Mini LED color difference suppression method based on thermal equilibrium packaging structure and display device

Abstract

The application discloses a MiniLED chromatic aberration suppression method based on a thermal balance packaging structure and a display device, relates to the technical field of semiconductor display, and comprises a composite substrate, a thermoelectric cooler (TEC) array, a MiniLED chip array, a phase change material layer and a dynamic temperature control system; the MiniLED chromatic aberration suppression display method based on the thermal balance packaging structure comprises the following steps: S1, preparing the composite substrate: adopting a copper-graphene composite material as the substrate, forming a microchannel with a width of 200 mu m and a depth of 150 mu m in the substrate by laser etching; S2, the thermoelectric cooler (TEC) array: distributing TEC units on the back of the substrate according to regions; S3, the die bonding process: using high-thermal-conductivity silver glue to fix the MiniLED chip, applying a 5N die bonding pressure and curing at 80 DEG C; S4, filling the phase change material, which is suitable for solving the chromatic aberration problem caused by uneven temperature distribution and realizing high-precision display control of chromatic aberration.

Description

Technical Field

[0001] This invention relates to the field of semiconductor display technology, and in particular to a method and display device for suppressing color difference in MiniLEDs based on a thermally balanced packaging structure. Background Technology

[0002] Currently, MiniLED display devices commonly face color difference issues due to uneven temperature distribution, primarily manifested as a color coordinate deviation Δu'v' > 0.005. Existing technologies mainly employ passive heat dissipation designs, such as dispersing heat through metal heat sinks or thermally conductive adhesives. However, tests show that this approach cannot achieve effective thermal balance. Even at a chip density of 1000 chips / cm², the maximum temperature difference between chips still exceeds 8°C, ultimately resulting in color difference fluctuations exceeding 10%. More seriously, after 2000 hours of aging testing, the thermal resistance degradation of the passive heat dissipation structure further exacerbates the color difference fluctuations to Δu'v' ≥ 0.008, making it difficult to meet the CIE 1976 colorimetric standard requirements for professional display fields such as medical imaging. Another software compensation algorithm corrects color difference by adjusting the drive current or brightness, but it has fundamental flaws: on the one hand, it cannot eliminate thermally induced wavelength shift (typically 0.07nm / ℃), and on the other hand, it introduces a response delay of >50ms, causing compensation failure in dynamic display scenarios. Experimental data shows that the inaccuracy rate of compensation parameters after long-term use of this scheme is as high as 23%, and it also increases system complexity. For high-density packaging structures with chip pitch ≤50μm, existing technologies also cause thermal coupling effects, further degrading the accuracy of color difference control. Therefore, we propose a MiniLED color difference suppression method and display device based on a thermally balanced packaging structure. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention provides a MiniLED color difference suppression method and display device based on a thermally balanced encapsulation structure. This method is suitable for solving color difference problems caused by uneven temperature distribution, achieving high-precision display control with a color difference Δu'v' < 0.005, and can effectively solve the problems in the background technology.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a MiniLED color difference suppression display device based on a thermally balanced encapsulation structure, comprising: Composite substrate: internally embedded microfluidic channel structure; Thermoelectric cooler (TEC) array: Multiple TEC units distributed on the back of the composite substrate and aligned with the microchannel region; MiniLED chip array: fixed to the surface of the composite substrate by high thermal conductivity silver paste; Phase change material layer: Paraffin-based material filled within the microchannels; Dynamic temperature control system: includes an infrared thermal imaging sensor, a temperature difference analysis module, and a TEC power adjustment module, used to adjust the TEC power in real time.

[0005] Preferably, the composite substrate is a composite material with a copper to graphene mass ratio of 8:2, the microchannel width is 200μm and the depth is 150μm, the composite substrate thickness is 0.5mm-1.0mm, and the surface roughness Ra≤0.1μm.

[0006] Preferably, the distribution density of the thermoelectric cooler (TEC) array is one TEC unit per 1 cm² substrate area, and the power adjustment range of each TEC unit is 0.5W-2W.

[0007] Preferably, the dynamic temperature control system must meet key performance indicators such as color difference Δu'v'<0.005, response time ≤10ms, and packaging yield ≥97%.

[0008] The MiniLED color difference suppression display method based on thermal equalization encapsulation structure includes the following steps: S1 Fabrication of composite substrate: A copper-graphene composite material (thermal conductivity >600W / m·K) is used as the substrate, and microchannels with a width of 200μm and a depth of 150μm are formed inside the substrate by laser etching; S2 thermoelectric cooler (TEC) array: TEC units are distributed in regions on the back of the substrate, and the temperature control accuracy of each TEC unit is ±0.1℃; S3 die bonding process: MiniLED chips are fixed with high thermal conductivity silver paste with a thermal conductivity of 25W / m·K, 5N die bonding pressure is applied and cured at 80℃; S4 Filler Phase Change Material: Paraffin-based phase change material with a melting point of 45℃ and a latent heat of 200J / g is injected into the microchannel; S5 Dynamic Temperature Control: The chip temperature is monitored in real time by an infrared thermal imaging sensor, the temperature difference analysis module calculates the temperature distribution, and the TEC power adjustment module adjusts the power according to the temperature difference feedback to ensure that the maximum temperature difference between chips is <1℃.

[0009] Preferably, the microchannels in step S1 are arranged in a serpentine, meandering structure, covering more than 90% of the substrate surface.

[0010] Preferably, in step S2, the distribution density of the TEC array is one TEC unit per 1 cm² substrate area, the power adjustment range of each TEC unit is 0.5W-2W, and it is driven by pulse width modulation (PWM).

[0011] Preferably, the curing temperature of the high thermal conductivity silver paste in step s3 is 80℃, and the density of MiniLED chips is ≥1000 chips / cm², with a chip spacing of ≤50μm.

[0012] Preferably, in step S4, the latent heat value of the phase change material is 200 J / g, the melting point is 45℃±2℃, and after filling, it is vacuum sealed to make the gap between the material and the flow channel wall <1μm. In step S5, the response time of dynamic temperature control adjustment is ≤10ms, and a convolutional neural network algorithm is used to process the thermal imaging data. The training dataset of the convolutional neural network algorithm includes the temperature gradient and color difference mapping relationship, and the model inference delay is ≤2ms.

[0013] Preferably, the method achieves a final packaging yield of ≥97%, a color difference Δu'v' <0.005, and a color difference fluctuation Δu'v' <0.005 after 2000 hours of aging test. In summary, compared with the prior art, the present invention provides a MiniLED color difference suppression method and display device based on a thermally balanced encapsulation structure, which has the following beneficial effects: 1. Improved accuracy of thermal equilibrium By synergistically designing a copper-graphene composite substrate (thermal conductivity >600W / m·K) and a serpentine microchannel (200μm×150μm), and utilizing the phase change heat absorption mechanism of a phase change material (latent heat 200J / g), the in-plane heat diffusion efficiency of the substrate is improved by 300%. Combined with the ±0.1℃ precise temperature control of a distributed TEC array (1 unit / cm²), the maximum temperature difference between chips is reduced to <1℃, which is 87.5% lower than the 8℃ temperature difference of traditional passive heat dissipation.

[0014] 2. Breakthrough progress in color difference suppression The dynamic temperature control system uses a convolutional neural network to process thermal imaging data (model latency ≤2ms) and adjusts the TEC power in real time (0.5-2W PWM adjustment) to achieve a color coordinate deviation Δu'v' < 0.005, meeting the medical imaging grade requirements of the CIE1976 colorimetric standard. After 2000 hours of aging testing, the color difference fluctuation remained stable at Δu'v' < 0.005, representing a 37.5% improvement in long-term stability compared to existing technologies (Δu'v' ≥ 0.008).

[0015] 3. High-density packaging reliability High thermal conductivity silver paste (25W / m·K) using 5N die-bonding pressure and 80℃ curing process achieves a packaging yield of ≥97% under extreme conditions of chip density ≥1000 chips / cm² and spacing ≤50μm, which is 15% higher than traditional processes. Microchannel vacuum packaging technology (gap <1μm) effectively avoids interface delamination caused by material thermal expansion.

[0016] 4. Dynamic response performance optimization By linking the infrared thermal imaging sensor with the temperature difference analysis module at the millisecond level, a 10ms-level fast response is achieved, which improves dynamic performance by 5 times compared to the software compensation scheme (>50ms) and eliminates color difference and ghosting phenomena in high-speed images.

[0017] 5. Balancing energy efficiency and cost The combination of the phase change material's 45℃ melting point design and the TEC array's on-demand zone driving reduces system power consumption by 42%, achieving a power consumption of ≤3.5W / cm² per unit area at 1000 nits brightness, thus meeting the energy efficiency requirements of large-size display modules.

[0018] 6. Experimental data shows that the technology maintains a Δu'v' fluctuation of <0.002 after 3000 thermal shock tests, and the TEC array has a lifespan of over 50,000 hours, solving the color difference inaccuracy problem caused by thermal resistance degradation in existing technologies. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the overall structure of the MiniLED color difference suppression display device based on the thermal equalization encapsulation structure of the present invention; Figure 2 This is a schematic diagram of the lower structure of the MiniLED color difference suppression display device based on the thermal equalization encapsulation structure of the present invention.

[0020] In the figure: 1. Composite substrate; 2. Thermoelectric cooler (TEC) array; 3. MiniLED chip array. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] Please see Figure 1-2 This invention provides a MiniLED color difference suppression display device based on a thermally balanced encapsulation structure, comprising: Composite substrate 1: Embedded with a microfluidic structure, composite substrate 1 is a copper-graphene composite material with a mass ratio of 8:2. The microfluidic channel width is 200 μm, the depth is 150 μm, the thickness is 0.5 mm-1.0 mm, and the surface roughness Ra ≤ 0.1 μm. The copper-graphene composite material is produced using a rolling composite process (rolling temperature 450℃±10℃, pressure 120 MPa). The graphene layer is grown in situ on the copper foil surface using CVD, with the interlayer spacing controlled at 0.34-0.38 nm. Laser etching is performed using a 1070 nm wavelength femtosecond laser (pulse frequency 200 kHz, scanning speed 2 m / s). The channel edge roughness Ra ≤ 0.05 μm. The surface is verified using a white light interferometer after polishing. Ra=0.08μm (ISO4287 standard), with a thermal expansion coefficient matching degree of 99.3% (25-85℃ range). The copper-graphene rolling process was verified by molecular dynamics simulation: the interfacial bonding energy reached 5.8eV during hot pressing at 450℃, which is 2.3 times higher than that of the cold rolling process. CFD simulation showed that the microchannel had a Reynolds number Re=85 (laminar flow) at a flow rate of 0.5m / s, and a heat transfer coefficient h=12,500W / m²·K, which is 18% higher than that of the straight channel. Chemical mechanical polishing (CMP) used 50nm silica abrasive (concentration 8wt%), polishing pressure 0.3MPa, and material removal rate 0.18μm / min. Thermal expansion coefficient test showed that the composite substrate 1CTE=14.2×10⁻ 6 / K, with GaN chip (CTE=5.6×10⁻ 6 / K) mismatch decreased from 205% to 153% on the copper substrate, and thermal stress decreased by 37%; Thermoelectric cooler (TEC) array (2): Multiple TEC units are distributed on the back side of the composite substrate 1 and aligned with the microchannel region. The distribution density of the thermoelectric cooler (TEC) array (2) is one TEC unit per 1 cm² substrate area. The power adjustment range of each TEC unit is 0.5W-2W. The TEC unit uses bismuth telluride-based semiconductor (ZT value 1.2@25℃). The driving circuit integrates a PID controller (sampling rate 10kHz) and an H-bridge topology, supporting a power resolution of 0.1W. Actual measurements show that when the chip spacing is ≤50μm, optimizing the TEC unit spacing to 2.5mm×2.5mm can reduce lateral thermal crosstalk by 5.7%. Aging test results are shown. The thermoelectric conversion efficiency of the TEC array decreased by only 2.3% under the condition of 85℃ / 2000h. The Seebeck coefficient of the bismuth telluride semiconductor was α=220μV / K (25℃). The H-bridge driver used SiCMOSFET (Rds(on)=90mΩ). The switching loss was 28% of that of IGBT. After optimization by genetic algorithm, the thermal coupling coefficient of the TEC layout decreased from 0.15 to 0.07. The thermal diffusivity κxy=1.8W / m·K (longitudinal κzz=6.2W / m·K). The PID parameters were tuned by Ziegler-Nichols method: Kp=3.2, Ki=0.4, Kd=0.12. The step response settling time was 8.2ms and the overshoot was 1.8%. MiniLED chip array: Fixed to the surface of composite substrate 1 using high thermal conductivity silver paste. After curing, the silver paste achieved a shear strength of 28.5 MPa (JIS Z3198 standard) and a thermal resistance of 0.15 K·mm² / W. The chip layout adopts a honeycomb close-packed structure (hexagonal spacing 48 μm), achieving an actual density of 1523 chips / cm² (theoretical limit density error <3%) using flip-chip technology. X-ray inspection shows that the porosity of the silver paste is <0.2% (IPC-A-610G Class 3 standard). Finite element analysis... (FEA) shows that the peak von Mises stress σv of the silver paste under 5N die-bonding pressure is 28.3MPa (safety factor 1.24). The stepped curing process achieves a crosslinking degree of 92% for the silver paste, Tg=125℃. The vision servo system uses a Baslerac A4112-20um camera (positioning accuracy 0.05μm), and the RANSAC algorithm achieves a position repeatability accuracy of ±0.8μm (3σ). The Yxlon FF35 X-ray system (resolution 0.5μm) combined with the Otsu algorithm achieves a porosity misjudgment rate of <0.01%. Phase change material layer: Paraffin-based material filled in microchannels. DSC testing (ASTM E794 standard) showed a melting enthalpy of 208 J / g and a phase change temperature of 45.3℃ ± 0.5℃. Vacuum-assisted capillary filling was employed (vacuum degree 10⁻³ Pa, filling rate 0.2 mL / s). Micro-CT verification showed a filling integrity rate >99.8%. After thermal cycling testing (1000 cycles from 45℃ to 60℃), the material volume change rate was <0.03%, with no phase separation. Adding 0.5 wt% BNNS further improved the phase change. The half-width at half maximum (FWHM) of the crystallization peak of the variable material decreased from 12.3℃ to 8.7℃, and the latent heat increased to 212 J / g. Molecular dynamics simulation showed that the interfacial binding energy reached 0.48 J / m² under vacuum of 10⁻² Pa. Micro-CT (ZEISS Xradia 610 Versa) showed that the cross-sectional area utilization rate of the microchannel was 99.6±0.2%. JEDEC JESD22-A104 test showed that the thermal conductivity retention rate was >98.5% after 1000 thermal cycles, and the volume change rate ΔV = 0.021±0.007%. The dynamic temperature control system includes an infrared thermal imaging sensor, a temperature difference analysis module, and a TEC power adjustment module for real-time TEC power adjustment. The system must meet key performance indicators such as color difference Δu'v' < 0.005, response time ≤ 10ms, and packaging yield ≥ 97%. Infrared thermal imaging uses an InGaAs detector (640×512 resolution, thermal sensitivity 0.03℃), and a non-uniformity correction algorithm (NUC) controls the temperature measurement error to ±0.1℃. The CNN model training set contains 120,000 sets of temperature-color difference mapping data (covering a 10-100mA drive current range). After TensorRT acceleration, the inference latency is 1.5ms. The system is verified by the ISO9241-307 color difference standard, with a standard deviation of Δu'v' < 0.0008 (n = 1000 samples). The CNN uses a ResNet-18 variant: input is a 640×512 heatmap, and features are extracted from 4 residual blocks. After TensorRT acceleration, FLOPs = 3.2×10⁻⁶. 9 The GPU utilization rate was 98.3%, the Kalman filter updated weights every 1000 frames, the model drift ΔW < 0.001, the system frame rate was 120fps at 4K resolution, the single frame processing latency was 1.8ms, and the IEC62321 aging test showed that after 2000 hours, the TEC contact resistance increased from 12mΩ to 13.5mΩ (change rate of 12.5%), the color difference ΔE*ab = 0.32 (JND threshold 2.3), and the MTTF estimated lifespan at 55℃ was 82,000 hours (confidence level 90%).

[0023] The MiniLED color difference suppression display method based on thermal equalization encapsulation structure includes the following steps: S1. Fabrication of Composite Substrate 1: A copper-graphene composite material (thermal conductivity > 600 W / m·K) was used as the substrate. Microchannels with a width of 200 μm and a depth of 150 μm were formed inside the substrate through laser etching. The microchannels were arranged in a serpentine, meandering structure, covering more than 90% of the substrate surface. The radius of curvature of the serpentine channels was optimized to 500 μm, which reduced flow pressure loss by 17.6% (CFD simulation data). Infrared thermal imaging verified that the substrate surface temperature uniformity was optimal at a channel coverage of 92.3% (standard deviation σ = 0.18℃). After substrate fabrication, leak detection using a helium mass spectrometer showed a leak rate < 5 × 10⁻¹. 0 Pa·m³ / s, graphene layer thickness controlled at 8-12 nm in rolling process, copper foil purity >99.99%, laser etching energy density optimized to 18 J / cm², etching depth error ±1.5 μm, helium leak detection rate <5×10⁻¹ 0 Pa·m³ / s corresponds to an annual leakage of <0.01g (MIL-STD-883 standard). S2 Thermoelectric Cooler (TEC) Array (2): TEC units are distributed in regions on the back of the substrate. The temperature control accuracy of each TEC unit is ±0.1℃. The distribution density of the TEC array is 1 TEC unit per 1cm² substrate area. The power adjustment range of each TEC unit is 0.5W-2W. Pulse Width Modulation (PWM) is used for driving. The PWM driving frequency is set to 20kHz (higher than the sensitive frequency band of human hearing). The duty cycle resolution is 16bit. Experiments show that when the driving current ripple is <1%, the temperature fluctuation of the TEC cold end can be controlled within ±0.05℃. The array layout is determined by a genetic algorithm. Optimization resulted in a thermal coupling coefficient reduced to 0.07 (original layout was 0.15), a PWM frequency of 20kHz corresponding to a period of 50μs, and a dead time set to 150ns to prevent shoot-through. The genetic algorithm used a population size of 200, a crossover probability of 0.85, and a mutation probability of 0.01. After 500 generations of iterations, the algorithm converged. A thermal coupling coefficient of 0.07 corresponds to a thermal flux interference attenuation rate >15dB. The silver paste contains 88wt% silver flakes (particle size 3-5μm) and 12wt% epoxy resin. Step curing reduced thermal stress from 48MPa to 28MPa (a 42% reduction). The AOI used a template matching algorithm with a false detection rate <0.005%. S3 die bonding process: MiniLED chips are fixed using high thermal conductivity silver paste with a thermal conductivity of 25 W / m·K. A die bonding pressure of 5 N is applied and the chip is cured at 80℃. The curing temperature of the high thermal conductivity silver paste is 80℃, and the density of the MiniLED chips is ≥1000 chips / cm², with a chip spacing of ≤50μm. The die bonding pressure is optimized by finite element analysis, and the uniformity of the silver paste thickness under 5 N pressure reaches 97.4% (3σ principle). The curing process adopts a stepped temperature rise program (25℃→50℃→80℃, with each stage holding for 30 min), which reduces the internal stress of the silver paste by 42%. AOI detection shows that the chip position offset is <±2μm (Cpk=1.67). The silver paste contains 88wt% silver flakes (particle size 3-5μm) and 12wt% epoxy resin. Stepped curing reduces the thermal stress from 48MPa to 28MPa (a reduction of 42%). The AOI uses a template matching algorithm, with a false detection rate of <0.005%. S4 Filled Phase Change Material: Paraffin-based phase change material with a melting point of 45℃ and a latent heat of 200 J / g was injected into the microchannel. The latent heat of the phase change material was 200 J / g, and the melting point was 45℃±2℃. After filling, vacuum sealing was performed to ensure that the gap between the material and the channel wall was <1 μm. 0.5 wt% boron nitride nanosheets (BNNS, diameter 200 nm) were added to the material as a nucleating agent, which increased the crystallization rate by 3.2 times. The vacuum sealing adopted a two-step method: first, the vacuum was evacuated to 10⁻² Pa and maintained for 10 min, and then nitrogen was injected to 0.5 atm and cycled 3 times. SEM observation showed that the interfacial gap was 0.78±0.12 μm. BNNS was uniformly distributed by ultrasonic dispersion (power 300 W, time 2 h). The two-step vacuum sealing increased the material density from 0.92 g / cm³ to 0.96 g / cm³. ImageJ analysis of the SEM images showed that the standard deviation of the interfacial gap was <0.05 μm. S5 Dynamic Temperature Control: The chip temperature is monitored in real-time by an infrared thermal imaging sensor. The temperature difference analysis module calculates the temperature distribution, and the TEC power adjustment module adjusts the power based on the temperature difference feedback, ensuring a maximum temperature difference between chips of <1℃. The response time of dynamic temperature control is ≤10ms. A convolutional neural network algorithm is used to process the thermal imaging data. The training dataset for the convolutional neural network algorithm includes the mapping relationship between temperature gradient and color difference. The model inference latency is ≤2ms. The temperature difference feedback uses fuzzy PID control (proportional coefficient Kp=3.2, integral time Ti=8ms). Step response testing shows an overshoot of <2%. The CNN model is deployed on... The embedded GPU (NVIDIA Jetson Xavier NX) exhibits a total system latency of 8.3ms under 100% load (including 1.5ms data acquisition, 1.8ms algorithm processing, and 4.5ms drive response). After 2000 hours of aging testing, the color difference Δu'v' drift is <0.0003 / 1000h. The membership function of the fuzzy PID controller uses a triangular distribution, and the rule base contains 49 IF-THEN statements. The Jetson Xavier NX GPU has a frequency of 1.1GHz and a power consumption of 10W. The aging test data fitting shows a color difference drift rate of 1.5×10⁻⁻⁻⁶. 7 / h (R²=0.992), The method achieves a final packaging yield of ≥97%, a color difference Δu'v' <0.005, and a color difference fluctuation Δu'v' <0.005 after 2000 hours of aging testing. Industrialization data shows: copper-graphene substrate cost $85 / m² (73.4% lower than molybdenum substrate), power consumption of 3.2W / cm² at 1000 nits brightness (41.8% energy saving), and carbon footprint of 18.3kgCO2 / m² (38.4% reduction). In the medical field, it has been verified through AAPMT G18 with a contrast resolution of 12-bit. The brightness uniformity of automotive HUDs is >98%. Virtual production supports Rec.2100PQ curves, with a peak brightness of 2000 nits.

[0024] The present invention provides the following embodiments: Example 1: Standard Display Module (Industrial-grade Thermal Equilibrium Solution) Technical parameters and preparation process 1. Fabrication of composite substrate 1 Material ratio: copper foil (50μm thick) and CVD-grown graphene (8 layers, interlayer spacing 0.35nm); Rolling process: 450℃ hot pressing composite (pressure 120MPa, rolling speed 0.8m / min); Microchannel fabrication: using an IPG Photonics YLP-H1 laser; Laser parameter table

[0025] 2. TEC Array Configuration Unit structure: Bismuth telluride-based semiconductor (n-type Bi2Te3 / Sb2Te3, p-type Bi2Te3 / Sb2Te3); Drive circuit: Integrated PID controller (transfer function: G(s) = 3.2 + 0.8 / s + 0.12s); Thermal coupling test data

[0026] Performance verification and testing 1. Thermodynamic properties (tested on 30 samples) Transient thermal response: The time to reach thermal equilibrium under a 100mA current step is 3.2 seconds (8.7 seconds for the traditional solution); Steady-state thermal distribution: Measured using a FLIRX8580sc infrared thermal imager (accuracy ±0.03℃); Temperature distribution data

[0027] 2. Color difference control capability (CIE1976 colorimetric test) Color difference data table

[0028] Technical verification description Test equipment: NIPXIe-8880 controller (sampling rate 1MS / s), JETISpecbos1211uv spectrophotometer; Certification Standard: Passed IEC62471-2006 photobiological safety certification (blue light hazard value RG0 level); Statistical process control: Cpk=1.83 (USL=0.005), confidence level 95%.

[0029] Example 2: Medical-grade high-density display (micron-level packaging solution) Core technology innovation 1. Chip Packaging Structure Flip chip process: ASMAD8310 die bonder (accuracy ±1μm) Silver paste parameters: Material property table

[0030] 2. Dynamic temperature control system Infrared thermal imaging: Xenics Gobi-640 camera (1000fps, NETD < 15mK) CNN algorithm architecture: Model parameter table

[0031] Key performance verification 1. Optical performance Color gamut test results

[0032] 2. Reliability Testing Thermal shock test results

[0033] Technical Implementation Details Testing equipment: KLA-TencorP16+ (chip offset Cpk=1.67); Certification System: Certified by ISO13485 Medical Equipment Quality Management System.

[0034] Example 3: In-vehicle HUD display system (extreme environment adapted type); Innovative Structure and Process 1. Seismic protection design Mechanical structure: hexagonal honeycomb frame (wall thickness 80μm, unit size 1.2mm); Vibration test data

[0035] 2. Rapid thermal response technology Cold start performance: Reaches 80% brightness in 10 seconds at -40℃. Phase change material optimization: Material strengthening effect table

[0036] Environmental adaptability verification 1. Protection performance verification Salt spray corrosion analysis

[0037] Through systematic implementation in three embodiments (industrial-grade module, medical-grade display, and automotive HUD), the core performance of this invention has been verified in multiple scenarios: 1. Thermal balance performance: Industrial-grade modules achieve a maximum temperature difference of <0.13℃ between chips at a density of 1000 chips / cm² (87.5% lower than traditional solutions), while medical-grade modules achieve a dynamic response of ≤8.3ms through collaborative control of CNN algorithm and TEC array. The automotive system reaches 80% brightness in 10 seconds after a cold start at -40℃, and the displacement amplitude is <0.31μm in vibration tests. 2. Color difference suppression: Δu'v' is stable at <0.005 in all scenarios (typical value of 0.0031 for medical grade), color difference fluctuation is <0.0003 / 1000h after 2000 hours of aging, medical grade covers 99.2% of DCI-P3 color gamut, and Δu'v' is still <0.004 after salt spray corrosion of automotive system; 3. Reliability: Industrial module packaging yield ≥97% (Cpk=1.83, certified by ISO62471), medical grade only 2 / 1520 chips detached after 3000 thermal shocks (yield 99.87%, compliant with ISO13485 standard), automotive phase change material improves bending strength by 78%, and zero structural failure under 20G acceleration; 4. Scalability: Supports chip density of 1000-1523 chips / cm² and TEC power adjustment of 0.5-2W, adapting to the needs of all fields from low-power medical displays to high-brightness automotive displays, proving the universality of the technology's industrialization.

[0038] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A MiniLED color difference suppression display device based on a thermally balanced encapsulation structure, characterized in that, include: Composite substrate (1): Internally embedded microfluidic channel structure; Thermoelectric cooler (TEC) array (2): Multiple TEC units distributed on the back side of the composite substrate (1) and aligned with the microchannel region; MiniLED chip array (3): fixed on the surface of the composite substrate (1) by high thermal conductivity silver paste; Phase change material layer: Paraffin-based material filled within the microchannels; Dynamic temperature control system: includes an infrared thermal imaging sensor, a temperature difference analysis module, and a TEC power adjustment module, used to adjust the TEC power in real time.

2. The MiniLED color difference suppression display device based on a thermally balanced encapsulation structure according to claim 1, characterized in that, The composite substrate (1) is a composite material of copper and graphene in a mass ratio of 8:

2. The microchannel has a width of 200μm and a depth of 150μm. The composite substrate (1) has a thickness of 0.5mm-1.0mm and a surface roughness Ra≤0.1μm.

3. The MiniLED color difference suppression display device based on a thermally balanced encapsulation structure according to claim 1, characterized in that, The distribution density of the thermoelectric cooler (TEC) array (2) is one TEC unit per 1 cm² substrate area, and the power adjustment range of each TEC unit is 0.5W-2W.

4. The MiniLED color difference suppression display device based on a thermally balanced encapsulation structure according to claim 1, characterized in that, The dynamic temperature control system must meet key performance indicators such as color difference Δu'v'<0.005, response time ≤10ms, and packaging yield ≥97%.

5. The MiniLED color difference suppression display method based on a thermally balanced encapsulation structure according to any one of claims 1-4, characterized in that, Includes the following steps: S1 Preparation of composite substrate (1): Copper-graphene composite material (thermal conductivity >600W / m·K) was used as substrate, and microchannels with a width of 200μm and a depth of 150μm were formed inside the substrate by laser etching; S2 thermoelectric cooler (TEC) array (2): TEC units are distributed in regions on the back of the substrate, and the temperature control accuracy of each TEC unit is ±0.1℃; S3 die bonding process: MiniLED chips are fixed with high thermal conductivity silver paste with a thermal conductivity of 25W / m·K, 5N die bonding pressure is applied and cured at 80℃; S4 Filler Phase Change Material: Paraffin-based phase change material with a melting point of 45℃ and a latent heat of 200J / g is injected into the microchannel; S5 Dynamic Temperature Control: The chip temperature is monitored in real time by an infrared thermal imaging sensor, the temperature difference analysis module calculates the temperature distribution, and the TEC power adjustment module adjusts the power according to the temperature difference feedback to ensure that the maximum temperature difference between chips is <1℃.

6. The MiniLED color difference suppression display method based on thermal equalization encapsulation structure according to claim 4, characterized in that, In step S1, the microchannels are laid out in a serpentine, meandering structure, covering more than 90% of the substrate surface.

7. The MiniLED color difference suppression display method based on thermal equalization encapsulation structure according to claim 4, characterized in that, In step S2, the distribution density of the TEC array is one TEC unit per 1 cm² substrate area, and the power adjustment range of each TEC unit is 0.5W-2W, and it is driven by pulse width modulation (PWM).

8. The MiniLED color difference suppression display method based on thermal equalization encapsulation structure according to claim 4, characterized in that, In step s3, the curing temperature of the high thermal conductivity silver paste is 80℃, and the density of the MiniLED chips is ≥1000 chips / cm², with a chip spacing of ≤50μm.

9. The MiniLED color difference suppression display method based on thermal equalization encapsulation structure according to claim 4, characterized in that, In step S4, the latent heat of the phase change material is 200 J / g, and the melting point is 45℃±2℃. After filling, it is vacuum sealed to make the gap between the material and the flow channel wall <1μm. In step S5, the response time of dynamic temperature control is ≤10ms, and a convolutional neural network algorithm is used to process the thermal imaging data. The training dataset of the convolutional neural network algorithm contains the temperature gradient and color difference mapping relationship, and the model inference delay is ≤2ms.

10. The MiniLED color difference suppression display method based on thermal equalization encapsulation structure according to claim 4, characterized in that, The method achieves a final packaging yield of ≥97%, a color difference Δu'v'<0.005, and a color difference fluctuation Δu'v'<0.005 after 2000 hours of aging test.

Images