A thermal stress equalization design method and module layout of a power supply modular system
By using a thermoelectric synergistic coupling network and dynamic load-thermal stress dual closed-loop control, the problems of electro-thermal decoupling, slow response, weak protection, poor adaptability and insufficient intelligence in the thermal management of power modules are solved, achieving efficient and fast thermal management and overload protection, and adapting to complex scenarios.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- TAIYUAN YONGMING HENGDONGYUAN ELECTRONICS CO LTD
- Filing Date
- 2025-08-18
- Publication Date
- 2026-07-03
AI Technical Summary
Existing thermal management technologies for power modules suffer from problems such as efficiency bottlenecks due to electro-thermal decoupling, delayed response of passive thermal management, inefficient overload protection mechanisms, insufficient environmental adaptability, and low level of intelligence.
It adopts a thermoelectric synergistic coupling network design, realizes low-loss power transmission and efficient heat dissipation through heat pipe-conductor composite structure, combines dynamic load-thermal stress dual closed-loop control and phase change cooling, integrates edge computing unit and machine learning algorithm for real-time thermal trend prediction and load optimization, and uses mechanical limit device for overload protection.
It achieves deep coupling of power transmission and heat dissipation, dynamically adjusts module power, quickly responds to overload conditions, improves system stability and adaptability, adapts to high power density, wide temperature range and complex scenarios, and broadens the application range.
Smart Images

Figure CN121031326B_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of power supply circuit technology, and particularly relates to a thermal stress equalization design method and module layout for a modular power supply system. Background Technology
[0002] Current power module thermal management technologies primarily rely on an architecture where power transmission and thermal management are designed independently. In traditional solutions, power transmission is achieved via copper busbars or wires, while the cooling system typically employs independent aluminum heat sinks, forced air cooling, or liquid cooling devices. For example, data center servers commonly use air-cooled heat sinks combined with thermal grease to reduce temperature by increasing the heat dissipation area and air convection; new energy vehicle power supplies often use liquid cooling plates combined with circulating coolant to remove heat. These technologies can meet basic heat dissipation requirements under steady-state conditions; for example, the thermal resistance of a single-sided heat sink can be controlled at 0.452 K / W, and liquid cooling systems can further reduce it to 0.203 K / W. In addition, some solutions improve heat dissipation efficiency by optimizing the packaging structure (such as double-sided heat-dissipating PCBs and integrated heat sinks), achieving stable operation within a temperature range of -20℃ to 85℃.
[0003] While existing technologies are effective in typical scenarios, their design principles have fundamental flaws:
[0004] Electro-thermal decoupling leads to efficiency bottlenecks: the electrical transmission path and heat dissipation channel are independent of each other, and the heat generated by power loss needs to be transferred to the external heat sink through multiple layers of interface materials, resulting in high system thermal resistance (e.g., the thermal resistance of traditional chip heat dissipation reaches 0.3K / W), making it difficult to meet 50kW / m 3 The above high power density requirements.
[0005] Passive thermal management suffers from lag in response: lacking a real-time thermal stress feedback mechanism, it cannot dynamically adjust load distribution. For example, traditional air-cooled systems have a response time exceeding 1 second during sudden load changes, and the temperature difference between modules can easily exceed 15°C, leading to localized hot spots that accelerate device aging.
[0006] The overload protection mechanism is inefficient: it relies on passive fuses (fusing time > 500ms) and external temperature control switches, which cannot quickly trigger protection under instantaneous high current impact, and lacks active thermal stress regulation capability, resulting in a high failure rate.
[0007] Insufficient environmental adaptability: Limited wide temperature range (e.g. -20℃ to 85℃), requiring additional heating / heat dissipation devices in extreme low or high temperature environments, and unable to adapt to special scenarios such as microgravity and salt spray.
[0008] Low level of intelligence: Traditional solutions lack edge computing and machine learning support, making it difficult to predict heat trends and optimize load strategies, and failing to meet the dynamic balancing requirements under high-frequency operating conditions. Summary of the Invention
[0009] To overcome the above-mentioned defects of the prior art, embodiments of the present invention provide a thermal stress equalization design method and module layout for a power modular system, which solves the problems of electrical-thermal decoupling, slow response, weak protection, poor adaptability and insufficient intelligence in the thermal management of existing power modules.
[0010] To achieve the above objectives, the present invention provides the following technical solution:
[0011] A thermal stress equalization design method for a modular power supply system includes the following:
[0012] Construction of thermoelectric co-coupling network
[0013] The high-current bus between power modules is designed as a heat pipe-conductor composite structure, with an inner conductive core (copper / aluminum / silver-based, etc., with conductivity ≥4×10⁻⁶). 7 The heat pipe (S / m) is integrated with the outer heat pipe (microchannel / pulsating / loop heat pipe, etc.), and the working fluid of the heat pipe is driven by the Joule heating of the current, so as to realize low power loss transmission and efficient heat dissipation at the same time, with a thermal conductivity of ≥800W / mK.
[0014] Dynamic load-thermal stress dual closed-loop control
[0015] The distributed temperature sensors collect module thermal stress data in real time, and the edge computing unit is equipped with machine learning algorithms (LSTM / reinforcement learning, etc.) to predict thermal trends. When the module temperature exceeds the threshold, the power balancing circuit automatically transfers the load to the low-temperature module, forming a closed loop of "thermal stress detection-load distribution-heat flow control". The temperature difference fluctuation between modules is ≤5℃.
[0016] Phase change cooling and mechanical limit integration
[0017] A low-melting-point metal alloy phase change unit (melting point ≤200℃, latent heat ≥25kJ / kg) is embedded in the bus junction node. When a sudden large current causes local overheating, the alloy melts and absorbs the latent heat, triggering a mechanical limit device driven by shape memory alloy (SMA). The overload circuit is cut off within 100ms to avoid thermal stress concentration and device damage.
[0018] A modular layout structure for a power supply modular system includes the following:
[0019] Heat pipe busbar
[0020] Employing a composite structure of conductive core and heat pipe, it supports liquid cooling-heat pipe hybrid heat dissipation or graphene radiation-enhanced coating, with a busbar cross-sectional area ≥50mm². 2 Through nano-silver sintering process (contact thermal resistance ≤0.05℃·cm), 2 / W) to achieve efficient thermal coupling, adaptable to 30-50kW / m 3 High power density layout.
[0021] Intelligent load balancing module
[0022] Each module has a built-in edge computing unit that enables thermal stress data sharing and dynamic load adjustment via CAN bus or wireless communication. The neural network model optimizes the allocation strategy in real time with a response time of ≤100ms and supports plug-and-play distributed nodes and topology reconfiguration.
[0023] Thermal field dynamic control expansion
[0024] Integrating infrared vision sensing devices and actuators (parallel robots / robotic arms), it scans the module temperature distribution in real time (accuracy ±0.1℃), automatically adjusts the module spacing and angle, or inserts thermally conductive bridge components, and completes system-level thermal field equalization reconstruction within 2 minutes. It is suitable for dynamic scenarios such as reconfigurable satellite power supplies.
[0025] The technical effects and advantages of the thermal stress equalization design method and module layout for a modular power supply system of the present invention are as follows:
[0026] 1. This invention features a thermoelectric integrated design that achieves deep coupling of power transmission and heat dissipation through a heat pipe-conductor composite bus, breaking through the limitations of traditional independent design of power transmission and thermal management, significantly reducing system thermal resistance and improving heat dissipation efficiency.
[0027] 2. This invention features dynamic thermal stress balancing. Based on a load distribution mechanism using thermal stress feedback, it adjusts module power in real time. Combined with the thermal coupling effect of heat pipes, it effectively reduces the temperature difference between modules, avoids local overheating, and improves the stability of system operation.
[0028] 3. This invention provides efficient overload protection. The phase change cooling node integrates a low-melting-point alloy, which, through the dual action of latent heat absorption and mechanical limiting, quickly responds to overheating problems caused by instantaneous large currents, shortens the overload protection time, and enhances system safety.
[0029] 4. This invention features intelligent regulation and adaptability. The edge computing unit, combined with machine learning algorithms, enables heat trend prediction and load strategy optimization, supports dynamic reconfiguration of module layout, and improves the system's adaptability to high-frequency load fluctuations and extreme operating conditions.
[0030] 5. This invention has broad applicability across multiple scenarios. Through the synergistic optimization of materials, structure, and control strategies, it covers complex scenarios such as high power density, wide temperature range, and microgravity, thus expanding the application scope of modular power supply systems and improving overall performance and reliability. Attached Figure Description
[0031] Figure 1 This is a flowchart illustrating the thermal stress balancing design method and module layout of a power modular system proposed in this invention. Detailed Implementation
[0032] 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.
[0033] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.
[0034] Example 1
[0035] refer to Figure 1 This embodiment provides a thermal stress balancing design method and module layout for a modular power supply system, used in the implementation of an 800V high-voltage platform power supply system for new energy vehicles. Specific implementation details include:
[0036] Objective: To address the issues of concentrated Joule heat on the busbar, excessive temperature difference between modules, and slow overload protection response in high-voltage fast charging scenarios, thereby improving the safety and efficiency of battery pack charging and discharging.
[0037] Implementation steps:
[0038] Preparation of thermoelectric synergistic composite busbar:
[0039] Conductive core: Made of 6061-T6 aluminum alloy, extruded into a cross-sectional area of 80mm². 2 The rectangular array is plated with a 5μm silver layer to reduce the contact resistance to below 10μΩ;
[0040] Loop heat pipe: The inner wall of the copper shell (1mm thick) is sintered with copper powder capillary structure (porosity 50%), filled with deionized water (filling rate 60%), and integrated with the conductive core by vacuum welding to form a composite busbar with thermal conductivity ≥1200W / mK.
[0041] Intelligent load balancing system integration:
[0042] Each power module has a built-in STM32H7 edge computing unit, equipped with a trained LSTM model (input layer is temperature, current and voltage data, output layer is load distribution coefficient), and collects module temperature in real time via CAN bus (accuracy ±0.5℃).
[0043] When the module temperature exceeds 85℃, the power balancing circuit (bidirectional DC-DC converter, efficiency ≥97%) will transfer part of the load to the low-temperature module within 50ms, realizing dynamic matching between electrical load and thermal load.
[0044] Phase change cooling node installation:
[0045] SnBi58 alloy phase change unit (size 10mm×10mm×5mm, melting point 138℃, latent heat 56kJ / kg) is embedded in the main bus of the battery pack and connected to the busbar through nano-silver sintering process (250℃, 30min).
[0046] The peripherally integrated mechanical limiting device driven by Ni-Ti shape memory alloy wire cuts off the overload circuit within 0.1ms when the phase change volume expansion rate reaches 10% (corresponding to a temperature of 140℃).
[0047] Implementation results:
[0048] Thermal stress equalization: Under 3C fast charging conditions, the maximum temperature difference between modules is reduced from 18℃ to 2.5℃, the bus temperature rise rate is reduced from 15℃ / s to 4℃ / s, and the thermal resistance is reduced by 40% compared with the traditional solution;
[0049] Overload protection: Under an instantaneous short-circuit current of 2500A, the phase change unit triggers the limit switch within 30ms, which is 6 times faster than traditional fuses, thus preventing thermal runaway of the battery pack;
[0050] System efficiency: Charge and discharge efficiency increased from 93% to 95.5%, and the equivalent driving range of the 70kWh battery pack increased by 12km.
[0051] Example 2
[0052] This embodiment provides a thermal stress equalization design method and module layout for a modular power supply system, used in a data center with a power supply capacity of 50kW / m². 3 The implementation of a high-density UPS system includes the following specific implementation details:
[0053] Objective: To overcome the problem of localized overheating caused by high-density stacking, achieve low-energy thermal management under high power density, and reduce the PUE value of data centers.
[0054] Implementation steps:
[0055] Hybrid liquid cooling and heat pipe heat dissipation structure:
[0056] Microchannel heat pipe: A 2mm wide and 0.5mm deep microchannel is etched onto a 3mm thick copper substrate, connected to a 5mm thick copper base busbar (100mm² cross-sectional area). 2 Reflow soldering integration, with surface brazed 3mm inner diameter liquid cooling channels (10mm spacing);
[0057] The liquid cooling system uses 3M fluorinated liquid (boiling point 56℃) and is driven by a centrifugal pump (head 10m). The flow rate is dynamically adjusted by a PID algorithm according to the module temperature gradient (adjustment accuracy ±5%).
[0058] Thermal field dynamic reconfiguration system:
[0059] A FLIRA655sc infrared camera (640×480 resolution, ±0.05℃ accuracy) is mounted on the top to generate a 3D thermal field cloud map in real time, and the YOLOv5 algorithm identifies overheated areas (temperature > 75℃).
[0060] After receiving the instruction, the parallel robot (DobotMagician, positioning accuracy ±0.5mm) adjusts the module spacing (maximum ±10mm) within 5 seconds and inserts a graphene thermally conductive bridge (thickness 0.2mm, thermal conductivity 1500W / mK).
[0061] Two-stage phase change cooling node:
[0062] The inner layer of the busbar junction is made of In52Sn48 alloy (melting point 118℃, latent heat 45kJ / kg), and the outer layer is wrapped with paraffin wax (melting point 58℃, latent heat 210kJ / kg). It is sealed by an aluminum shell, and the volume expansion rate is controlled to ≤12% by elastic gaskets.
[0063] Implementation results:
[0064] Heat dissipation efficiency: Module density 50kW / m 3 At that time, the maximum temperature rose from 90℃ to 65℃, the local hot spot elimination rate reached 95%, and the heat flux density increased to 200W / cm². 2 ;
[0065] Energy efficiency improvement: The power consumption of the cooling system has decreased from 15% to 5%, and the PUE value has decreased from 1.25 to 1.08, saving 800,000 yuan in electricity costs per 1MW data center per year;
[0066] Response speed: It takes 12 seconds from overheat detection to layout adjustment, which is 50 times faster than manual intervention and supports 24-hour uninterrupted high-load operation.
[0067] Example 3
[0068] This embodiment provides a thermal stress equalization design method and module layout for a modular power supply system, used in the implementation of high-reliability power supply systems for aerospace applications. Specific implementation details include:
[0069] Objective: To meet the long-life thermal management requirements of deep space exploration missions under extreme temperature variations (-100℃~1200℃) and microgravity environments.
[0070] Implementation steps:
[0071] High-temperature heat pipe-conductor composite structure:
[0072] Sodium heat pipe: Haynes 230 alloy cladding (resistant to 1200℃ high temperature), inner wall spiral groove capillary structure (depth 0.3mm), filled with metallic sodium (filling rate 70%), thermal conductivity ≥2000W / mK;
[0073] Conductive core: Uranium-molybdenum alloy (density 18.9 g / cm³) 3 Electrical conductivity 2.5×10 7 The surface of the S / m is coated with a 3μm iridium layer for radiation resistance and integrated with the heat pipe by electron beam welding.
[0074] Passive thermal equilibrium design:
[0075] The reactor fuel rods and power modules are connected by an NTC thermal bridge (made of manganese dioxide). For every 10°C increase in temperature, the thermal resistance automatically increases by 20%, achieving power self-balancing.
[0076] The integrated semi-Hersler thermoelectric generator (ZT value 1.2) converts waste heat into electrical energy for the control circuit with an efficiency of 8%.
[0077] Fault-tolerant phase-change unit:
[0078] Gallium-based liquid metal (melting point 29.8℃) is embedded in the heat pipe evaporation section. When the working fluid leaks and causes local overheating (>150℃), the liquid metal expands and triggers a microswitch, cutting off non-critical loads within 0.1ms.
[0079] Implementation results:
[0080] Extreme environment adaptability: Under microgravity, the module temperature difference is stable at 1.5℃, which is 3 times better than the traditional fluid loop solution, and the performance degradation is less than 5% under temperature change of -100℃ to 1200℃.
[0081] Long life and reliable: Within a 15-year lifespan, the heat pipe heat transfer efficiency decays by less than 5%, and there is no false triggering of the phase change unit, meeting the requirements of deep space missions such as Jupiter exploration.
[0082] Lightweight design: The system weight is reduced by 30% compared to the traditional solution (1.2 tons → 0.84 tons), and the equivalent load is increased by 15%.
[0083] Example 4
[0084] This embodiment provides a thermal stress balancing design method and module layout for a modular power supply system, used for implementing dynamic load balancing in distributed energy systems. Specific implementation details include:
[0085] Objective: To address the problem of thermal stress concentration caused by uneven load distribution among distributed photovoltaic / energy storage nodes and improve the stability of microgrids.
[0086] Implementation steps:
[0087] Fabrication of Pulsating Heat Pipe-Silver-Based Busbar:
[0088] The pulsating heat pipe uses a 1mm inner diameter glass capillary tube filled with gallium indium tin alloy (50% by volume of working fluid), bent into a serpentine shape and attached to a silver-based busbar (60mm² cross-sectional area). 2 Electrical conductivity 6.3×10 7 S / m), fixed with 3W / mK thermally conductive silicone;
[0089] The modular interface uses a magnetically coupled connector (95% transmission efficiency) and supports plug-and-play and topology reconfiguration.
[0090] Blockchain collaborative control:
[0091] The edge computing unit runs the Hyperledger Fabric protocol to share temperature and load data (AES-256 encrypted), and the BP neural network (3 layers, 20 hidden nodes) updates the load distribution strategy every 10ms.
[0092] By adjusting the output power of each inverter using solid-state relays (response time 1ms), the power balance error between nodes is reduced to less than 2%.
[0093] Phase change energy storage node integration:
[0094] The photovoltaic combiner box is equipped with a Bi43Sn48In9 alloy phase change unit (20mm×20mm×3mm, melting point 117℃, latent heat 50kJ / kg), which is connected to the busbar by spring-type crimp terminals to absorb heat from the day-night temperature difference.
[0095] Implementation results:
[0096] Load balancing accuracy: The thermal stress imbalance between distributed nodes is reduced from 30% to 5%, and load redistribution is completed within 20ms under the islanding effect to avoid single-point overheating;
[0097] Energy utilization rate: The phase change unit recovers 12% of the diurnal temperature difference energy, the system energy efficiency is improved by 10%, and the annual power generation of a 100kW distributed power station is increased by 4%;
[0098] Communication reliability: Blockchain data sharing reduces fault location time to 20ms, which is 3 times faster than centralized control and supports large-scale node expansion.
[0099] Example 5
[0100] This embodiment provides a thermal stress equalization design method and module layout for a modular power supply system, used in industrial power supply systems operating in extreme environments (-40℃ to 125℃). Specific implementation details include:
[0101] Objective: To address the thermal stress problem across a wide temperature range in both cold and high-temperature industrial settings, and to improve the reliability of equipment in outdoor and metallurgical environments.
[0102] Implementation steps:
[0103] Ultra-wide temperature heat pipe-conductor composite structure:
[0104] The heat pipe working fluid is a methanol-water mixture (volume ratio 7:3, freezing point -45℃), and the inner wall of the copper-nickel alloy (CuNi25) shell is electroplated with a 1μm diamond coating for corrosion resistance.
[0105] The conductive core is made of copper-nickel alloy (conductivity 4.5 × 10⁻⁶). 7 The heat pipe is integrated using a cold pressing process, with a polyimide insulation layer (temperature resistance -55℃~200℃) and a surface area of S / m.
[0106] Dual-mode thermal management control:
[0107] Low temperature mode (< -20℃): 10W resistance heating film preheats the heat pipe, and the temperature rises to -10℃ within 30 seconds, starting the working fluid circulation;
[0108] High temperature mode (>85℃): Sn58Bi alloy phase change unit (melting point 140℃, latent heat 52kJ / kg) is triggered, and axial fan (2000rpm) enhances convection cooling.
[0109] Intelligent anti-freeze protection:
[0110] A temperature sensor (accuracy ±0.1℃) monitors the heat pipe temperature. When the temperature drops below -40℃, the SMA wire (austenite transformation temperature -35℃) contracts, cutting off the main circuit and sending a fault signal within 200ms.
[0111] Implementation results:
[0112] Wide temperature range adaptability: -40℃ start-up time <5min, module temperature stabilizes at 110℃±5℃ during continuous operation at 125℃, the temperature range is 60℃ wider than the traditional solution;
[0113] Improved reliability: After salt spray test (96h), the contact thermal resistance increases by <10%, meeting the IP68 protection level, and the structure does not loosen due to vibration (10-2000Hz, 2g).
[0114] Fault response speed: The anti-freeze protection mechanism under extreme low temperatures is triggered within 200ms, which is 100 times faster than manual inspection and reduces the equipment freezing damage rate by 90%.
[0115] Comparative Example 1
[0116] This comparison model provides a traditional independent cooling power supply system.
[0117] Purpose of implementation: To compare and verify the shortcomings of traditional solutions in terms of thermal stress balancing and load response.
[0118] Implementation steps:
[0119] Independent heat dissipation design:
[0120] The busbar is a pure copper busbar (100mm² cross-sectional area). 2 Tin-plated, thermal resistance 0.5℃·m / W, aluminum heat sink (thickness 2mm, spacing 5mm) relies on forced air cooling (fixed speed 2500rpm);
[0121] It has no heat pipes or phase change units, and the load distribution ratio is fixed (1:1:1). The temperature sensor is only used for over-limit alarm (threshold 90℃).
[0122] Implementation results:
[0123] Thermal stress problem: Under high load, the temperature difference between modules reaches 20℃, the bus temperature rise rate is 10℃ / s, and the capacitor failure rate is 5 times higher than that of Example 1.
[0124] System defects: Voltage fluctuation >5% during dynamic load adjustment, efficiency only 85%, external heating device required below -20℃ (additional power consumption 5%);
[0125] Protection mechanism: Relies on fuse (fusing time 500ms), which is prone to cascading failures under overload and has no active thermal field control capability.
[0126] Compared with Examples 1-5 and Comparative Example 1, Examples 1-5 of the present invention, through core technologies such as thermoelectric synergistic coupling networks, dynamic load balancing, and phase change cooling, significantly outperform traditional independent heat dissipation solutions (Comparative Example 1) in terms of heat dissipation efficiency, response speed, reliability, and environmental adaptability. Specific comparisons are as follows:
[0127] Examples: All examples employ heat pipe-conductor composite busbars (such as loop heat pipes, microchannel heat pipes, pulsating heat pipes, etc.) to deeply couple electrical energy transfer with heat dissipation, achieving a thermal conductivity of 800-2000 W / mK. For instance, the loop heat pipe in Example 1 reduces the busbar thermal resistance by 40%, while the liquid-cooling-heat pipe hybrid heat dissipation in Example 2 increases the module density to 50 kW / m². 3 The temperature rise is controlled at 65℃, which is 25℃ lower than the traditional air-cooled solution (temperature rise of 90℃).
[0128] Comparative example: Relying on independent aluminum heat sinks and forced air cooling, it has high thermal resistance (0.5℃·m / W), low heat dissipation efficiency, and the temperature difference between modules can reach 20℃ under high load. Local hot spots are prone to failure.
[0129] Load regulation and response speed:
[0130] Examples: Integrating edge computing units with intelligent algorithms (LSTM, reinforcement learning, BP neural networks, etc.) enables dynamic load distribution driven by thermal stress feedback, with a response time of <100ms. For example, Example 1 completes load transfer within 50ms during 3C fast charging, reducing the module temperature difference from 18℃ to 2.5℃; Example 4 uses blockchain collaboration to control the distributed node load balancing error to within 2%.
[0131] Comparative example: Fixed load distribution (ratio 1:1:1), no dynamic adjustment capability, voltage fluctuation >5% when load changes suddenly, response time reaches the second level, unable to adapt to high frequency operating conditions.
[0132] Overload protection and reliability:
[0133] Example: Integrating a low-melting-point alloy phase change unit (such as SnBi58, In52Sn48) into the bus junction node, which has both latent heat absorption and mechanical limiting functions, the overload protection time is <50ms (such as the 30ms trigger limit in Example 1), and avoiding cascading failures through intelligent fault-tolerant design (such as SMA drive, liquid metal trigger).
[0134] Comparative example: Relying on passive fuses (fusing time 500ms), without active thermal stress control, it is prone to single-point failure under overload, and the failure rate is more than 5 times higher than that of the example.
[0135] The above embodiments can be implemented in whole or in part by software, hardware, firmware or other arbitrary combinations. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product.
[0136] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.
[0137] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.
[0138] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of protection of the claims.
[0139] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A thermal stress equalization design method for a modular power supply system, characterized in that, Includes the following steps: Constructing a thermoelectric synergistic coupling network: The high-current bus between power modules is designed as a heat pipe-conductor composite structure, with an inner conductive core and an outer heat pipe structure. The Joule heat generated by the current drives the circulation of the heat pipe working fluid, simultaneously realizing power transmission and heat dissipation. Dynamic load balancing control: Dynamic load balancing control: By collecting module thermal stress data through distributed temperature sensors and adjusting module power distribution based on thermal stress feedback, when the temperature of a module exceeds the threshold, a predictive control algorithm based on the edge computing unit will transfer part of the load to the low-temperature module. Combined with the thermal coupling effect of the bus heat pipe, a dual-balance closed loop of electrical load and thermal load is formed. Phase change cooling node integration: Low melting point metal alloy phase change unit is integrated into the bus junction node. When a sudden large current causes local overheating, the alloy melts and absorbs latent heat to avoid thermal stress concentration. At the same time, mechanical limiting is triggered by the phase change volume expansion.
2. The thermal stress equalization design method for a power modular system as described in claim 1, characterized in that, The heat pipe structure includes microchannel heat pipes, pulsating heat pipes, or loop heat pipes, with the working fluid being water, methanol, or liquid metal, and a filling rate of 40-80%.
3. The thermal stress equalization design method for a power supply modular system as described in claim 1, characterized in that, The conductive core material is copper, aluminum, silver, or their alloys, with a conductivity ≥4× S / m, cross-sectional area ≥50 .
4. The thermal stress equalization design method for a power supply modular system as described in claim 1, characterized in that, The predictive control algorithm includes a machine learning model to predict module thermal trends in advance and to achieve dynamic load distribution through a power balancing circuit.
5. The thermal stress equalization design method for a power supply modular system as described in claim 1, characterized in that, The low-melting-point metal alloy has a melting point ≤200℃, a latent heat absorption ≥25kJ / kg, and a volume expansion rate ≤20% during phase transformation.
6. A modular layout structure for a power supply modular system, characterized in that, include: Heat pipe busbar: It consists of an inner conductive core and an outer heat pipe structure. The heat pipe working fluid circulation channel is thermally coupled with the conductive core, and the thermal conductivity is ≥800W / mK. Phase change cooling busbar: Integrated at the busbar junction, it includes a low melting point metal alloy phase change unit and a mechanical limiting device. The phase change unit is connected to the busbar conductor through an embedded or encased connection. Intelligent load balancing module: Each power module has a built-in edge computing unit, which realizes thermal stress data sharing and dynamic load adjustment through CAN bus or wireless communication bus, and the temperature difference fluctuation between modules is <5℃.
7. The modular layout structure of a power supply modular system as described in claim 6, characterized in that, The heat pipe surface of the heat pipe busbar is coated with a radiation-enhancing coating with an emissivity ≥0.7, or an integrated liquid-cooling-heat pipe hybrid heat dissipation channel is provided.
8. The modular layout structure of a power supply modular system as described in claim 6, characterized in that, The mechanical limiting device of the phase change cooling junction node is driven by a shape memory alloy. When the phase change volume expansion triggers the limit, the overload circuit is cut off.
9. The modular layout structure of a power supply modular system as described in claim 6, characterized in that, The edge computing unit of the intelligent load balancing module integrates a neural network model and optimizes the load distribution strategy through reinforcement learning algorithms, with a response time of <100ms.
10. The modular layout structure of a power supply modular system as described in claim 6, characterized in that, The system also includes a thermal field control module, which uses a visual sensing device to scan the temperature distribution in real time and drives the actuator to adjust the spacing or angle of the modules to achieve dynamic thermal field equilibrium.