A multi-mode power management method and device for a gallium nitride adapter

Abstract

The present invention relates to the technical field of thermal management of power adapters, and discloses a multi-mode power management method and device for a gallium nitride adapter. The multi-mode power management method for a gallium nitride adapter comprises: dividing the interior of the adapter into multiple cooling priority areas according to component thermal sensitivity and integrating a three-dimensional layered microchannel structure; real-time monitoring of the temperature and cooling system status of each area; constructing an adaptive flow distribution algorithm to dynamically adjust the coolant flow distribution according to the real-time temperature status of each area; calculating the maximum allowable power under the current cooling capacity and dynamically adjusting the power parameters; and dynamically switching the operating mode of the adapter. The present invention achieves the technical effects of improved temperature management, extended reliability and service life, and enhanced power output capacity through the synergistic effect of the regionally differentiated microchannel cooling structure and multi-mode power management.

Description

Technical Field

[0001] The present invention relates to the technical field of thermal management of power adapters, and more particularly to a multi-mode power management method and device for a gallium nitride adapter. Background Art

[0002] As electronic devices evolve towards higher performance and smaller size, power adapters are also facing demands for miniaturization and high power density. Gallium nitride (GaN), a new generation of wide-bandgap semiconductor materials, boasts excellent properties such as high breakdown electric field, high electron mobility, and low on-resistance, enabling GaN-based power adapters to achieve higher power density.

[0003] However, the concentrated heat load caused by high power density poses a serious challenge to the adapter's reliability, efficiency, and lifespan. In miniaturized GaN adapters, the dense arrangement of power components leads to uneven distribution of hotspots, and traditional uniform cooling strategies are unable to specifically address the differentiated heat dissipation needs of each area. Furthermore, existing cooling systems employ a uniform cooling strategy that fails to provide customized cooling based on the varying thermal sensitivity and heat generation rates within different areas of the adapter, resulting in irrational cooling resource allocation and reduced overall performance.

[0004] At the same time, existing thermal management systems and power management systems are often independent of each other, lacking a coordinated control mechanism and unable to intelligently adjust operating modes based on real-time thermal conditions. This is particularly true in ultra-small adapters with limited heat dissipation channels. Traditional passive cooling or simple temperature control protection methods struggle to meet the demands of sustained, stable high-power output. The high power density of GaN devices leads to concentrated hotspots, delayed temperature control response, and inability to prevent overheating in a timely manner. Traditional temperature control methods simply reduce power or shut down when critical temperatures are reached, failing to achieve a smooth transition and impacting the user experience.

[0005] Therefore, there is an urgent need for a multi-mode power management method for a GaN adapter that can solve the above problems, so as to improve the thermal management efficiency and power output stability of the adapter. Summary of the Invention

[0006] The present invention provides a multi-mode power management method and device for a gallium nitride adapter, which solves the technical problems of low thermal management efficiency and unstable power control of miniaturized gallium nitride adapters in the related art.

[0007] The present invention provides a multi-mode power management method and device for a gallium nitride adapter, comprising:

[0008] The interior of the adapter is divided into multiple cooling priority areas according to the thermal sensitivity of the components and a three-dimensional layered microchannel structure is integrated;

[0009] Based on the division of cooling priority areas and the three-dimensional layered microchannel structure, a sensor network system is constructed to monitor the temperature of each area and the status of the cooling system in real time;

[0010] Based on the monitoring data of each zone's temperature and cooling system status, an adaptive flow distribution algorithm is constructed to dynamically adjust the coolant flow distribution according to the real-time temperature status of multiple cooling priority zones;

[0011] Based on the monitoring data of each area's temperature and cooling system status and the coolant flow distribution results, the maximum allowable power under the current cooling capacity is calculated and the power parameters are dynamically adjusted;

[0012] The working mode of the adapter is dynamically switched based on the result of calculating the maximum allowable power under the current cooling capacity, the dynamically adjusted power parameters, and the load status and temperature conditions.

[0013] Furthermore, the division into multiple cooling priority areas includes:

[0014] Divide the interior of the adapter into high-priority cooling zones, medium-priority cooling zones, and low-priority cooling zones;

[0015] The high-priority cooling zone includes GaN power devices and driver ICs, the medium-priority cooling zone includes transformers and high-power resistors, and the low-priority cooling zone includes filter capacitors and interface circuits.

[0016] Furthermore, the integrated three-dimensional hierarchical microchannel structure includes:

[0017] Microchannels of different densities and directions are configured for different priority areas. High-priority areas use high-density serpentine microchannels, medium-priority areas use medium-density linear microchannels, and low-priority areas use low-density radial microchannels.

[0018] A dielectric liquid having high thermal conductivity, low electrical conductivity and high dielectric strength properties is selected as the cooling medium.

[0019] Furthermore, the flow distribution of the adaptive flow distribution algorithm is to distribute the total flow of the coolant according to the proportion of the cooling demand index of each area to the total demand;

[0020] The cooling demand index of the multiple cooling priority areas is calculated based on the difference between the current temperature of the multiple cooling priority areas and the reference temperature and the weight coefficient of the area. The weight coefficient of the high priority area is 1.5 to 2.0, the weight coefficient of the medium priority area is 0.8 to 1.2, and the weight coefficient of the low priority area is 0.3 to 0.6.

[0021] Furthermore, the maximum allowable power under the current cooling capacity is calculated to determine the power upper limit according to the real-time measured thermal resistance value and the target maximum temperature; the maximum allowable power is equal to the difference between the target maximum temperature and the ambient temperature divided by the thermal resistance value.

[0022] Furthermore, the dynamic adjustment of power parameters includes:

[0023] When the temperature rise trend is obvious, reduce the switching frequency and drive voltage;

[0024] When the temperature is predicted to stabilize or decrease, increase the switching frequency and drive voltage;

[0025] The switching frequency can be adjusted from 65kHz to 1MHz, and the drive voltage can be adjusted from 3.3V to 6.5V.

[0026] Furthermore, the working mode of the dynamic switching adapter has four working modes, including:

[0027] Normal mode: The temperature is within the safe range, i.e. , with full power output as the main goal;

[0028] Low temperature optimization mode: When the temperature is low, , optimize energy efficiency by increasing switching frequency;

[0029] High temperature derating mode: When the temperature approaches the threshold, , reduce power output and prioritize heat dissipation;

[0030] Critical protection mode: When the temperature approaches the limit, , quickly reduce power and increase cooling efforts to ensure system safety.

[0031] Furthermore, the operating mode of the adapter is dynamically switched to avoid power fluctuations caused by switching between modes. During the switching process, power and cooling parameters gradually change according to a set time constant. The specific steps are as follows:

[0032] Define a state transition function so that the power and cooling parameters during the switching process change gradually according to the set time constant;

[0033] The power parameters use shorter time constants, and the cooling parameters use longer time constants;

[0034] During the mode switching process, the parameter status is updated every 5ms to ensure a smooth transition.

[0035] Furthermore, the dynamic adjustment of power parameters also includes implementing a power management strategy based on heat redistribution, the specific steps of which are:

[0036] Through the topological structure of the microchannel network, a heat transfer path is formed from the heat-sensitive area to the heat-resistant area;

[0037] establishing a direct connection between the microchannel outlet of the heat-sensitive region and the microchannel inlet of the heat-resistant region;

[0038] By controlling the flow distribution ratio of each microchannel segment, directional heat transfer is achieved, which reduces the temperature of GaN power devices while increasing the temperature of transformers and capacitors to achieve a more uniform temperature distribution.

[0039] The present invention provides a multi-mode power management device for a gallium nitride adapter, comprising a memory and one or more processors. The memory stores executable code, and when the one or more processors execute the executable code, they are used to perform the multi-mode power management method for a gallium nitride adapter.

[0040] The beneficial effects of the present invention are: through the synergistic effect of regionally differentiated microchannel cooling structures and multi-mode power management, the technical difficulties of thermal management and power control in miniaturized gallium nitride adapters are solved, achieving the technical effects of improved temperature management, extended reliability and service life, and enhanced power output capability;

[0041] The maximum temperature of the adapter is reduced, the temperature difference of hot spots is reduced, and efficient operation is achieved in the full temperature range; the thermal protection trigger rate is reduced, and the service life is extended; with the volume unchanged, the continuous power density and peak power are improved due to the improved heat dissipation efficiency; the working mode is adaptively adjusted according to the load and temperature without manual intervention; through precise temperature management and dual-mode heat dissipation system, the energy consumption of the cooling system is reduced, and the efficiency of the entire load range is improved; smooth transition between various working modes, avoid the sudden frequency reduction or shutdown of traditional adapters, and provide stable and continuous power supply. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a flow chart of a multi-mode power management method of a gallium nitride adapter in the present invention;

[0043] Figure 2 This is a flow chart of dividing multiple cooling priority areas and integrating a three-dimensional layered microchannel structure in the present invention;

[0044] Figure 3 This is a flow chart of the sensor network system constructed in the present invention for real-time monitoring of the temperature of each area and the status of the cooling system;

[0045] Figure 4 This is a flow chart of dynamically adjusting coolant flow distribution in the present invention;

[0046] Figure 5This is a flow chart of calculating the maximum allowable power under the current cooling capacity and dynamically adjusting the power parameters in the present invention;

[0047] Figure 6 This is a flow chart of the working mode of the dynamic switching adapter in the present invention. DETAILED DESCRIPTION

[0048] The subject matter described herein will now be discussed with reference to example embodiments. It should be understood that these embodiments are discussed solely to enable those skilled in the art to better understand and implement the subject matter described herein, and that the functions and arrangements of the elements discussed may be varied without departing from the scope of this specification. Various examples may omit, substitute, or add various processes or components as needed. Furthermore, features described in some examples may be combined in other examples.

[0049] At least one embodiment of the present invention discloses a multi-mode power management method for a gallium nitride adapter, such as Figures 1 to 6 Shown, including:

[0050] Step 1: Divide the interior of the adapter into multiple cooling priority areas based on component thermal sensitivity and integrate a three-dimensional layered microchannel structure;

[0051] Specifically include:

[0052] Step 1.1: Classify the regions according to the thermal sensitivity and heating characteristics of the components;

[0053] The interior of the adapter is divided into a high-priority cooling zone (including GaN power devices, driver ICs and other key components), a medium-priority cooling zone (including transformers, high-power resistors, etc.) and a low-priority cooling zone (including filter capacitors, interface circuits and other high-temperature resistant components);

[0054] Optionally, in some embodiments, the high-priority cooling zone can be further subdivided into an ultra-high-priority area (such as GaN power tube chips) and a high-priority area (such as driver ICs), each of which adopts different microchannel densities and cooling strategies;

[0055] Step 1.2, integrating a three-dimensional hierarchical microchannel structure inside the PCB;

[0056] Microchannels of different densities and directions are configured for different priority areas. High-priority areas use high-density serpentine microchannels, medium-priority areas use medium-density linear microchannels, and low-priority areas use low-density radial microchannels.

[0057] In some embodiments, the microchannels can be multi-layered, for example, by constructing independent microchannel networks on different layers of a PCB to achieve three-dimensional cooling by vertically connecting channels; or by using variable-section microchannels with a larger cross-section at the inlet and a gradually decreasing cross-section at the outlet to enhance the flow rate of the coolant;

[0058] Step 1.3, select dielectric liquid as cooling medium;

[0059] The dielectric liquid has high thermal conductivity, low electrical conductivity and high dielectric strength, ensuring safety for use around electronic components;

[0060] Optionally, the cooling medium can be a fluorinated liquid (such as FC-72, HFE-7100, etc.) or mineral insulating oil, depending on the specific temperature range and insulation requirements. In higher power density applications, nanoparticles (such as aluminum oxide, boron nitride, etc.) can also be added to enhance thermal conductivity.

[0061] Step 1.4, integrating a micro temperature sensor array;

[0062] At least five micro temperature sensors are placed at key locations inside the adapter to monitor the temperature status of each area in real time.

[0063] In specific application scenarios, such as medical device chargers or server power supplies that require more accurate temperature monitoring, the number of temperature sensors can be increased to 8-12; or a combination of different types of sensors can be used, such as a mixed application of thermocouples, thermistors, and bandgap temperature sensors, to improve measurement accuracy and reliability.

[0064] Step 2: Based on the division of cooling priority areas and the three-dimensional layered microchannel structure, a sensor network system is constructed to monitor the temperature of each area and the status of the cooling system in real time;

[0065] Based on the regional differentiated micro-channel cooling structure, a complete sensor network system is built to monitor the internal temperature distribution and cooling system status of the adapter in real time, providing data support for subsequent flow distribution and temperature prediction. Specifically, it includes:

[0066] Step 2.1, integrating micro pressure and flow sensors;

[0067] Miniature pressure and flow sensors are installed at the coolant inlet and outlet respectively to monitor the pressure difference and flow changes of the cooling system;

[0068] Step 2.2, calculate the cooling efficiency based on the monitoring data;

[0069] Cooling efficiency The calculation formula is:

[0070] ;

[0071] in Indicates the cooling efficiency, Removal of heat, Indicates the pump power, represents the specific heat capacity of the coolant, represents the coolant mass flow, Indicates the temperature difference between the coolant inlet and outlet, represents the pump driving voltage, Indicates the pump drive current.

[0072] Step 2.3, constructing a temperature monitoring network data processing unit;

[0073] Filter, standardize and detect anomalies on collected temperature, pressure and flow data to ensure data accuracy;

[0074] Step 2.4, establishing a thermal characteristics database;

[0075] Store temperature distribution characteristics under different working conditions to provide data support for subsequent power management strategies.

[0076] Step 3: Based on the monitoring data of the temperature and cooling system status of each zone, an adaptive flow distribution algorithm is constructed to dynamically adjust the coolant flow distribution according to the real-time temperature status of the multiple cooling priority zones;

[0077] Based on the real-time temperature data collected by the multi-layer distributed temperature monitoring network, an adaptive flow distribution algorithm and temperature prediction model are constructed to dynamically optimize the cooling resource allocation in each area. Specifically, the following features are included:

[0078] Step 3.1, configure the piezoelectric micropump and microvalve array system;

[0079] The micro pump is responsible for providing cooling liquid circulation power, and the micro valve array controls the flow distribution of micro channels in each area;

[0080] In high-end applications, such as data center server power supplies, a backup micropump system can be optionally configured to automatically switch to the main pump in the event of a failure, ensuring cooling system reliability. For size-constrained applications such as wearable device chargers, a valveless microfluidic system driven by surface acoustic waves can be used to replace traditional micropumps and valves, further reducing size.

[0081] Step 3.2, construct an adaptive traffic distribution algorithm;

[0082] The coolant distribution is dynamically adjusted according to the real-time temperature status of each area. The flow distribution calculation formula is:

[0083] ;

[0084] in Indicates the assignment to Coolant flow rate in each area, Indicates the total flow, and Respectively represent and The weight coefficient of the cooling priority of each zone, Indicates the The temperature of the area, represents the reference temperature, Indicates the The temperature of the area, Indicates that all The sum of the regions is calculated. Indicates the total number of regions;

[0085] The implementation process of this adaptive flow allocation algorithm involves: first, normalizing the temperature data of each cooling zone and calculating the difference between each zone's temperature and the reference temperature; then, calculating the cooling demand index for each zone based on preset weight coefficients (1.5-2.0 for high-priority zones, 0.8-1.2 for medium-priority zones, and 0.3-0.6 for low-priority zones); and finally, allocating flow based on the proportion of each zone's cooling demand index to the total demand. In practice, this algorithm effectively avoids the problem of excessively high temperatures in hot spots by updating the flow allocation every 100ms.

[0086] Optionally, different traffic allocation strategy variations can be adopted for different usage scenarios. For example, for gaming laptop chargers that frequently experience sudden loads, a prediction factor can be added to adjust traffic allocation in advance based on historical load patterns. For safety-critical scenarios such as industrial control system power supplies, a conservative allocation strategy can be adopted to keep the cooling flow in key areas at or above the minimum safety threshold.

[0087] Step 3.3, construct a thermal diffusion prediction model based on the finite difference method;

[0088] To predict the temperature change inside the adapter in the next 10-300ms, the discrete form of the heat diffusion equation is:

[0089] ;

[0090] in Indicates location At time step The temperature at the next moment; Indicates location At time step The temperature at the current moment; represents the thermal diffusivity, which describes the rate at which heat diffuses in a material; Represents the time step, that is, the interval between two adjacent time points in the calculation; Indicates the spatial grid size in the x-direction, that is, the spacing of the calculation grid in the x-direction; Indicates the spatial grid size in the y direction, that is, the spacing of the calculation grid in the y direction; Indicates location At time step The temperature of the point adjacent to the current point in the x direction; Indicates location At time step The temperature of the point adjacent to the current point in the negative x direction; Indicates location At time step The temperature of the point adjacent to the current point in the y direction; Indicates location At time step The temperature of the point adjacent to the current point in the negative y direction; Indicates location The heat source term, that is, the heat generated per unit volume; Indicates the density of a material, describing the ratio of its mass to its volume; It represents the specific heat capacity of the material, which describes the amount of heat required to raise the unit temperature of a unit mass of material;

[0091] The specific implementation of this thermal diffusion prediction model is as follows: the internal space of the adapter is divided into a 10×10×3 three-dimensional grid, and a thermal network consisting of temperature nodes, heat source nodes, and boundary nodes is established. The discretized thermal diffusion equation is applied to each node, and the temperature field is solved using an explicit iterative method. The time step is set to 1ms, and the spatial grid size is 1mm×1mm×0.5mm. The thermal diffusion coefficient of each area is calibrated by recording the historical power data and corresponding temperature response of each power component. In practical applications, this model can accurately predict the temperature change trend caused by sudden load changes, predicting temperature changes 10-300ms in advance, providing sufficient response time for power parameter adjustments.

[0092] In some embodiments, this thermal diffusion prediction model can be coupled with a simplified fluid dynamics model to calculate the fluid flow and heat transfer characteristics in the microchannel. Alternatively, for low-cost applications with limited computing resources, a simplified lumped parameter thermal model can be used to divide the interior of the adapter into a small number of thermal zones, represented by a thermal resistance-capacitance network to reduce computational complexity.

[0093] Step 3.4, input the current power data into the prediction model as a heat source term;

[0094] Combined with past temperature data, temperature trend prediction is performed. In this implementation, the power data of major heat-generating components, such as GaN power devices, driver ICs, and transformers, is converted into heat generation rates per unit volume. This is then input into the prediction model as a heat source term to predict future temperatures.

[0095] For example, in a high-performance laptop charging scenario, when a user launches a graphics-intensive application, the laptop's power demand can quickly increase from 45W to 95W. The prediction model in this embodiment can predict within approximately 50ms of the initial power demand increase that the temperature in the high-priority zone will rise by approximately 15°C within the next 250ms. This allows proactive power management measures to be initiated in advance, ensuring that a safe temperature range is maintained despite the sudden load increase.

[0096] Step 4: Calculate the maximum allowable power under the current cooling capacity based on the monitoring data of the temperature and cooling system status of each area and the coolant flow distribution results, and dynamically adjust the power parameters;

[0097] The intelligent flow distribution and temperature prediction system, implemented by the intelligent flow distribution and temperature prediction unit, provides temperature monitoring data and prediction results to achieve coordinated control of cooling capacity and power parameters, ensuring that the system maximizes power output while ensuring heat dissipation. Specifically, the following are included:

[0098] Step 4.1, calculate the maximum allowable power under the current cooling capacity;

[0099] The power upper limit is determined based on the real-time measured thermal resistance value and the target maximum temperature. The calculation formula is:

[0100] ;

[0101] in Indicates the maximum allowable power, Indicates the maximum temperature allowed. Indicates the ambient temperature, Indicates the thermal resistance currently measured;

[0102] Optionally, in applications with large ambient temperature fluctuations (such as charging portable devices outdoors), an ambient temperature sensor can be integrated into the adapter to monitor ambient temperature changes in real time and dynamically adjust the maximum allowable power. In extreme temperature environments (such as high altitude or polar environments), multi-level safety redundancy can be used to increase the safety factor of the thermal resistance value (for example, 1.2-1.5 times), ensuring a more conservative power limit.

[0103] Step 4.2, dynamically adjust the power parameters according to the temperature prediction results;

[0104] The frequency adjustment range is 65kHz-1MHz, and the drive voltage adjustment range is 3.3V-6.5V. The adjustment strategy is: when the predicted temperature rises significantly, the switching frequency and drive voltage are reduced; when the predicted temperature tends to stabilize or decrease, the switching frequency and drive voltage are increased to improve performance.

[0105] The specific implementation method is as follows: Dynamic adjustment of power parameters is achieved through a digitally controlled frequency synthesizer and a programmable drive voltage regulator. A mapping relationship between the temperature change rate and the power parameter adjustment range is established. When the temperature prediction results show that the temperature rise rate exceeds 1°C / s within the next 100ms, the switching frequency is reduced by 5% and the drive voltage is reduced by 0.2V. When the temperature change rate is less than 0.2°C / s and the temperature is below the threshold, the switching frequency is increased by 3% and the drive voltage is increased by 0.1V. Parameter adjustments are performed every 50ms to ensure that the system can respond to temperature changes in a timely manner.

[0106] For example, in a fast-charging scenario, when a high-power phone (such as a flagship phone supporting 65W fast charging) is connected to an adapter, this embodiment can intelligently adjust the switching frequency based on temperature prediction: maintaining a higher frequency (approximately 800kHz) in the early stages of charging to improve efficiency; when a rapid temperature rise is predicted, the frequency is reduced to approximately 500kHz in advance, and the drive voltage is gradually reduced to keep the temperature rise rate within a controllable range. This avoids the charging speed fluctuation problem caused by traditional methods that reduce power only after the temperature exceeds a certain level.

[0107] Step 4.3, implement a power management strategy based on heat redistribution;

[0108] By actively controlling the flow of coolant, heat is transferred from heat-sensitive areas (such as GaN power devices) to heat-resistant areas (such as transformers and capacitors), expanding the overall power capacity of the system.

[0109] This heat redistribution strategy is implemented by creating a heat transfer pathway from heat-sensitive to heat-resistant areas through the topology of the microchannel network; establishing a direct connection between the microchannel outlets in the heat-sensitive areas and the microchannel inlets in the heat-resistant areas; and achieving directional heat transfer by controlling the flow distribution ratio of each microchannel segment. In practical applications, this strategy has reduced the maximum temperature of GaN power devices by 8°C, while increasing the temperatures of transformers and capacitors by 5-7°C, achieving a more uniform temperature distribution overall.

[0110] In some embodiments, the heat redistribution strategy can be used in conjunction with dynamic adjustment of power parameters. For example, when it is detected that the temperature of the GaN power device is approaching a critical value but the overall thermal capacity of the system still has margin, the power load sharing of the heat-resistant area can be increased while reducing the power load of the GaN device, thereby balancing the temperatures of various parts without reducing the overall power output.

[0111] Step 5: Dynamically switch the working mode of the adapter based on the result of calculating the maximum allowable power under the current cooling capacity, the dynamically adjusted power parameters, and the load state and temperature conditions;

[0112] Based on the power thermal management results performed by the multi-mode collaborative power thermal management unit, combined with load status and temperature conditions, the adapter's operating mode is dynamically switched to achieve a balance between energy efficiency and heat dissipation, optimizing system performance under different operating conditions. Specifically, the following features are included:

[0113] Step 5.1, construct a multi-level temperature response strategy;

[0114] Divided into four working modes:

[0115] Normal mode: The temperature is within the safe range, i.e. , with full power output as the main goal;

[0116] Low temperature optimization mode: When the temperature is low, , optimize energy efficiency by increasing switching frequency;

[0117] High temperature derating mode: When the temperature approaches the threshold, , reduce power output and prioritize heat dissipation;

[0118] Critical protection mode: When the temperature approaches the limit, , quickly reduce power and increase cooling efforts to ensure system safety;

[0119] temperature This refers to the real-time temperature monitoring of key heat-generating components within the adapter (such as GaN power devices and driver ICs), representing the temperature nodes of high-priority cooling areas. This temperature, collected in real time by a distributed temperature sensor network, reflects the actual operating temperature of the system's most heat-sensitive areas and serves as a criterion for switching operating modes.

[0120] Optionally, temperature thresholds can be adjusted for different application scenarios. For example, for high-reliability applications such as avionics chargers, the temperature thresholds for each mode can be lowered by 5-10°C to provide a greater safety margin. For scenarios such as server power supplies that require maximum power output, the thresholds can be raised by 3-5°C to extend the operating range while ensuring safety.

[0121] Step 5.2: Implement a dual-mode cooling system according to the load status;

[0122] In light load mode (load < 30% of rated power), heat dissipation is achieved through conduction and natural convection, and the active cooling system is disabled to save energy.

[0123] Active microfluidic cooling is activated in heavy load mode (load ≥ 30% rated power) to ensure temperature control at high power output;

[0124] In some embodiments, an intermediate transition mode (load range of 20-40% rated power) can be added, using a pulsed active cooling strategy, that is, intermittently starting the microfluidic circulation system to balance cooling effect and energy consumption. For applications requiring low noise (such as bedroom chargers), a night mode can be implemented, prioritizing passive cooling and running the active cooling system at a lower speed when necessary to reduce noise.

[0125] Step 5.3, implement the mode smooth switching algorithm;

[0126] To avoid power fluctuations caused by switching between modes, the power and cooling parameters change gradually according to the set time constant during the switching process to ensure output stability.

[0127] The specific implementation of the mode smooth switching algorithm is as follows: define the state transition function:

[0128] ;

[0129] in Indicates time The state parameter value at time is the initial state parameter value, is the target state parameter value, is the time constant (usually set to 50-200ms), Indicates the time from the beginning to the current time. represents the exponential factor that decays over time, represents the base of natural logarithms;

[0130] Appropriate time constants are selected for different parameter types, with shorter time constants used for power parameters (such as switching frequency and drive voltage) and longer time constants for cooling parameters (such as flow rate and pump speed). During mode switching, parameter status is updated every 5ms to ensure a smooth transition. In practical applications, this algorithm effectively eliminates output voltage fluctuations during mode switching, reducing the fluctuation range from the traditional 5-8% to 0.5-1%.

[0131] For example, in a portable medical device charging scenario, when the device suddenly enters a high-power diagnostic mode from a standby state, the mode smooth switching algorithm of this embodiment can complete a smooth transition from a low-temperature optimization mode to a normal mode within 200ms, while turning on the active cooling system. During the entire process, the output voltage fluctuation is controlled within ±0.8%, ensuring stable power supply to the medical device and avoiding diagnostic data distortion or device restart problems caused by power supply fluctuations.

[0132] A multi-mode power management device for a gallium nitride adapter includes a memory and one or more processors. The memory stores executable code. When the one or more processors execute the executable code, they are used to perform the multi-mode power management method for a gallium nitride adapter.

[0133] Here, the present invention provides an implementation example:

[0134] This implementation has been applied and verified in the development of a 65W high-power-density GaN adapter. This adapter measures just 40×35×30mm and employs a single-stage LLC topology, offering an output voltage range of 5-20V and a maximum output current of 5A. The following describes the specific application of this implementation in this product.

[0135] This GaN adapter is primarily used for fast charging in high-end laptops, tablets, and smartphones. Extreme operating environments, such as prolonged periods of high load and high ambient temperatures, pose significant challenges to the adapter's thermal management and power control capabilities. For example, a high-performance gaming laptop, when operating at full load (simultaneously running high-load games and rendering software), requires a continuous power supply of 60-65W. This operating mode can persist for hours, easily causing traditional adapters to overheat, reduce power output, or even shut down.

[0136] In the adapter, the specific implementation process of this embodiment is as follows:

[0137] The microchannel cooling structure is implemented within the adapter's four-layer PCB, integrating a total length of approximately 210mm. The GaN power device and driver IC area (high-priority area) utilizes 0.8mm-wide serpentine microchannels with a density of 3.5mm² / mm²; the LLC transformer and rectifier diode area (medium-priority area) utilizes 1.2mm-wide linear microchannels with a density of 2.2mm² / mm²; and the input and output filter capacitor area (low-priority area) utilizes 1.5mm-wide radial microchannels with a density of 1.0mm² / mm². The cooling medium used is modified fluorinated liquid HFE-7000, with a thermal conductivity of 0.075W / (m·K) and a dielectric strength of 35kV / mm. A micro piezoelectric pump (8×8×3mm) provides a maximum flow rate of 15ml / min and a maximum pressure of 25kPa.

[0138] The temperature monitoring system utilizes seven micro-thermistor temperature sensors within the adapter: three in the GaN power device area, two in the transformer area, and one each in the control IC and output interface area. The sampling frequency is 200Hz, and the temperature measurement accuracy is ±0.5°C. Micro-pressure and flow sensors are installed at the coolant inlet and outlet, respectively, to monitor system differential pressure and flow. Data is acquired using a 12-bit ADC and processed by a built-in 32-bit microcontroller (STM32F042, 48MHz).

[0139] Adaptive traffic allocation is based on real-time temperature data, with the system adjusting traffic allocation every 100ms. For example, when a laptop suddenly enters a high-load operating state from standby mode, the temperature rise rate in the GaN power device area reaches 2.5°C / s. The traffic allocation algorithm immediately adjusts the allocation ratio, increasing the traffic in the high-priority area from the initial 40% to 65%, maintaining 35% in the medium-priority area, and reducing the low-priority area to only 10% of the total traffic. Through this dynamic adjustment, the temperature rise in the GaN area is limited to 8°C, instead of the 15°C increase that would have been achieved with a fixed allocation scheme.

[0140] Multi-mode collaborative power thermal management is implemented in this adapter, enabling dynamic power parameter adjustment with a switching frequency range of 75kHz-850kHz and a drive voltage range of 3.8V-6.2V. Taking a real-world use case as an example, when the adapter is charging a gaming laptop while the laptop is simultaneously running a game and video exporting tasks, the adapter detects a high load demand and a rapid temperature increase in the GaN power device area. The system first calculates the maximum allowable power under the current cooling capacity to be 68W (at the time, the GaN area temperature was 65°C and the thermal resistance was 0.32°C / W). It then reduces the switching frequency from the initial 800kHz to 600kHz and the drive voltage from 6.0V to 5.2V. At the same time, active cooling at maximum flow is initiated to stabilize the temperature of the GaN area at around 70°C, maintaining a continuous output power of 60W.

[0141] Adaptive operating mode switching enables the system to automatically and smoothly switch between four operating modes in different usage scenarios. For example, when a user connects a mobile phone for fast charging (20W load), the system initially operates in low-temperature optimization mode (approximately 45°C), maintaining a high switching frequency (approximately 820kHz) to improve energy efficiency. When the user subsequently connects a laptop and begins downloading a large file, the load quickly rises to 50W. The system predicts that the temperature will exceed 65°C in a short period of time, so it smoothly switches to normal mode in advance and gradually activates the active cooling system. The entire switching process takes approximately 180ms, and the output voltage fluctuation is controlled within ±0.6%, ensuring stable operation of the device.

[0142] Two key technical effects of this implementation were fully verified: temperature management effect and power output capability.

[0143] When the ambient temperature is 25°C, the maximum temperature comparison under different load conditions is shown in Table 1:

[0144] Table 1: Comparison of maximum temperatures under different load conditions (ambient temperature 25°C)

[0145]

[0146] The comparison of continuous output power under different ambient temperatures is shown in Table 2:

[0147] Table 2: Comparison of continuous output power at different ambient temperatures

[0148]

[0149] From the above data, it can be seen that this embodiment has achieved improvements in temperature management. Under various load conditions, the maximum temperature is reduced by an average of about 21%. In terms of power output capacity, especially under high ambient temperature conditions, the continuous output power is increased by as much as 47.4%-62.5%, fully verifying the technical effect of this embodiment.

[0150] The above describes an embodiment of the present invention, but this embodiment is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Ordinary technicians in this field can also make more forms of equivalent embodiments based on the inspiration of this embodiment, all of which are protected by this embodiment.

Claims

1. A multi-mode power management method for a gallium nitride adapter, characterized in that: include: The interior of the adapter is divided into multiple cooling priority areas according to the thermal sensitivity of the components and a three-dimensional layered microchannel structure is integrated; Based on the division of cooling priority areas and the three-dimensional layered microchannel structure, a sensor network system is constructed to monitor the temperature of each area and the status of the cooling system in real time; Based on the monitoring data of each zone's temperature and cooling system status, an adaptive flow distribution algorithm is constructed to dynamically adjust the coolant flow distribution according to the real-time temperature status of multiple cooling priority zones; Based on the monitoring data of each area's temperature and cooling system status and the coolant flow distribution results, the maximum allowable power under the current cooling capacity is calculated and the power parameters are dynamically adjusted; The working mode of the adapter is dynamically switched based on the result of calculating the maximum allowable power under the current cooling capacity, the dynamically adjusted power parameters, and the load status and temperature conditions.

2. The multi-mode power management method of a gallium nitride adapter according to claim 1, characterized in that: The division into multiple cooling priority areas includes: Divide the interior of the adapter into high-priority cooling zones, medium-priority cooling zones, and low-priority cooling zones; The high-priority cooling zone includes GaN power devices and driver ICs, the medium-priority cooling zone includes transformers and high-power resistors, and the low-priority cooling zone includes filter capacitors and interface circuits.

3. The multi-mode power management method of a gallium nitride adapter according to claim 1, characterized in that: The integrated three-dimensional hierarchical microchannel structure comprises: Microchannels of different densities and directions are configured for different priority areas. High-priority areas use high-density serpentine microchannels, medium-priority areas use medium-density linear microchannels, and low-priority areas use low-density radial microchannels. A dielectric liquid having high thermal conductivity, low electrical conductivity and high dielectric strength properties is selected as the cooling medium.

4. The multi-mode power management method of a gallium nitride adapter according to claim 1, characterized in that: The flow distribution of the adaptive flow distribution algorithm is to distribute the total flow of coolant according to the proportion of the cooling demand index of each area to the total demand; The cooling demand index of the multiple cooling priority areas is calculated based on the difference between the current temperature of the multiple cooling priority areas and the reference temperature and the weight coefficient of the area. The weight coefficient of the high priority area is 1.5 to 2.0, the weight coefficient of the medium priority area is 0.8 to 1.2, and the weight coefficient of the low priority area is 0.3 to 0.

6.

5. The multi-mode power management method of a gallium nitride adapter according to claim 1, characterized in that: The maximum allowable power under the current cooling capacity is calculated by determining the power upper limit according to the thermal resistance value measured in real time and the target maximum temperature; the maximum allowable power is equal to the difference between the target maximum temperature and the ambient temperature divided by the thermal resistance value.

6. The multi-mode power management method of a gallium nitride adapter according to claim 1, characterized in that: The dynamic adjustment of power parameters includes: When the temperature rise trend is obvious, reduce the switching frequency and drive voltage; When the temperature is predicted to stabilize or decrease, increase the switching frequency and drive voltage; The switching frequency can be adjusted from 65kHz to 1MHz, and the drive voltage can be adjusted from 3.3V to 6.5V.

7. The multi-mode power management method of a gallium nitride adapter according to claim 1, characterized in that: There are four working modes in the working mode of the dynamic switching adapter, including: Normal mode: The temperature is within the safe range, i.e. , with full power output as the main goal; Low temperature optimization mode: When the temperature is low, , optimize energy efficiency by increasing switching frequency; High temperature derating mode: When the temperature approaches the threshold, , reduce power output and prioritize heat dissipation; Critical protection mode: When the temperature approaches the limit, , quickly reduce power and increase cooling efforts to ensure system safety.

8. The multi-mode power management method of a gallium nitride adapter according to claim 1, characterized in that: The working mode of the dynamic switching adapter avoids power fluctuations caused by switching between modes. During the switching process, the power and cooling parameters gradually change according to the set time constant. The specific steps are as follows: Define a state transition function so that the power and cooling parameters during the switching process change gradually according to the set time constant; The power parameters use shorter time constants, and the cooling parameters use longer time constants; During the mode switching process, the parameter status is updated every 5ms to ensure a smooth transition.

9. The multi-mode power management method of a gallium nitride adapter according to claim 1, characterized in that: The dynamic adjustment of power parameters also includes implementing a power management strategy based on heat redistribution, and the specific steps are as follows: Through the topological structure of the microchannel network, a heat transfer path is formed from the heat-sensitive area to the heat-resistant area; establishing a direct connection between the microchannel outlet of the heat-sensitive region and the microchannel inlet of the heat-resistant region; By controlling the flow distribution ratio of each microchannel segment, directional heat transfer is achieved, which reduces the temperature of GaN power devices while increasing the temperature of transformers and capacitors to achieve a more uniform temperature distribution.

10. A multi-mode power management device for a gallium nitride adapter, characterized in that: The device comprises a memory and one or more processors, wherein the memory stores executable code, and when the one or more processors execute the executable code, the device is configured to execute the multi-mode power management method for a gallium nitride adapter according to any one of claims 1 to 9.

Images