insulating block 210, the first heat sink 220, the first heating stage 230, the first insulating heat-conducting block 240, and the first copper busbar 250 are all coaxially arranged.

[0063] Example 2.

[0064] This embodiment 2 discloses an integrated experimental IGBT device press-fit fixture, wherein the second clamping pair 300 includes a second insulating block 310, a second heat sink 320, a second heating stage 330, a second insulating heat-conducting block 340, and a second copper busbar 350.

[0065] A second insulating heat-conducting block 340 is detachably fixed above the second copper busbar 350. A second heating platform 330 is detachably fixed above the second insulating heat-conducting block 340. A second heat sink 320 is detachably fixed above the second heating platform 330. A pressure sensor 600 is detachably fixed above the second heat sink 320. A second insulating block 310 is detachably fixed above the pressure sensor 600.

[0066] A support cylinder is fixedly installed on the outside of the second insulating block 310, the second heat dissipation block 320, the second heating platform 330, the second insulating heat-conducting block 340, and the second copper busbar 350, and the second copper busbar 350 is fixedly installed inside the support cylinder.

[0067] The support cylinder is slidably mounted on four double-ended screws 800. Four electric push rods are fixedly mounted on the upper pressure plate 400. The movable rods of the electric push rods slide in conjunction with the support cylinder, and a support plate is detachably fixed to the lower end of the movable rod. A laser rangefinder is fixedly mounted on the support cylinder. The support cylinder is made of 45# steel, heat-treated, and has four guide grooves machined on its inner wall. These grooves mate with the guide bosses on the outer side of the double-ended screws 800, with a clearance of 0.05-0.1mm, ensuring only vertical sliding and no horizontal movement or rotation. Three L-shaped brackets (made of Q235 steel) are welded internally to fix the second copper busbar 350; two lugs are welded externally for mounting the laser rangefinder. The function of the support plate is that after the experiment, activating the electric push rods moves the support plate upwards, which in turn moves the second clamping assembly 300 upwards, facilitating the loading and unloading of the IGBT device 500. A shock-absorbing rubber pad is fixedly mounted above the support plate.

[0068] The second copper busbar 350 is symmetrical to the first copper busbar 250, made of T2 copper, and fixed to an L-shaped bracket inside the support cylinder with hexagonal bolts. The second insulating heat-conducting block 340 is made of alumina ceramic and is coaxially fixed to the second copper busbar 350 with locating pins, and coated with high-temperature insulating thermal grease. The second heating platform 330 uses a 500W ceramic heating element and is fixed above the second insulating heat-conducting block 340 with bolts. An aviation connector leads out a control line to link with an external temperature controller, enabling synchronous heating or independent heating. The second heat sink 320 is made of 6061 aluminum alloy with an annular heat dissipation groove and is fixed to the second heating platform 330 with bolts. The pressure sensor 600 is a spoke-type pressure sensor, model HBM C9B, with a range of 0-5000N and an accuracy of 0.1 grade. It is fixed between the second heat sink 320 and the second insulating block 310 with hexagonal bolts (with matching anti-loosening nuts), and stainless steel gaskets are installed on the upper and lower surfaces to prevent uneven stress on the sensor. The pressure sensor 600 signal line is connected to an external data acquisition instrument (model NI cRIO-9039) via a shielded twisted-pair cable to display and record pressure data in real time. The second insulating block 310 is made of aluminum nitride ceramic and is coaxially fixed to the pressure sensor 600 by a locating pin, achieving electrical isolation between the pressure sensor and the pressure application component 700. A laser displacement sensor is fixedly mounted on the second insulating block 310.

[0069] In this technical solution, an electric actuator synchronously extends and retracts, causing the support cylinder to move downwards, positioning the second copper busbar 350 directly above the IGBT device 500. The movable rod of the electric actuator then continues to move downwards, reserving space for applying pressure. The IGBT device 500 is heated by the second heating platform 330, and the pressure on the IGBT device 500 is monitored by a pressure sensor 600. The second insulating block 310, the second heat sink 320, the second heating platform 330, the second insulating heat-conducting block 340, and the second copper busbar 350 are all coaxially arranged. The electric actuator is a DTZ300 model, with a thrust of 300N and encoder feedback. A PLC controller (Siemens S7-200 SMART) is used to achieve synchronous extension and retraction of four electric actuators with a synchronization accuracy of ≤0.1mm. A laser rangefinder (Keyence IL-600, accuracy ±0.01mm) monitors the displacement of the support cylinder in real time. When the displacement reaches a preset value (adjusted according to the IGBT device thickness), the actuator stops moving, ensuring that the second copper busbar 350 is precisely aligned with the upper end of the IGBT device 500. Then, the movable rod of the electric actuator continues to move downward, reserving a travel range for the subsequent pressure loading of the pressure application component 700 and avoiding interference with pressure transmission. In special application scenarios such as process verification, extreme condition testing, and failure analysis of the IGBT device 500, the first heating platform 230 and the second heating platform 330 are used to heat the IGBT device 500, and a fixed heating temperature can be set. During the experiment, the first insulating heat-conducting block 240 and the second insulating heat-conducting block 340 can effectively conduct heat to the IGBT device 500. The second insulating heat-conducting block 340 has sufficient thickness to prevent the high voltage from damaging the low-voltage withstand components in the first heating stage 230, the second heating stage 330, and the pressure sensor 600 under high-voltage experimental conditions. The first heating stage 230 and the second heating stage 330 use aviation plugs to lead out power lines and connect to an external temperature controller to achieve closed-loop temperature control. The temperature controller is set with an over-temperature protection threshold of 210℃, automatically cutting off the power to the heating stage when the threshold is reached. Simultaneously, a thermal fuse (rated temperature 220℃) is installed on the surface of the heating stage, achieving dual over-temperature protection through both software and hardware.

[0070] Example 3.

[0071] Reference Figure 1 , Figure 2 and Figure 4 This embodiment 3 discloses an integrated experimental IGBT device press-fit fixture, wherein the pressure application component 700 includes a trapezoidal conductive block 710, a ball head bolt 720, a guide post 730, and a pad sleeve 750.

[0072] The guide post 730 is detachably fixed to the upper pressure plate 400. A lower limit block 732 is fixedly installed at the lower end of the guide post 730. A butterfly spring 740 is installed above the lower limit block 732, and a pad sleeve 750 is installed above the butterfly spring 740. The pad sleeve 750 and the butterfly spring 740 are fitted onto the outside of the guide post 730. An annular groove is formed on the guide post 730. A positioning block is detachably fixed to the upper pressure plate 400, and the guide post 730 is fixed by the positioning block and the annular groove. The guide post 730 is made of 40Cr material and has undergone surface quenching treatment (hardness HRC55-60).

[0073] The lower limit block 732 is detachably fixed to the upper pressure plate 400. The outer diameter of the lower limit block 732 is larger than the outer diameter of the pad sleeve 750 and the disc spring 740. The guide post 730 is located directly above the trapezoidal transmission block 710.

[0074] A circular recess 711 is provided at the center of the trapezoidal conductive block 710.

[0075] The material of the disc spring is 65Mn, the material of the 750 pad is 45 steel, and the 740 disc spring provides elastic cushioning to avoid rigid impact during pressure loading.

[0076] A ball-end bolt 720 is threaded onto the guide post 730. The ball-end bolt 720 passes through the central hole 731 of the guide post 730 from top to bottom. A trapezoidal conductive block 710 is connected to the lower end of the ball-end bolt 720 via a ball joint. The lower surface of the trapezoidal conductive block 710 is a large-area plane, and the upper end is a small-area plane. The ball joint diffuses the point load generated by the ball-end bolt 720 into a uniformly distributed surface pressure, thereby maintaining reasonable stress on the IGBT device 500. During loading, angular errors are automatically compensated to further ensure the uniformity of the press-fit force.

[0077] The ball head bolt is made of 40Cr material with a polished surface (Ra≤0.2μm). The ball joint uses a miniature ball joint bearing with an IP54 sealing rating and is filled with high-temperature grease (temperature resistance 150℃). The ball joint is embedded and fixed in the circular recess 711 of the trapezoidal transmission block 710 (interference fit H7 / p6). The ball head of the ball head bolt 720 mates with the inner ring of the ball joint to achieve 360° omnidirectional angle compensation (compensation range ±5.0°).

[0078] In this technical solution, during loading, the operator uses an Allen wrench to rotate the ball head bolt 720. The bolt moves downward along the threaded hole of the guide post 730, driving the trapezoidal transmission block 710 to press the second insulating block 310 through the ball joint. The pressure is sequentially transmitted to the pressure sensor 600, the second clamping pair 300, and the IGBT device 500. The pressure sensor collects pressure data in real time and feeds it back to the data acquisition instrument. When the preset pressure value is reached, loading stops. The torque wrench can be set with a torque threshold to assist in pressure control. The pressure borne by the IGBT device 500 can be read in real time through the pressure sensor 600. The trapezoidal transmission block 710 is made of 45# steel and has undergone heat treatment. A recess is machined in the center of the upper surface for embedding the ball joint; the roughness Ra of the lower surface is ≤0.8μm, coated with high-temperature grease, and it fits against the upper surface of the second insulating block 310. The concentrated load of the ball head bolt 720 is gradually diffused into a uniform surface load along the conical surface, improving the pressure distribution uniformity by ≥30%; in conjunction with the ball joint, it realizes automatic compensation of angular error during the loading process, and offsets the coaxiality deviation during assembly or loading.

[0079] Example 4.

[0080] like Figure 1 , Figure 3 and Figure 4 As shown, this embodiment 4 discloses an integrated experimental IGBT device press-fit fixture. A positioning hole 110 is provided at the center of the lower pressure plate 100. A positioning pin 900 is inserted into the positioning hole 110 and fixed to the central axis of the press-fit fixture.

[0081] The first insulating block 210, the first heat sink 220, the first heating platform 230, the first insulating heat-conducting block 240, the first copper busbar 250, the second heat sink 320, the second heating platform 330, the IGBT device 500, the second insulating heat-conducting block 340, and the second copper busbar 350 are provided with round holes that are adapted to the positioning pins 900.

[0082] In this technical solution, after the positioning pin 900 is installed, its upper end extends 5 mm beyond the first copper busbar 250, and the IGBT device 500 is positioned by the positioning pin 900. A replaceable positioning sleeve is detachably fixed to the upper end of the positioning pin 900, improving the versatility of the fixture by replacing the positioning sleeve.

[0083] Furthermore, a temperature sensor is fixedly mounted above the first copper busbar 250 and below the second copper busbar 350, avoiding the electrode pins; it is fixed with high-temperature thermally conductive adhesive to ensure a tight fit between the sensor and the device surface, reducing temperature measurement errors. Shielded high-temperature wires are used for the sensor leads to avoid contact with conductive components such as the copper busbars, ensuring electrical isolation. The temperature sensor is a surface-mount PT100 temperature sensor, which can be directly attached to the IGBT device surface, with a temperature measurement range covering -50℃ to 200℃, suitable for the experimental temperature range.

[0084] Example 5.

[0085] This embodiment discloses an integrated experimental IGBT device press-fit fixture. A servo motor is fixedly mounted above the upper pressure plate 400 via a flange bracket. The motor output shaft is connected to a hexagonal head drive rod via a coupling. The hexagonal head drive rod is adapted to the hexagonal groove at the upper end of the ball head bolt 720. A guide sleeve and a universal joint are provided between the hexagonal head drive rod and the upper pressure plate 400 to ensure stable transmission even when the coaxiality deviation between the hexagonal head drive rod and the ball head bolt is ≤0.1mm. At the same time, an electromagnetic brake device is integrated between the motor and the reducer to prevent torque loss during loading, which could lead to pressure fluctuations.

[0086] In this technical solution, a servo motor drives a hexagonal head drive rod to rotate the ball head bolt 720, replacing traditional manual operation. After torque amplification via a planetary reducer, the motor drives the ball head bolt 720 to move uniformly downwards along the threaded hole of the guide post 730. Through a ball joint, the trapezoidal transmission block 710 presses against the second insulating block 310, and the pressure is sequentially transmitted to the pressure sensor 600, the second clamping pair 300, and the IGBT device 500. The pressure sensor collects pressure data in real time and feeds it back to the PLC controller, forming a closed-loop control with the preset pressure value. When the actual pressure reaches the preset value, the PLC issues a command to stop the motor rotation, and simultaneously, the electromagnetic brake locks the drive rod to prevent pressure rebound. The entire loading process can be automated and precise, with loading speed and target pressure value set via a touchscreen, requiring no manual intervention. Torque amplification via a planetary reducer (reduction ratio 1:50) ensures stable loading torque and avoids pressure fluctuations. A 400W servo motor (model: Panasonic MSMF042L1U2M) with a built-in absolute encoder is selected to meet the closed-loop control requirements of pressure loading. The servo drive is set to an overload current threshold of 1.5 times the rated current. When overloaded, it will stop immediately and trigger an audible and visual alarm, and record the fault code. The hexagonal drive rod is made of 40Cr material. A universal joint (model: WS20) is installed in the drive rod to compensate for the slight coaxiality deviation between the drive rod and the ball head bolt during assembly. A guide sleeve is installed on the outside of the hexagonal drive rod to ensure that there is no radial movement during transmission. The motor integrates an electromagnetic brake device (power-off braking). After the target pressure is applied, the motor is powered off and locked by the brake, which can maintain the pressure stability for a long time and solve the problem of easy loosening during long-term pressure holding in manual operation. The pressure sensor collects pressure data in real time and feeds it back to the PLC control module to form a closed-loop control with the preset pressure value. At the same time, the PLC control module sets a pressure two-level protection: (1) the warning threshold is 4800N, triggering an audible and visual alarm; (2) the shutdown threshold is 5200N, immediately cutting off the motor power supply and locking the drive rod with the electromagnetic brake to avoid pressure overload.

[0087] Furthermore, the double-ended screw 800 has a lower end 81 and an upper end 820 fixedly installed at both ends; the lower pressure plate 100 has four first threaded holes 120, and the lower end 81 of the double-ended screw 800 is threadedly connected to the first threaded holes 120. The upper pressure plate 400 has four second threaded holes 410, and the upper end 820 of the double-ended screw 800 is threadedly connected to the second threaded holes 410. A nut 420 is threadedly installed on the upper end 820 of the double-ended screw 800. A limit block 830 is fixedly installed at the lower end of the double-ended screw 800.

[0088] Furthermore, a high-definition camera is fixedly installed on the lower outer side of the upper pressure plate 400, and the high-definition camera is aimed at the IGBT device 500 to capture high-definition images. A PLC control module and a temperature controller are fixedly installed on the upper pressure plate 400.

[0089] In this technical solution, a high-definition camera data cable is connected to a PLC control module. The camera's focal length and exposure parameters (exposure time adjustable from 1 / 100s to 1 / 50s) are adjusted, and the image acquisition frequency is set to 10fps, synchronized with the pressure and temperature acquisition frequencies. A linkage program is written through the PLC control module to bind the timestamps of pressure, temperature, and image data, ensuring synchronized data recording. During the experiment, the high-definition camera continuously acquires high-definition images of the IGBT device 500, which are synchronously stored in the database along with the pressure, temperature, and device test data. If the PLC control module detects an abnormality in the IGBT device through an image recognition algorithm (preset crack and deformation thresholds), it immediately triggers an emergency stop command, stopping pressure loading and heating to protect the experimental sample and equipment. A high-temperature resistant transparent protective cover (temperature resistant to 250℃) is installed at the lens of the high-definition camera to prevent thermal grease splashing and high-temperature airflow from affecting image quality.

[0090] Furthermore, miniature insulation monitoring electrodes, gold-plated electrodes with an area of ​​one square centimeter, are respectively disposed on the upper and lower surfaces of the first insulating block 210 and the second insulating block 310, and are attached to the surfaces of the first insulating block 210 and the second insulating block 310. The electrodes are connected to a high-voltage insulation monitoring module (model: FLUKE 1555) via a shielded high-voltage wire, with a measurement range of 100MΩ-1TΩ and an adjustable test voltage of 0-10.0kV. The high-voltage insulation monitoring module can detect the insulation status in real time and provide early warning of high-voltage breakdown risks, making it particularly suitable for high-voltage IGBT (≥1200V) experimental scenarios.

[0091] Early warning logic: Before the experiment, a preset insulation resistance threshold is established (≥1000MΩ, adjusted according to the experimental voltage). During the experiment, the insulation monitoring module collects insulation resistance data once per second and feeds it back to the PLC control module. If the insulation resistance is ≤80% of the threshold, an audible and visual warning is triggered; if it is ≤50% of the threshold, heating and pressure loading are immediately cut off, and emergency pressure relief is initiated (servo motor reverses to unload). Insulation resistance data is stored synchronously with temperature and pressure data, allowing analysis of the impact of temperature-pressure coupling on insulation performance and providing data support for IGBT high-voltage reliability design.

[0092] Example 6.

[0093] This embodiment discloses a method for using an integrated experimental IGBT device press-fit fixture, wherein step S5, comprehensive evaluation and anomaly identification, includes the following steps.

[0094] S51. Synchronous Multi-Source Data Acquisition: The PLC control module controls pressure sensors, temperature sensors, and a high-definition camera to simultaneously acquire three-dimensional data of the IGBT device 500: pressure, temperature, and high-definition images. For high-pressure experimental scenarios, an anti-interference filtering module is set at the sensor signal end. Pressure data is preprocessed using a Kalman filter algorithm, then displayed and recorded in real time. Temperature data is accurately captured using a multi-point acquisition and fusion algorithm to capture the surface temperature distribution of the device. The high-definition camera acquires images of the IGBT device surface through a high-temperature resistant transparent protective cover (temperature resistant to 250℃, with an anti-fog coating), focusing on monitoring appearance anomalies such as crack initiation and encapsulation deformation. This solves the problem of inaccurate direct display of IGBT pressure in existing technologies and further improves data reliability under extreme experimental conditions through anti-interference and data preprocessing optimization.

[0095] S52. Performance Data Acquisition and Synchronization Verification: Start the oscilloscope (with a high-voltage differential probe / current probe), power analyzer, multimeter, and other external equipment to monitor the static and dynamic performance parameters of the IGBT 500 device in real time according to preset parameters, including on-state voltage drop VCE(sat), leakage current ICEO, switching delay time, rise and fall time, and power loss; achieve millisecond-level synchronization of performance parameters with acquired pressure, temperature, and high-definition image data using the NIcRIO-9039 data acquisition instrument; after synchronization, verify the data and eliminate abnormal data points caused by signal interference (using the 3σ criterion) to ensure data reliability.

[0096] S53. Multi-dimensional data fusion processing: Standardized preprocessing is performed on the synchronized pressure, temperature, performance, and image data to eliminate dimensional differences; for image data, surface feature parameters of the device (such as package flatness and area of ​​suspected cracks) are extracted using image segmentation algorithms, and the image information is transformed into quantifiable numerical features; finally, a unified format fusion dataset is generated using a data fusion algorithm (emphasizing a weighted average fusion strategy, with weights determined by the contribution of each parameter to the device performance), providing high-quality input for subsequent modeling.

[0097] S54. Constructing a Dynamic Relationship Model: Based on the fused dataset, a dynamic relationship model among pressure, temperature, performance parameters, and appearance status is constructed using a machine learning algorithm (random forest regression model, balancing fitting accuracy and interpretability). First, the dataset is divided into a training set (70%), a validation set (20%), and a test set (10%). Performance parameters and appearance status are used as output variables, while pressure and temperature are used as input variables for model training. During training, the model's hyperparameters (deep decision tree, number of leaf nodes, etc.) are optimized using grid search, and an adaptive correction module is added. By inputting basic parameters (such as rated voltage and rated current) of different IGBT devices (500), the model can automatically adjust parameter weights to accommodate device characteristics and improve model versatility.

[0098] The adaptive correction module calculates the correction amount based on the ratio of the rated voltage and rated current of the current experimental IGBT device 500 to the average of common IGBT models, combined with the preset calibration coefficient. The correction amount is used to adjust the basic weights of input variables such as pressure, temperature, and image features in the model, so as to avoid model prediction deviations caused by differences in parameters of different device models.

[0099] S55. Quantitative Analysis of Temperature-Pressure Coupling Effect: For temperature-pressure synergistic experimental scenarios, multiple comparative experiments were conducted based on an optimized dynamic correlation model. The target temperature range (-50℃ to 200℃, covering the preset experimental temperature range) was divided into 5℃ intervals. At each temperature node, ten pressure gradients were applied progressively from 0 to 5000N, and the performance parameter changes under each temperature-pressure combination were recorded. The influence of individual temperature and pressure effects, as well as the coupling effect, on performance was separated using the controlled variable method. The temperature-pressure coupling effect coefficient was quantified (defined as the ratio of performance degradation under coupling to the sum of performance degradation under individual effects), clarifying the correspondence between pressure thresholds and performance degradation in different temperature ranges. For example, in the high-temperature range (150℃-200℃), when the pressure exceeds 3000N, the conduction voltage drop degradation is ≥10%, providing accurate quantitative basis for process verification. A carrier mobility modulation model was constructed to evaluate the temperature-pressure coupling effect. The carrier mobility modulation model is as follows.

[0100] .

[0101] In the formula, This is the carrier mobility modulated by temperature and pressure (T / P), the core output of the model, representing the ease with which electrons or holes move in a semiconductor under given temperature T and pressure P. A higher value indicates stronger conductivity. It is the reference mobility, the intrinsic carrier mobility under reference conditions (typically room temperature T0 = 300 K and standard atmospheric pressure or zero mechanical stress). It is the absolute temperature (Kelvin), the actual operating temperature of the IGBT device 500. It is a temperature index, dimensionless, describing the power-law decay coefficient of mobility as a function of temperature. For phonon scattering-dominated mechanisms, it is relevant in the intermediate temperature range (50°C to 150°C). ≈1.5-2.5. As temperature increases, lattice vibrations intensify, scattering is enhanced, and mobility decreases. This is the pressure applied to the IGBT device 500; take one atmosphere (approximately 0.1 MPa) or the minimum loading pressure in the experiment; It is reference pressure. This is the pressure modulation coefficient, dimensionless, which quantifies the strength of the effect of pressure on mobility. A positive value (typically between 0 and 0.5) indicates that increased pressure reduces mobility. Its magnitude is determined by the piezoresistive coefficient and band structure of the material. The activation energy is a core physical parameter that serves as a bridge between temperature and pressure effects. It represents the energy threshold at which pressure-induced lattice defects or band deformations require thermal activation to affect carrier transport. Its value is obtained through fitting experimental data. It is Boltzmann's constant. ≈8.617×10 -5 eV / K is used to convert absolute temperature T into an energy scale.

[0102] S56. Extreme Condition Failure Analysis: For extreme condition experiments and failure analysis scenarios, the focus is on monitoring the entire process data of devices approaching or failing. When cracks (length ≥ 0.5 mm), deformations (amount ≥ 0.1 mm), or performance parameters exceeding the rated range (e.g., leakage current ICEO ≥ 1 mA) are detected by image recognition, they are marked as critical failure points. The critical pressure threshold and critical temperature peak corresponding to the critical failure point are derived in reverse using a dynamic correlation model, and the failure evolution path is traced by combining time series data (e.g., the complete process from local overheating → package deformation → crack initiation → sudden performance degradation). At the same time, the failure probability under different extreme conditions (e.g., low temperature and high pressure, high temperature and low pressure) is statistically analyzed to establish a failure risk assessment matrix, providing data support for device reliability design.

[0103] S57. Evaluation Result Output: The comprehensive evaluation results will be compiled into a standardized report, including the temperature-pressure coupling effect curve, the performance parameter comparison table under different pressures / temperatures, the critical parameter table for failure under extreme conditions, and the failure evolution path diagram. If the model prediction error is found to be ≥5% during the evaluation process, the synchronous dataset of this experiment will be added to the model training set, the model parameters will be re-optimized, and the dynamic correlation model will be iteratively upgraded to continuously improve the accuracy of quantitative evaluation.

[0104] Example 7.

[0105] This embodiment discloses a method for using an integrated experimental IGBT device press-fit fixture, wherein step S54 includes the following steps.

[0106] Step 1: Set reference parameters: Predetermine the average rated electrical parameters of common IGBT device models as the calculation benchmark, specifically: average rated voltage 1200V, average rated current 500A (the parameters are determined by calibration of 50 sets of experimental data of mainstream IGBT models, which are suitable for most common experimental scenarios).

[0107] Step 2: Extract the rated parameters of the IGBT device 500 to be tested, and calculate the ratios to the reference parameters.

[0108] 2.1 Voltage Ratio Calculation: The ratio of the rated voltage of the current experimental IGBT device to the 1200V reference value.

[0109] 2.2 Current Ratio Calculation: The ratio of the rated current of the current experimental IGBT device to the 500A reference value.

[0110] Step 3: Calculate the initial correction amount: Construct a model for dynamic voltage calibration coefficient and dynamic current calibration coefficient, and calculate the initial correction amount.

[0111] .

[0112] .

[0113] . .

[0114] In the formula, It is the dynamic voltage calibration coefficient, which characterizes the calibration intensity corresponding to the degree of voltage deviation from the reference. It is a voltage ratio, dimensionless, which is the ratio of the rated voltage of the IGBT device 500 under test to the reference voltage. It is a natural constant; It is the dynamic current calibration coefficient, which characterizes the calibration intensity corresponding to the degree of current deviation from the reference. It is a current ratio, dimensionless, which is the ratio of the rated current of the IGBT device 500 under test to the reference current. It is the initial correction quantity, dimensionless; it is the unconstrained correction quantity calculated jointly by the voltage term, the current term, and the voltage-current coupling term. It is the coupling term calibration coefficient, a fixed value, determined through verification using experimental data from fifty sets of mainstream IGBT models; It is a voltage-current coupling term, dimensionless, and is the product of the voltage ratio and the current ratio; Step 4: Boundary Constraint Handling of Correction Quantities: To avoid overcorrection leading to model parameter shifts, the initial correction quantity is limited in range. Effective Correction Quantity The calculation model is as follows.

[0115] .

[0116] Step 5: Adjust the basic weights of input variables based on the correction amount: Based on the model's preset basic weights (pressure 0.35, temperature 0.35, image features 0.1), allocate effective correction amounts according to the current characteristics of the IGBT device 500 to achieve dynamic weight adjustment.

[0117] Prioritize allocating correction amounts based on the rated parameters of the device. For example, for high-voltage IGBTs, 60%-70% of the correction amount can be added to the temperature weight to improve the prediction accuracy under high-temperature conditions; for high-current models, 50%-60% of the correction amount can be added to the pressure weight.

[0118] 5.1 For high-voltage IGBTs (U dev (≥1200V), the weighted dynamic adjustment model is as follows.

[0119] ; .

[0120] 5.2 For high-current IGBTs (I dev (≥500A), the weighted dynamic adjustment model is as follows.

[0121] ; .

[0122] 5.3 Constraint Verification: The adjusted constraint must meet the following requirements. + + + =1, ensuring a closed-loop logic for model weights.

[0123] In the formula, For performance parameter weights, =0.2, which is fixed and is the preset weight of IGBT device performance parameters (such as on-state voltage drop, switching delay, etc.) in the model. It is a temperature-sensitive factor, characterizing the degree to which the device voltage is sensitive to temperature weighting; These are the temperature base weights and the initial weights of the pressure input variables preset by the model. These are the pressure base weights, the initial weights of the pressure input variables preset by the model; These are the basic weights of image features, the initial weights of the image feature input variables preset by the model; It is the intermediate weight of temperature; under high pressure conditions, after Unnormalized temperature weights after assigning correction values; It is the intermediate pressure weight, which is the unnormalized pressure weight after the remaining correction amount is allocated under high pressure conditions. It is the intermediate weight of image features, and the unnormalized image feature weights after the remaining correction amount is allocated under high-pressure conditions. It is a stress-sensitive factor; This is the current rated current of the device. This is the current reference value; It is a normalization factor; The final weight is temperature; It is the final weight of pressure; These are the final weights of the image features; It is the effective correction amount.

[0124] This embodiment discloses a method for using an integrated experimental IGBT device press-fit fixture, wherein the extreme condition failure analysis includes the following steps.

[0125] Step 1, Preset Extreme Operating Condition Types and Configure Experimental Parameters: Based on the actual application scenarios of IGBT devices 500, preset typical extreme operating condition types: (1) Low temperature and high pressure condition (T≤-30℃, P≥4000N); (2) High temperature and low pressure condition (T≥180℃, P≤1000N); (3) High temperature and high pressure condition (T≥180℃, P≥4000N); (4) Low temperature and low pressure condition (T≤-30℃, P≤1000N). For each operating condition, set the target temperature value, pressure loading gradient (0-5000N in ten levels), and pressure holding time (30-60min preset for long-term experiments) through the PLC control module. At the same time, configure the sampling frequency of equipment such as oscilloscopes and power analyzers (performance parameter sampling frequency ≥1kHz) and the frame rate of high-definition cameras (increased to 20fps to ensure the accuracy of failure process capture).

[0126] Step 2: Synchronous Acquisition of Multi-Dimensional Data Throughout the Entire Process: The NI cRIO-9039 data acquisition instrument initiates simultaneous acquisition of four dimensions of data: pressure, temperature, performance, and image. Pressure data is preprocessed using a Kalman filter and then uploaded in real time. Temperature data is obtained through a multi-point acquisition and fusion algorithm to capture the temperature field distribution on the device surface. Performance parameters are acquired in real time using a high-voltage differential probe / current probe. A high-definition camera captures images of the device surface through a high-temperature resistant transparent protective cover. All data is bound to a unified timestamp and stored in a related database to ensure the temporal consistency of the failure process data.

[0127] Step 3, Real-time identification and marking of failure critical points: A dual judgment mechanism of intelligent image recognition and performance parameter threshold comparison is adopted to monitor and mark failure critical points in real time: (1) Image recognition: The collected images are analyzed by deep learning image segmentation algorithm. When the crack length is detected to be ≥0.5mm or the deformation is ≥0.1mm, the appearance failure judgment is triggered; (2) Parameter comparison: The performance parameters are compared with the rated range of the device in real time. When the performance parameters exceed the rated threshold, the performance failure judgment is triggered. After the failure is judged, the timestamp of the failure critical point, the current pressure value, temperature value and performance parameters are automatically marked to generate a failure warning signal.

[0128] Step 4: Reverse derivation of critical pressure and temperature thresholds: Call the constructed dynamic correlation model, input the performance / appearance failure threshold of the failure critical point, and combine it with the constraint conditions (pressure 0-5000N, temperature -50℃-200℃). Through reverse solving, the critical pressure threshold and critical temperature peak corresponding to the failure critical point are accurately derived, forming the correspondence between failure criteria and critical parameters.

[0129] Step 5, Failure Evolution Path Tracing: Based on time series data, the failure evolution path is traced through a process of time gradient division, parameter extraction, and path fitting: ① Time Gradient Division: Starting from the failure critical point, four key time nodes are extracted backward (local temperature rise initiation point, encapsulation deformation initiation point, crack initiation point, and performance drop point); ② Parameter Extraction: Pressure, temperature, performance, and appearance data for each time node are extracted to construct state characteristics; ③ Path Fitting: The curves of each parameter changing over time are fitted using data visualization tools to form a complete failure evolution path, such as local temperature rise → encapsulation deformation → crack initiation → performance drop.

[0130] Step 6, Failure probability statistics under different extreme working conditions: For the four preset extreme working conditions, multiple sets of parallel experiments are carried out (the number of experimental samples in each working condition is ≥30): (1) Count the number of failure samples under each working condition, and calculate the corrected failure probability in combination with the working condition severity coefficient, where the severity coefficient of low temperature and high pressure working condition is 1.2, high temperature and low pressure working condition is 1.1, and other extreme working conditions are 1.0; (2) Statistically analyze the critical pressure / temperature threshold of each working condition to obtain the distribution range of critical parameters under each working condition.

[0131] Step 7: Using the temperature range as the vertical axis and the pressure gradient as the horizontal axis, fill in the failure probabilities corresponding to each temperature-pressure combination into a matrix to construct a three-dimensional failure risk assessment matrix of temperature-pressure-failure probability; simultaneously, compile a failure evolution path diagram and a critical parameter table for each operating condition to form an extreme operating condition failure analysis report, providing quantitative data support for IGBT device reliability design and process parameter optimization. The analysis report results will then be visualized and output.

[0132] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for using an integrated experimental IGBT device press-fit fixture, characterized in that, include: Place the IGBT device above the copper busbar of the first mounting pair; The second clamping assembly moves down, so that the copper busbar of the second clamping assembly is in contact with the IGBT device; The first and second heating stages are activated to heat the IGBT device to the target temperature via the first and second clamping pairs. The pressure application assembly applies pressure to the IGBT device to the target pressure value through the second clamping pair; Simultaneously acquire pressure and temperature data of IGBT devices.

2. The method of using the integrated experimental IGBT device press-fit fixture according to claim 1, characterized in that... Simultaneously acquire images and performance parameters of IGBT devices to obtain a multi-dimensional fusion dataset; construct a state dynamic correlation model based on the multi-dimensional fusion dataset; obtain the rated parameters of the IGBT devices, combine them with preset calibration coefficients to obtain an effective correction amount, and adjust the input weights of pressure, temperature, and image features in the dynamic correlation model through the effective correction amount; The fused dataset and adjusted input weights are used as inputs to the dynamic correlation model to perform quantitative analysis of temperature-pressure coupling effects and extreme condition failure analysis on IGBT devices, and output evaluation reports and abnormal conditions.

3. The method of using the integrated experimental IGBT device press-fit fixture according to claim 2, characterized in that... The construction of a dynamic state association model includes the following steps: Synchronously acquire pressure, temperature, and image data from IGBT devices; perform synchronization and verification processing; Standardized preprocessing is performed on the pressure, temperature, performance, and image data after synchronous processing; Based on the fused dataset, a dynamic correlation model of stress, temperature, performance parameters and appearance status is constructed using a random forest regression model machine learning algorithm. Quantitative analysis of temperature-pressure coupling effect; Failure analysis under extreme conditions; The comprehensive evaluation results will be compiled into a standardized report and visualized.

4. The method of using the integrated experimental IGBT device press-fit fixture according to claim 2, characterized in that... In the constructed dynamic correlation model, the adaptive correction module calculates the correction amount based on the ratio of the rated voltage and rated current of the current experimental IGBT device to the average value of common IGBT models, combined with the preset calibration coefficient. The correction amount is used to adjust the basic weights of pressure, temperature and image features in the model to avoid model prediction deviations caused by differences in parameters of different device models.

5. The method of using the integrated experimental IGBT device press-fit fixture according to claim 4, characterized in that... Adjusting the base weights for pressure, temperature, and image features involves the following steps; Set the baseline reference parameters; Extract the rated parameters of the IGBT device under test and calculate the ratios to the reference parameters respectively; Construct models for dynamic voltage calibration coefficients and dynamic current calibration coefficients, and calculate the initial correction values; ； ； ; In the formula, It is the dynamic voltage calibration coefficient; It is the voltage ratio; It is a natural constant; It is the dynamic current calibration coefficient; It is the current ratio; This is the initial correction amount; These are the coupling term calibration coefficients; It is a voltage-current coupling term; Correction quantity boundary constraint processing limits the range of the initial correction quantity, and the effective correction quantity. The calculation model is as follows: ； Adjust the base weights of input variables based on the correction amount.

6. The method of using the integrated experimental IGBT device press-fit fixture according to claim 3, characterized in that... Thermo-baric coupling effect quantitative analysis evaluates the thermo-baric coupling effect by constructing a carrier mobility modulation model. The carrier mobility modulation model is as follows: ； In the formula, It is the carrier mobility after temperature and pressure modulation; It is the reference migration rate; It is absolute temperature; It is a temperature index; It is the pressure applied to the IGBT device; It is reference pressure. It is the pressure modulation coefficient; It is the activation energy, and its value is obtained by fitting experimental data; It is the Boltzmann constant.

7. The method of using the integrated experimental IGBT device press-fit fixture according to claim 1, characterized in that... The first clamping assembly includes a first insulating block, a first heat sink, a first heating platform, a first insulating heat-conducting block, and a first copper busbar. A first heat sink is fixedly installed above the first insulating block, a first heating platform is fixedly installed above the first heat sink, a first insulating heat-conducting block is fixedly installed above the first heating platform, and a first copper busbar is fixedly installed above the first insulating heat-conducting block.

8. The method of using the integrated experimental IGBT device press-fit fixture according to claim 1, characterized in that... The pressure application assembly includes a trapezoidal transmission block, ball head bolts, guide posts, and pad sleeves; An upper pressure plate is fixedly installed above the guide post, a lower limit block is fixedly installed at the lower end of the guide post, a butterfly spring is installed above the lower limit block, a pad sleeve is installed above the butterfly spring, and the pad sleeve and the butterfly spring are sleeved on the outside of the guide post; an annular groove is opened on the guide post, and a positioning block is detachably fixedly installed on the upper pressure plate. The guide post is threaded with a ball head bolt, which passes through the center hole of the guide post from top to bottom. The lower end of the ball head bolt is connected to a trapezoidal transmission block via a ball joint.

9. The method of using the integrated experimental IGBT device press-fit fixture according to claim 1, characterized in that... The second clamping assembly includes a second insulating block, a second heat sink, a second heating platform, a second insulating heat-conducting block, and a second copper busbar. A second insulating heat-conducting block is fixedly installed above the second copper busbar. A second heating platform is fixedly installed above the second insulating heat-conducting block. A second heat dissipation block is fixedly installed above the second heating platform. A pressure sensor can be fixedly installed above the second heat dissipation block. A second insulating block is fixedly installed above the pressure sensor. A support cylinder is fixedly installed on the outside of the second insulating block, the second heat dissipation block, the second heating platform, the second insulating heat-conducting block, and the second copper busbar; the support cylinder is slidably mounted on four double-headed screws; multiple electric push rods are fixedly installed on the upper pressure plate, the movable rods of the electric push rods slide in cooperation with the support cylinder, and a support plate is detachably fixedly installed at the lower end of the movable rods of the electric push rods; a laser rangefinder is fixedly installed on the support cylinder.

10. An experimental IGBT press-fit fixture using the integrated experimental IGBT device press-fit fixture described in claims 1-9, comprising a lower pressure plate, a first clamping pair, a second clamping pair, an upper pressure plate, an IGBT device, a pressure sensor, and a pressure application assembly, characterized in that... : The lower pressure plate and the upper pressure plate are fixedly connected by multiple double-headed screws to form a rigid force-bearing frame; A first clamping assembly is detachably fixed above the lower pressure plate; A pressure-applying component is detachably fixed on the upper pressure plate; a second clamping assembly is slidably mounted on the double-ended screw; and a pressure sensor is fixedly mounted on the second clamping assembly.

Images