Optimization Design Method for Heat Dissipation Structure of Power Circuit Boards in New Energy Vehicles
By configuring a stress-collecting groove array, a flexible thin-walled region, phase change microcapsules, and a piezoelectric pump drive frequency on the power circuit board of new energy vehicles, the problems of independent operation of the heat dissipation module, slow response of phase change materials, and thermal stress damage are solved, achieving efficient, low-energy heat dissipation and improved reliability.
Patent Information
- Application Number
- CN202611481768.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-22
- Publication Date
- 2026-11-10
AI Technical Summary
The heat dissipation modules of existing power circuit boards in new energy vehicles work independently without synergy. The phase change materials have a slow response speed, and microchannel cooling relies on external high-energy-consuming pumps. Differences in thermal expansion coefficients cause interfacial thermal stress that damages reliability. Furthermore, there is a lack of technical means to convert harmful thermal stress into useful work.
By configuring a stress-collecting groove array, a flexible thin-walled region, phase change microcapsules, and a piezoelectric pump driving frequency, thermal stress is directionally converted into driving force, self-excited oscillation of coolant flow is achieved, pressure waves trigger the phase change material to respond in advance, and phase-locked loop forces the piezoelectric pump to resonate and enhance coolant circulation.
It achieves synergistic enhancement of heat dissipation capacity, extremely low pump power consumption, improved transient response speed, enhanced thermal cycling reliability, and a compact structure that requires no external equipment.
Smart Images

Figure CN123075797A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of thermal management technology for electronic control systems of new energy vehicles, and in particular to a method for optimizing the heat dissipation structure of a power circuit board for new energy vehicles. Background Technology
[0002] With the rapid development of new energy vehicles towards higher power density and higher integration, the thermal load problem of electronic control power modules is becoming increasingly prominent. The widespread application of wide bandgap power devices such as silicon carbide and gallium nitride has significantly improved power density, but it has also placed higher demands on the heat dissipation design of power circuit boards. Currently, heat dissipation solutions for power circuit boards mainly include thermally conductive via arrays, metal substrates, embedded copper blocks, and liquid cooling plates.
[0003] However, existing technologies still have the following shortcomings: the heat dissipation structures operate independently, lacking synergistic effects. In existing solutions, heat dissipation methods such as microchannel liquid cooling, phase change thermal storage, and jet impingement are often offered as independent parallel options for designers to choose from or simply stack. There is no physical coupling or energy transfer relationship between the modules; the heat dissipation capacity is merely the arithmetic sum of the effects of each module, failing to produce a synergistic enhancement effect.
[0004] Phase change materials (PCCs) have a slow response speed, making them difficult to handle transient thermal shocks. PCCs rely on thermal conduction to trigger melting and heat absorption, with a response lag time typically reaching hundreds of milliseconds. Under transient conditions such as rapid acceleration in new energy vehicles, chip temperatures can rise sharply within tens of milliseconds. PCCs do not have enough time to absorb sufficient heat, resulting in the inability to effectively suppress transient temperature rise peaks.
[0005] Microchannel liquid cooling relies on external pump power, resulting in high energy consumption and response delays. Traditional microchannel cooling systems depend on electric pumps to drive coolant circulation, consuming 80 to 150W of power, directly reducing the vehicle's driving range. Furthermore, pump flow regulation depends on the electronic control link of sensors, controllers, and actuators, with response delays reaching hundreds of milliseconds.
[0006] Mismatch in coefficients of thermal expansion leads to interfacial thermal stress and reliability issues. The significant difference in coefficients of thermal expansion between the power chip and the heat sink substrate generates substantial thermal stress under frequent temperature cycling conditions, resulting in failure modes such as solder fatigue and interfacial cracking.
[0007] There is a lack of technical means to convert harmful thermal stress into useful work. In existing designs, thermal stress is passively endured or mitigated through material matching as a harmful factor, and there is no technical solution to directionally convert thermal stress into a driving force to drive the flow of heat dissipation medium. Summary of the Invention
[0008] In view of this, the present invention addresses the deficiencies of the existing technology, and its main objective is to provide an optimized design method for the heat dissipation structure of the power circuit board of new energy vehicles. This method solves the problems in the existing technology, such as the lack of synergistic effect due to the independent operation of the heat dissipation module, the slow response speed of the phase change material, the reliance on external high-energy-consuming pumps for microchannel cooling, the damage to reliability caused by thermal stress, and the lack of technical means to convert harmful thermal stress into useful work.
[0009] To achieve the above objectives, the present invention adopts the following technical solution: A method for optimizing the heat dissipation structure of a power circuit board for new energy vehicles includes the following steps: S1. Stress collection groove array parameter configuration: Obtain the heat source distribution and thermal expansion coefficient gradient distribution of the power chip, determine the layout area of the stress collection groove according to the heat source distribution, determine the expected thermal stress amplitude of each area according to the thermal expansion coefficient gradient distribution, configure V-shaped or U-shaped grooves as stress collection grooves at the substrate interlayer interface between the power chip and the coolant channel, and convert the in-plane thermal shear stress generated by the power chip into periodic pulsating extrusion pressure, which serves as the power source for driving the heat dissipation structure. S2. Flexible thin-walled region wall thickness parameter configuration: Taking the amplitude of the pulsating extrusion pressure in each region in S1 as input, the wall thickness of the flexible thin-walled region is configured at the position of the stress collection groove in each region. The wall thickness of the flexible thin-walled region is inversely proportional to the amplitude of the pulsating extrusion pressure, so that the flexible thin-walled region undergoes periodic deformation under the action of the pulsating extrusion pressure, resulting in the alternating contraction and expansion of the cross-sectional area of the coolant channel, causing the coolant to generate self-sustaining fluid self-excited oscillation. S3. Phase change microcapsule parameter configuration: Taking the frequency and amplitude of the pressure wave generated by the fluid self-excited oscillation in S2 as input, configure the shell wall material, shell wall thickness and core material composition of the phase change microcapsule, so that the shell of the microcapsule undergoes elastic deformation under the action of pressure wave, and the phase change material is triggered in advance under the condition of lower than the conventional melting temperature by pressure-induced dynamic shift of phase change temperature, and the phase change response time is set based on the pressure wave frequency. S4. Piezoelectric pump drive frequency configuration: Collect the fundamental frequency of fluid self-excited oscillation at the inlet of the coolant channel in S2, use this fundamental frequency as the locking reference for the piezoelectric pump drive frequency, and force the drive frequency of the piezoelectric pump to be locked to this fundamental frequency through the phase-locked loop circuit, so as to generate a resonance enhancement effect at the jet inlet to maintain coolant circulation.
[0010] As a preferred embodiment: In S1, the density of the stress current collection grooves is positively correlated with the local heat flux density of the power chip, and the peak region of the groove density corresponds to the hot spot region of the chip; the groove width of the stress current collection groove is 50-200μm, the groove depth is 30-100μm, the groove spacing is 200-500μm, and the extension direction of the groove is at an angle of 30°-60° with the chip arrangement direction.
[0011] As a preferred embodiment: in S2, the wall thickness of the flexible thin-walled region is 0.10-0.20 mm, and the wall thickness of the non-flexible region is 0.40-0.60 mm; the wall thickness d_i of the flexible thin-walled region satisfies an inverse relationship with the amplitude of the pulsating extrusion pressure P_i, and the minimum wall thickness value is configured in the region where the amplitude of the expected pulsating extrusion pressure is the largest.
[0012] As a preferred embodiment: In S2, the coolant channel adopts an H-shaped fractal topology with 3-4 fractal levels and a hydraulic diameter of 0.6-1.5 mm.
[0013] As a preferred embodiment: In step S3, the shell wall material of the phase change microcapsules is high-density polyethylene or urea-formaldehyde resin, with a shell wall thickness of 2-10 μm. The core material is a composite phase change material of paraffin and expanded graphite, wherein the mass fraction of paraffin is 75%-85%, the mass fraction of expanded graphite is 15%-25%, the microcapsule particle size is 30-100 μm, and the melting temperature of the phase change material is 80-100℃.
[0014] As a preferred embodiment: in S3, the phase change response time τ and the pressure wave frequency f satisfy τ≤(2πf)⁻¹, so that the phase change response time matches the temperature rise rate of the transient thermal shock.
[0015] As a preferred embodiment: in S4, the phase-locked loop circuit tracks the fundamental frequency of the fluid self-excited oscillation in the range of 50-900Hz, and the tracking rate is greater than or equal to 100Hz / ms; the driving voltage of the piezoelectric pump in the resonant driving mode is 15%-20% of the conventional non-resonant driving voltage.
[0016] As a preferred embodiment: In S4, the frequency locking condition of the phase-locked loop circuit is: the piezoelectric pump driving frequency f_p is equal to the fundamental frequency f_0 of the fluid self-excited oscillation, and the phase difference Δφ ≤ ±5°; when the frequency is lost, the phase-locked loop searches for and locks a new fundamental frequency again in a frequency sweep manner.
[0017] As a preferred option, it also includes S5, integrated manufacturing parameter configuration: based on the stress collection groove parameters, flexible thin-wall region wall thickness parameters, phase change microcapsule parameters and piezoelectric pump drive frequency parameters configured in S1 to S4, integrated manufacturing process parameters for the power circuit board heat dissipation structure are generated. The manufacturing process parameters include the temperature and pressure parameters of powder metallurgy stacked sintering, the groove shape parameters of femtosecond laser etching, the scanning speed and laser power parameters of selective laser melting, and the filling rate and sealing parameters of micro-injection filling.
[0018] As a preferred embodiment: the amplitude distribution of the pulsating extrusion pressure output by S1 is used as the input parameter of S2, the frequency and amplitude of the pressure wave output by S2 are used as the input parameter of S3, the fundamental frequency of the self-excited oscillation output by S2 is used as the input parameter of S4, and the output parameters of the previous stage are directly used as the input parameters of the next stage, forming a step-by-step driving parameter transfer relationship between each stage.
[0019] Compared with the prior art, the present invention has obvious advantages and beneficial effects. Specifically, as can be seen from the above technical solution: The first and fourth cascaded stages achieve a synergistic enhancement effect. S1 converts harmful thermal stress into useful driving force, S2 oscillates under the driving force and generates pressure waves, S3 absorbs heat quickly in synergy with the pressure waves, and S4 maintains the cycle efficiently through resonance. Forced physical coupling is formed between each stage, and the overall system's heat dissipation capacity far exceeds the arithmetic summation of the effects of each module.
[0020] Secondly, the pump power consumption is extremely low. The fluid self-excited oscillation of S2 is driven by the thermal stress captured by S1, and S4 only compensates for the system damping loss under resonance conditions. The pump power consumption of the entire system is less than or equal to 5W, which is more than 95% lower than the 80 to 150W of traditional liquid cooling solutions.
[0021] Third, the transient response speed has achieved a qualitative leap. Under the synergy of pressure waves, the response lag time of S3 phase change material has been shortened from 300 to 500 ms in the traditional scheme to less than or equal to 20 ms, which can effectively match the millisecond-level transient thermal shock conditions such as rapid acceleration of new energy vehicles.
[0022] Fourth, thermal cycling reliability is greatly improved. The S1 stress collector transforms the harmful thermal stress that would otherwise cause interface cracking into useful work that drives microchannel oscillation, significantly releasing residual stress at the interface and increasing thermal cycling life to over 5000 cycles.
[0023] Fifth, the fully passive adaptive system requires no complex electronic control. The system forms a purely physical positive feedback closed loop: the greater the heat generated, the greater the stress; the greater the stress, the stronger the oscillation; the stronger the oscillation, the faster the phase change material responds; the faster the phase change material responds, the higher the jet resonance frequency; and the higher the resonance frequency, the stronger the heat dissipation.
[0024] Sixth, the compact structure facilitates weight reduction. All heat dissipation functions are integrated inside and at the edge of the power circuit board substrate, eliminating the need for external large heat sinks and complex liquid cooling pipelines.
[0025] To more clearly illustrate the structural features and effects of the present invention, a detailed description is provided below in conjunction with the accompanying drawings and specific embodiments. Attached Figure Description
[0026] Figure 1This is a flowchart illustrating the overall method of optimizing the heat dissipation structure of the power circuit board according to the present invention.
[0027] Figure 2 This is a data flow diagram of the four-level cascaded collaborative parameter configuration chain S1 to S4 of the present invention.
[0028] Figure 3 This is a flowchart of the phase-locked loop frequency locking control in S4 of the present invention. Detailed Implementation
[0029] The present invention is as follows Figures 1 to 3 As shown, a heat dissipation structure optimization design method for a power circuit board in a new energy vehicle is described, wherein: Figure 1 This paper demonstrates the overall method flow of the power circuit board heat dissipation structure optimization design method of the present invention. Figure 1 As can be seen, the method includes the following steps in sequence: First, the heat source distribution and thermal expansion coefficient gradient distribution of the power chip are obtained, and the parameters of the stress gathering groove array are configured according to the heat source distribution and thermal stress distribution, including the layout area, groove width, groove depth, groove spacing and extension direction; Second, the wall thickness of the flexible thin-walled region is configured with the amplitude of the pulsating extrusion pressure in each region as input, so that the wall thickness of the flexible thin-walled region is inversely proportional to the amplitude of the pulsating extrusion pressure; Then, the shell wall material and core material composition of the phase change microcapsule are configured with the pressure wave frequency and amplitude as input, and the phase change response time is set; Finally, the fundamental frequency of the fluid self-excited oscillation is collected, and the frequency of the piezoelectric pump driven by the phase-locked loop is locked to the fundamental frequency. Figure 1 This demonstrates the complete process of the method from parameter input to parameter configuration completion.
[0030] Figure 2 This demonstrates the data flow of the four-level cascaded collaborative parameter configuration chain from S1 to S4. Figure 2 As can be seen, the amplitude distribution of the pulsating extrusion pressure output by S1 serves as the input parameter of S2, the frequency and amplitude of the pressure wave output by S2 serve as the input parameter of S3, the fundamental frequency of the self-excited oscillation output by S2 serves as the input parameter of S4, and the resonance drive signal output by S4 is fed back to the coolant circulation system. Figure 2 The parameter passing and causal relationships between different levels were explained from the perspective of data flow.
[0031] Figure 3 This demonstrates the detailed flow of the phase-locked loop frequency locking control in S4. (By...) Figure 3 As can be seen, the fundamental frequency f0 of the fluid self-excited oscillation is first collected by the pressure sensor. The phase-locked loop determines whether the current piezoelectric pump drive frequency f_p is equal to f0 and the phase difference Δφ ≤ ±5°. If so, the current drive frequency is maintained and monitoring continues. Otherwise, a frequency sweep search is started, and f_p is adjusted until the locking condition is met. After locking, the drive is maintained and monitoring continues. Figure 3The implementation process of frequency adaptation was explained from the perspective of control logic.
[0032] The present invention will be further described below with reference to specific embodiments.
[0033] Example 1:
[0034] This embodiment takes the power circuit board of a traction inverter for a certain type of new energy vehicle as an example and optimizes its heat dissipation structure.
[0035] In step S1, the heat source distribution and thermal expansion coefficient gradient distribution of the power chips are first obtained. The power circuit board contains six SiC power chips arranged in two rows and three columns, with each chip having a rated power of 200W and a total heat generation power of 1200W. Based on the layout and heat generation power distribution of the power chips, the hot spot region is determined to be located directly below the two chips in the middle column, where the heat flux density reaches 180W / cm². Based on the thermal expansion coefficient gradient distribution, the expected thermal stress amplitude for each region is determined. The region directly below the chips in the middle column is expected to have the highest thermal stress amplitude, reaching 45MPa; the edge region is expected to have a lower thermal stress amplitude, approximately 15MPa. A stress-collecting groove array is configured at the substrate interlayer interface between the power chips and the coolant channels. This substrate is a three-layer gradient SiCp / Al composite substrate, with a SiC volume fraction of 50% in the top layer, 30% in the middle layer, and 15% in the bottom layer. The stress current collectors employ a V-shaped groove structure with a groove width of 100 μm, a groove depth of 60 μm, and a groove spacing of 300 μm. The extension direction of the grooves forms a 45° angle with the chip arrangement direction. The density of the stress current collectors is positively correlated with the local heat flux density of the chip. The highest density is found in the central hotspot area of the chip (directly below the two chips in the middle row), with a groove spacing of 200 μm. The density decreases in the edge area, with a groove spacing of 500 μm.
[0036] When the chip generates heat and creates a temperature gradient, each gradient layer expands to different degrees due to differences in their coefficients of thermal expansion. The top layer has a coefficient of thermal expansion of approximately 7 ppm / K, while the bottom layer has a coefficient of thermal expansion of approximately 16 ppm / K. When the temperature difference reaches 80°C, in-plane shear thermal stress accumulates at the interlayer interface. The stress distribution channel directs and rectifyes the originally disordered thermal stress into periodic pulsating extrusion pressure perpendicular to the coolant channel wall. The amplitude of the pulsating extrusion pressure is positively correlated with the heat generated by the chip; the greater the heat generation, the greater the temperature difference, and the higher the amplitude of the pulsating extrusion pressure. The peak amplitude of the pulsating extrusion pressure in the hot spot area is approximately 38 MPa, with a frequency of approximately 200 Hz.
[0037] In S2, the amplitude of the pulsating extrusion pressure in each region of S1 is used as input, and the wall thickness of the flexible thin-walled region is configured at the position of the stress collection groove corresponding to each region. In the region where the expected amplitude of the pulsating extrusion pressure is the largest, i.e., directly below the middle row of chips, the wall thickness is 0.10 mm. In the edge region where the expected amplitude of the pulsating extrusion pressure is smaller, the wall thickness is 0.20 mm. The wall thickness of the flexible thin-walled region is inversely proportional to the amplitude of the pulsating extrusion pressure. The relationship between the wall thickness d_i and the amplitude of the pulsating extrusion pressure P_i is d_i = k·P_i⁻¹, where k is 3.8 mm·MPa. Calculations show that d = 3.8 / 38 = 0.10 mm in the hot spot region and d = 3.8 / 19 = 0.20 mm in the edge region, achieving a non-linear gain in the wall deformation depth under increased heat load, with the self-excited oscillation amplitude increasing positively with increasing heat generation. The wall thickness of the non-flexible region is 0.50 mm. The coolant channels employ an H-shaped fractal topology with three fractal stages. The main channel has a hydraulic diameter of 1.2 mm and a rectangular cross-section, measuring 1.2 mm wide and 1.0 mm deep. The secondary branch channels have a hydraulic diameter of 0.8 mm, and the final branch channel has a hydraulic diameter of 0.5 mm. The coolant is a 1:1 volume ratio of deionized water and ethylene glycol, with an inlet temperature of 40°C and an inlet flow rate of 2.5 L / min.
[0038] When pulsating extrusion pressure is applied to the flexible thin-walled region, the channel wall undergoes periodic inward deformation of approximately 15 μm, resulting in alternating contraction and recovery of the channel cross-sectional area, with a reduction of approximately 20%. This dynamic boundary condition causes the flowing coolant to automatically form vortex shedding and fluid resonance at a Reynolds number of approximately 250, generating self-sustaining fluid self-excited oscillations with an oscillation frequency of approximately 200 Hz and a pressure wave amplitude of approximately 0.5 MPa.
[0039] In S3, the frequency and amplitude of the pressure wave generated by the self-excited oscillation of the fluid in S2 are used as inputs to configure the parameters of the phase change microcapsules. High-density polyethylene is used as the shell wall material, with a shell wall thickness of 5 μm, and a paraffin and expanded graphite composite phase change material is used as the core material. The composite phase change material contains 80% paraffin by mass, 20% expanded graphite by mass, and has a thermal conductivity of 4.5 W / (m·K). The melting temperature of the phase change material is 90℃, and its latent heat is 180 kJ / kg. The microcapsule particle size is 50 μm, and the inner surface of the shell wall has a micron-sized pit structure with a pit diameter of 1 μm, a depth of 0.5 μm, and an areal density of 10. 8The microcapsules, arranged in units per cm², are used to generate a stress concentration effect under pressure wave action, further reducing the phase change nucleation energy barrier. Phase change microcapsules are encapsulated in copper-based microcavity arrays and positioned at the low-speed reflux blind ends and fractal branch ends of the coolant channels. Each microcavity is a cylindrical cavity with a diameter of 1.0 mm and a depth of 2.0 mm. The microcapsule filling rate within the cavity is 70%, with the remaining space filled with FC-3283 dielectric coolant as the pressure transmission medium. A total of 48 microcavity arrays are used, evenly distributed at the ends of six fractal branches, with 8 microcavities arranged at each branch end.
[0040] When the 200Hz pressure wave generated by S2 is transmitted to the surface of the microcapsule through the coolant medium, it generates alternating pulse pressure on the high-density polyethylene shell wall, with a pressure amplitude of approximately 0.5 MPa. This pulse pressure causes elastic deformation of the shell wall, with a volume change rate of approximately 1.5%, and the phase change material inside the shell is subjected to alternating adiabatic compression and expansion. According to the Clapeyron equation dT / dp = T·Δv / ΔH, where T is the phase change temperature, Δv is the volume change before and after the phase change, and ΔH is the latent heat of the phase change, the melting temperature decreases with increasing pressure; for every 0.1 MPa increase, the melting temperature decreases by approximately 1 °C. When the pressure wave amplitude is 0.5 MPa, the phase change temperature dynamically shifts by approximately 5 °C, causing the phase change material to trigger a solid-liquid phase change earlier at approximately 82 °C. This pressure-induced dynamic shift in phase transition temperature lowers the energy barrier for uniform nucleation of paraffin crystals, reducing the response hysteresis time of the phase change material to 15ms, satisfying the condition τ≤(2πf)⁻¹, where (2π×200)⁻¹≈0.8ms. The actual response time of 15ms matches the temperature rise rate of the transient thermal shock.
[0041] In S4, a microprocessor-controlled voltage pressure sensor with a response frequency of 2kHz and a pressure resolution of 10Pa is installed at the inlet of the coolant channel. The sensor acquires the fundamental frequency f0 of the fluid self-excited oscillation at the channel inlet, which is approximately 200Hz in this embodiment. The phase-locked loop circuit extracts this fundamental frequency in real time and forces the drive frequency f_p of the piezoelectric pump to lock to f0. The tracking frequency range of the phase-locked loop circuit is 50-900Hz, and the tracking rate is 150Hz / ms. When f_p and f0 are precisely matched, the phase difference is locked at 0°±5°, and a Helmholtz resonance enhancement effect is generated at the jet inlet, amplifying the flow amplitude at the inlet by 4 times, and the jet impact velocity reaches 10m / s. The piezoelectric pump adopts a multi-layer stacked piezoelectric ceramic structure with a total of 120 layers, each layer being 0.1mm thick and 12mm in total length. In the conventional non-resonant drive mode, the drive voltage is 120V and the output flow rate is 1.2L / min. In the resonant drive mode, the drive voltage drops to 25V, which is about 21% of the conventional non-resonant drive voltage, and the output flow rate reaches 4.8L / min. The pump power consumption is reduced to 3W, calculated by the formula P=V·I, where V=25V, I=0.12A, and P=3W.
[0042] In S5, integrated manufacturing process parameters are generated based on the parameters configured in S1 to S4. The powder metallurgy laminate sintering temperature is 620℃, the pressure is 40MPa, the holding time is 1.5 hours, and the vacuum degree is better than 10⁻³Pa. The V-groove parameters for femtosecond laser etching are: groove width 100μm, groove depth 60μm, groove spacing 300μm, laser wavelength 800nm, pulse width 120fs, and power 5W. The laser power for selective laser melting is 400W, the scanning speed is 1000mm / s, the powder layer thickness is 40μm, and the scanning speed is reduced to 500mm / s at the corresponding stress current collection groove to form a flexible thin-walled region. A thin-walled region with a thickness of 0.10-0.20mm is formed directly below the stress current collection groove. The micro-injection filling rate was 70%, the injection pressure was 0.6 MPa, the injection temperature was 120℃, the sealing method was laser welding, the laser power was 200W, and the welding speed was 10mm / s. The power chip was mounted on the top layer of the substrate by silver sintering at a temperature of 250℃, a pressure of 40 MPa, and a holding time of 10 minutes.
[0043] With the complete parameter configuration of S1 to S5 above, the total thermal resistance (from chip junction temperature to coolant) of the power circuit board heat dissipation structure designed in this embodiment is about 1.5K / W, the highest chip junction temperature is controlled below 125℃ (coolant inlet temperature 40℃), and the temperature non-uniformity between chips is controlled within ±3℃, which meets the heat dissipation requirements of the traction inverter of new energy vehicles under continuous high power output and frequent transient impact conditions.
[0044] Example 2: The difference between this embodiment and Embodiment 1 lies in the use of a U-shaped trench structure for the stress current collection groove in S1. The power circuit board in this embodiment contains four SiC power chips arranged in a single row of four columns. Each chip has a rated power of 300W, resulting in a total heat dissipation of 1200W. The chip layout is more compact, with an adjacent chip spacing of only 5mm, leading to a more concentrated hotspot area. The heat flux density directly beneath the two middle chips reaches 220W / cm². The bottom of the U-shaped trench is arc-shaped, with a radius half the width of the trench. This reduces the stress concentration factor by approximately 30% compared to a V-shaped trench, making it suitable for applications with higher thermal stress amplitudes. The trench width is 150μm, the trench depth is 80μm, the trench spacing is 400μm, and the trench extension direction forms a 45° angle with the chip arrangement direction. The substrate is made of four-layer gradient SiCp / Al composite material. The volume fractions of SiC from the side closest to the chip to the side furthest from the chip are 55%, 40%, 25%, and 10%, respectively. Each layer has a thickness of 0.4 mm, and the total thickness is 1.6 mm. U-shaped stress current collection grooves are provided at the interlayer interfaces.
[0045] In S2, the wall thickness of the flexible thin-walled region is 0.12 mm in the area of maximum pulsating extrusion pressure, while the wall thickness of the non-flexible region is 0.55 mm. The coolant channel has four fractal stages: the hydraulic diameter of the main channel is 1.5 mm, the secondary branch is 1.1 mm, the tertiary branch is 0.7 mm, and the final branch is 0.4 mm, resulting in a more complex structure and more uniform flow distribution. The coolant is perfluorinated liquid FC-72, with an inlet temperature of 35℃ and an inlet flow rate of 1.8 L / min.
[0046] In S3, the phase change microcapsule shell material is urea-formaldehyde resin with a shell thickness of 3 μm, which is 40% thinner than that in Example 1, resulting in higher pressure wave transmission efficiency. The phase change material has a melting temperature of 85℃, a latent heat of 165 kJ / kg, and a microcapsule particle size of 40 μm. The microcapsules are encapsulated in a copper-based microcavity array plate, with a microcavity diameter of 0.8 mm and a depth of 1.5 mm, resulting in a microcapsule filling rate of 75%. The total number of microcavities is 72, evenly distributed at the ends of eight fractal branches, with 9 microcavities arranged at each branch end.
[0047] In S4, the phase-locked loop circuit tracks frequencies ranging from 50 to 900 Hz, with a tracking rate of 150 Hz / ms. The piezoelectric pump drive voltage is reduced to 20V, approximately 17% of the conventional non-resonant drive voltage of 120V, and pump power consumption is reduced to 2.5W.
[0048] In this embodiment, the fluid self-excited oscillation frequency generated by S2 is approximately 350 Hz, and the pressure wave amplitude is approximately 0.8 MPa. According to the Clapeyron equation, the melting temperature decreases by approximately 0.8 °C for every 0.1 MPa increase. The 0.8 MPa pressure shift causes the phase change material to trigger a phase change earlier at approximately 79 °C, shortening the response hysteresis time to 10 ms, satisfying the condition τ ≤ (2πf)⁻¹, where (2π × 350)⁻¹ ≈ 0.45 ms.
[0049] In applications requiring higher power density and a more compact layout, this embodiment provides finer thermal stress matching through a four-layer gradient substrate, more uniform flow distribution through four-level fractal channels, and a shorter phase change response time (10ms) through pressure wave triggering at a higher frequency (350Hz), resulting in better overall heat dissipation.
[0050] Example 3: The difference between this embodiment and Embodiment 1 lies in the precise inverse relationship between the wall thickness d_i of the flexible thin-walled region and the amplitude of the pulsating extrusion pressure P_i in S2: d_i = k·P_i⁻¹, where k is a material constant. For the AlSi10Mg material used in this embodiment by selective laser melting, k is 4.2 mm·MPa. In the region where the expected amplitude of the pulsating extrusion pressure is the largest, i.e. directly below the chip, P_i = 42 MPa, and the wall thickness is d_i = 4.2 / 42 = 0.10 mm, resulting in the highest deformation sensitivity. In the region where the expected amplitude of the pulsating extrusion pressure is smaller, P_i = 21 MPa, and the wall thickness is d_i = 4.2 / 21 = 0.20 mm. This precise inverse relationship ensures that the relationship between the wall deformation depth δ and the amplitude of the pulsating extrusion pressure P_i is δ = α·P_i·d_i, where α is the material elastic coefficient. For AlSi10Mg, α ≈ 0.015 mm² / N. Since d_i is inversely proportional to P_i, δ has an approximately linear relationship with P_i rather than an inverse relationship, i.e., δ≈α·k. Maintaining a constant deformation of approximately 15μm, the deformation frequency increases with increasing P_i, and the gain coefficient is 2.5. This design ensures a positive feedback characteristic where the amplitude of the self-excited oscillation is positively amplified as heat generation increases.
[0051] In S3, the pits on the inner surface of the phase change microcapsule shell have a diameter of 2 μm, a depth of 1 μm, and an areal density of 5 × 10⁻⁶. 7 The number of indentations per cm² is doubled compared to Example 1, which is used to match the pressure wave excitation at a higher frequency (approximately 250 Hz), making the stress concentration effect effective over a wider frequency range.
[0052] Example 4: The difference between this embodiment and Embodiment 1 is that the inner surface of the shell of the phase change microcapsule in S3 is provided with a micron-sized pit structure, with a pit diameter of 1 μm, a depth of 0.5 μm, and an areal density of 10. 8 The structure generates a stress concentration effect under the action of pressure waves, which amplifies the local stress of the shell wall by about 3 times, further reducing the nuclear energy barrier of phase change, enabling the phase change material to trigger the phase change earlier at a temperature of about 78°C, and shortening the response lag time to 10ms.
[0053] In this embodiment, the core material of the phase change microcapsules is a composite phase change material of paraffin and expanded graphite, with the mass fraction of expanded graphite increased to 25% and the mass fraction of paraffin decreased to 75%, resulting in an increased thermal conductivity of 5.2 W / (m·K). The porous structure of the expanded graphite undergoes minute elastic deformation of the skeleton under pressure wave action, further enhancing the compression effect of the pressure wave on the paraffin and making the pressure-induced dynamic shift in phase change temperature more significant. The microcapsule particle size is 35 μm, the shell wall thickness is 4 μm, and the microcapsule filling rate is 72%.
[0054] In S4, the phase-locked loop circuit extends the range of the piezoelectric pump drive frequency to 50-1200Hz, with a tracking rate of 200Hz / ms, enabling faster tracking of the rapid drift of the fundamental frequency under high power density conditions. The piezoelectric pump's drive voltage in resonant drive mode is 18% of the conventional non-resonant drive voltage, and its pump power consumption is 2.8W.
[0055] The phase change material response hysteresis time (10ms) in this embodiment is further shortened compared to that in Embodiment 1 (15ms), making it suitable for more severe transient thermal shock conditions.
[0056] Example 5: The difference between this embodiment and Embodiment 1 lies in the refinement of the phase-locked loop (PLL) frequency locking control strategy in S4. In this embodiment, the power circuit board's heating conditions are divided into three modes: light load mode (total heating power less than 400W), medium load mode (400W to 800W), and heavy load mode (greater than 800W). The fundamental frequency f0 of the fluid self-excited oscillation in S2 differs between these modes: approximately 120Hz in light load mode, approximately 200Hz in medium load mode, and approximately 350Hz in heavy load mode. Three frequency locking thresholds are preset in the PLL circuit. When a frequency drift is detected exceeding ±15% of the currently locked frequency, a mode switch is automatically determined. When the frequency is lost, the PLL scans the frequency from 10Hz to 1200Hz to re-search for and lock a new fundamental frequency. The scan step size is 5Hz, the dwell time per step is 10ms, and the time to complete the full-band search is approximately 2.4 seconds. This scan process is automatically executed during the cold start phase without external intervention. After locking, the phase-locked loop continuously monitors frequency changes. When the change in heat load causes the fundamental frequency to drift beyond the locking bandwidth (±10%), the frequency sweep locking process is automatically re-executed, and the re-locking time is less than 500ms.
[0057] The key design feature of this invention is to achieve step-by-step optimization of each component of the heat dissipation structure through a four-level cascaded collaborative parameter configuration chain. The first level, based on the heat source distribution and thermal expansion coefficient gradient distribution of the power chip, configures a stress-collecting trench array at the substrate interlayer interface. The trench width is 50-200 μm, the trench depth is 30-100 μm, and the trench spacing is 200-500 μm, ensuring a positive correlation between the trench density and the local heat flux density. This transforms the in-plane thermal shear stress into periodic pulsating extrusion pressure. The trench extension direction forms a 30°-60° angle with the chip arrangement direction. The second stage takes the amplitude of the pulsating extrusion pressure P_i of each region output from the first stage as input, and configures the wall thickness d_i of the flexible thin-walled region at the position of the stress collection channel corresponding to each region. The wall thickness is inversely proportional to the amplitude of the pulsating extrusion pressure, d_i=k·P_i⁻¹. The minimum wall thickness value of 0.10-0.20mm is configured in the region with the largest amplitude of pulsating extrusion pressure, and the wall thickness of the non-flexible region is 0.40-0.60mm. This allows the coolant to generate self-sustaining fluid self-excited oscillation in the channel of the H-shaped fractal topology (fractal level 3-4, hydraulic diameter 0.6-1.5mm). The inverse wall thickness configuration ensures that the wall deformation depth exhibits a nonlinear gain when the heat load is increased.
[0058] In addition, the third stage takes the pressure wave frequency f and amplitude output by the second stage as input, and configures the shell wall material and core material of the phase change microcapsule. The shell wall material is made of high-density polyethylene or urea-formaldehyde resin with a wall thickness of 2-10μm. The core material is a composite phase change material of paraffin (75%-85%) and expanded graphite (15%-25%). The microcapsule particle size is 30-100μm and the melting temperature of the phase change material is 80-100℃. The phase change material is triggered in advance by the dynamic shift of the phase change temperature induced by pressure. The phase change response time τ satisfies τ≤(2πf)⁻¹ with the pressure wave frequency f, so that the phase change response time matches the temperature rise rate of the transient thermal shock. The fourth stage acquires the fundamental frequency f0 of the fluid self-excited oscillation output from the second stage. A phase-locked loop (PLL) circuit forces the piezoelectric pump drive frequency f_p to lock onto this fundamental frequency. The PLL tracks frequencies from 50-900Hz with a tracking rate ≥100Hz / ms. The frequency locking condition is f_p=f0 and a phase difference Δφ≤±5°. After locking, the piezoelectric pump drive voltage drops to 15%-20% of the conventional non-resonant drive voltage, generating a Helmholtz resonance enhancement effect at the jet inlet to maintain coolant circulation. When the frequency is lost, the PLL re-searches for and locks onto a new fundamental frequency using a frequency sweep method. The four-stage parameter configurations form a chain-like dependency relationship where the output parameters of the previous stage directly serve as the input parameters of the next stage, constituting a complete closed-loop system for the cascade utilization of thermo-mechanical-fluid-phase change energy.
[0059] The above description is merely a preferred embodiment of the present invention and does not constitute any limitation on the technical scope of the present invention. Therefore, any minor modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention shall still fall within the scope of the technical solution of the present invention.
Claims
1. A method for optimizing the heat dissipation structure of a power circuit board for new energy vehicles, characterized in that, Includes the following steps: S1. Stress collection groove array parameter configuration: Obtain the heat source distribution and thermal expansion coefficient gradient distribution of the power chip, determine the layout area of the stress collection groove according to the heat source distribution, determine the expected thermal stress amplitude of each area according to the thermal expansion coefficient gradient distribution, configure V-shaped or U-shaped grooves as stress collection grooves at the substrate interlayer interface between the power chip and the coolant channel, and convert the in-plane thermal shear stress generated by the power chip into periodic pulsating extrusion pressure, which serves as the power source for driving the heat dissipation structure. S2. Flexible thin-walled region wall thickness parameter configuration: Taking the amplitude of the pulsating extrusion pressure in each region in S1 as input, the wall thickness of the flexible thin-walled region is configured at the position of the stress collection groove in each region. The wall thickness of the flexible thin-walled region is inversely proportional to the amplitude of the pulsating extrusion pressure, so that the flexible thin-walled region undergoes periodic deformation under the action of the pulsating extrusion pressure, resulting in the alternating contraction and expansion of the cross-sectional area of the coolant channel, causing the coolant to generate self-sustaining fluid self-excited oscillation. S3. Phase change microcapsule parameter configuration: Taking the frequency and amplitude of the pressure wave generated by the fluid self-excited oscillation in S2 as input, configure the shell wall material, shell wall thickness and core material composition of the phase change microcapsule, so that the shell of the microcapsule undergoes elastic deformation under the action of pressure wave, and the phase change material is triggered in advance under the condition of lower than the conventional melting temperature by pressure-induced dynamic shift of phase change temperature, and the phase change response time is set based on the pressure wave frequency. S4. Piezoelectric pump drive frequency configuration: Collect the fundamental frequency of fluid self-excited oscillation at the inlet of the coolant channel in S2, use this fundamental frequency as the locking reference for the piezoelectric pump drive frequency, and force the drive frequency of the piezoelectric pump to be locked to this fundamental frequency through the phase-locked loop circuit, so as to generate a resonance enhancement effect at the jet inlet to maintain coolant circulation.
2. The heat dissipation structure optimization design method for the power circuit board of a new energy vehicle according to claim 1, characterized in that: In S1, the density of the stress current collector grooves is positively correlated with the local heat flux density of the power chip, and the peak area of the groove density corresponds to the hot spot area of the chip. The groove width of the stress current collector groove is 50-200μm, the groove depth is 30-100μm, the groove spacing is 200-500μm, and the extension direction of the groove is at an angle of 30°-60° with the chip arrangement direction.
3. The heat dissipation structure optimization design method for the power circuit board of a new energy vehicle according to claim 1, characterized in that: In S2, the wall thickness of the flexible thin-walled region is 0.10-0.20 mm, and the wall thickness of the non-flexible region is 0.40-0.60 mm. The wall thickness d_i of the flexible thin-walled region is inversely proportional to the amplitude of the pulsating extrusion pressure P_i. The minimum wall thickness value is configured in the region where the amplitude of the expected pulsating extrusion pressure is the largest.
4. The heat dissipation structure optimization design method for the power circuit board of a new energy vehicle according to claim 1, characterized in that: In S2, the coolant channel adopts an H-shaped fractal topology with 3-4 fractal levels and a hydraulic diameter of 0.6-1.5 mm.
5. The heat dissipation structure optimization design method for the power circuit board of a new energy vehicle according to claim 1, characterized in that: In step S3, the shell wall material of the phase change microcapsules is high-density polyethylene or urea-formaldehyde resin, with a shell wall thickness of 2-10 μm. The core material is a composite phase change material of paraffin and expanded graphite, wherein the mass fraction of paraffin is 75%-85%, the mass fraction of expanded graphite is 15%-25%, the microcapsule particle size is 30-100 μm, and the melting temperature of the phase change material is 80-100℃.
6. The heat dissipation structure optimization design method for the power circuit board of a new energy vehicle according to claim 1, characterized in that: In S3, the phase change response time τ and the pressure wave frequency f satisfy τ≤(2πf)⁻¹, so that the phase change response time matches the temperature rise rate of the transient thermal shock.
7. The heat dissipation structure optimization design method for the power circuit board of a new energy vehicle according to claim 1, characterized in that: In S4, the phase-locked loop circuit tracks the fundamental frequency of the fluid self-excited oscillation in the range of 50-900Hz, and the tracking rate is greater than or equal to 100Hz / ms; the driving voltage of the piezoelectric pump in the resonant driving mode is 15%-20% of the conventional non-resonant driving voltage.
8. The heat dissipation structure optimization design method for the power circuit board of a new energy vehicle according to claim 1, characterized in that: In S4, the frequency locking condition of the phase-locked loop circuit is: the piezoelectric pump driving frequency f_p is equal to the fluid self-excited oscillation fundamental frequency f_0, and the phase difference Δφ ≤ ±5°; when the frequency is lost, the phase-locked loop searches for and locks a new fundamental frequency again in a frequency sweep manner.
9. The heat dissipation structure optimization design method for the power circuit board of a new energy vehicle according to claim 1, characterized in that: It also includes S5, integrated manufacturing parameter configuration: based on the stress collection groove parameters, flexible thin-wall region wall thickness parameters, phase change microcapsule parameters and piezoelectric pump drive frequency parameters configured in S1 to S4, the integrated manufacturing process parameters of the power circuit board heat dissipation structure are generated. The manufacturing process parameters include the temperature and pressure parameters of powder metallurgy stacked sintering, the groove shape parameters of femtosecond laser etching, the scanning speed and laser power parameters of selective laser melting, and the filling rate and sealing parameters of micro-injection filling.
10. The heat dissipation structure optimization design method for the power circuit board of a new energy vehicle according to claim 1, characterized in that: The amplitude distribution of the pulsating extrusion pressure output by S1 is used as the input parameter of S2. The frequency and amplitude of the pressure wave output by S2 are used as the input parameter of S3. The fundamental frequency of the self-excited oscillation output by S2 is also used as the input parameter of S4. The output parameters of the previous stage are directly used as the input parameters of the next stage, forming a step-by-step driving parameter transfer relationship between each stage.
