CN-121282442-B - Power supply multipoint thermal management method and device, computer equipment and storage medium
Abstract
The invention relates to a power supply multipoint heat management method, a device, computer equipment and a storage medium, wherein the power supply multipoint heat management method is based on distributed sensing and hierarchical control, temperature acquisition is realized through a temperature sensing array, temperature-power dual-mode control is adopted, when single-board-level self-adaptive cooling is executed, control quantity is generated through a segmentation PWM algorithm based on temperature data, energy efficiency optimization is realized based on temperature reference in a small current mode through a mode self-adaptive mechanism, an average value is calculated through FRM parameter configuration selection or a fixed rotating speed control strategy is set in a regional coordination level, the running state of a fan of an electric core board is converted into a control instruction of a regional big fan, and a small fan guiding big wind direction synergistic heat dissipation effect is formed. According to the invention, through a closed-loop control framework, 18 electric core plates can keep stable working temperature of 30-40 ℃ within 15-35 ℃, so that the problem of overheating of the electric core plates is effectively solved, the reliability of the system is improved, and the energy consumption is reduced.
Inventors
- LI TAO
- Yu dengyun
Assignees
- 深圳市卓讯达科技发展有限公司
Dates
- Publication Date
- 20260508
- Application Date
- 20251209
Claims (6)
- 1. A method of power multipoint thermal management for a high density battery cell cluster comprising a plurality of battery core boards, each battery core board configured with a temperature sensor and a turbofan, the system further comprising a plurality of regional large fans, the method comprising: Acquiring temperature data of each electric core board in real time through a distributed temperature sensing network, wherein the temperature sensing network is composed of a plurality of temperature sensors arranged on each electric core board; The method comprises the steps of executing single-board-level self-adaptive cooling control based on temperature data of each electric core board, wherein the single-board-level self-adaptive cooling control comprises temperature data fusion processing, over-temperature protection judgment, temperature offset correction and subsection PWM calculation, and independently controlling the rotating speed of a corresponding turbofan, executing an over-temperature processing flow if over-temperature protection conditions are triggered, performing offset correction on the current temperature by adopting a compensation algorithm in a fan unopened state, setting PWM to be a preset minimum rotating speed when the board temperature is lower than a first temperature threshold value, setting PWM to be a preset maximum rotating speed when the board temperature is higher than a second temperature threshold value, calculating a target PWM value according to the linear relation between a temperature value and the PWM value when the board Wen Chuyu is between the first temperature threshold value and the second temperature threshold value, wherein the temperature offset used in the temperature offset correction is a compensation value preset based on the thermal characteristics of a system and is used for eliminating measurement errors caused by heat accumulation when the fan is not started; Based on the real-time power load of the electric core plate, executing feedforward compensation control, including acquiring the electric working parameters of the electric core plate in real time through a voltage sensor and a current sensor, calculating real-time output power according to the acquired voltage value and current value, and converting the output power into a corresponding target PWM value based on the mapping relation between the preset power and PWM value; based on the working state of the system, executing mode self-adaptive control, including identification of a calibration mode and a small current mode and corresponding fan control, detecting whether the system is in the calibration mode, if so, controlling all fans to stop, detecting whether the system is in the small current mode, if so, further detecting the temperature of the radiator and the temperature of the small current module, and when the temperature of the radiator and the temperature of the small current module are lower than a set threshold value, controlling the fans to stop; Executing regional collaborative heat dissipation control based on the fan running state of the adjacent electric core plate group, wherein the regional collaborative heat dissipation control comprises real-time acquisition of PWM duty ratio values, data processing based on preset rules and rotating speed adjustment of regional big fans, real-time acquisition of current PWM duty ratio values of the turbofans of the adjacent electric core plate group, setting fan response mode parameters, and executing corresponding control operation according to the set fan response mode parameters; The regional cooperative heat dissipation control adopts a hierarchical linkage mechanism, and the rotating speed of the regional large fan is dynamically adjusted according to the running states of the turbofans of the adjacent electric core plates to form a distributed cooperative heat dissipation network.
- 2. The method of multipoint heat management for power supply according to claim 1, wherein the first temperature threshold is 30 ℃ and the second temperature threshold is 50 ℃, and wherein the linear relationship is such that PWM value increases linearly with temperature value within a range of 30 ℃ to 50 ℃.
- 3. The power multipoint thermal management method according to claim 1, wherein said setting fan response mode parameters supports the following dynamic configuration: The system presets a plurality of working modes, including an equalization mode, a control operation corresponding to the calculated average value, and at least one fixed rotating speed mode, a control operation corresponding to the set fixed value; The fan response mode parameter is supported to be manually set through a system configuration interface to directly select to execute corresponding control operation.
- 4. A power supply multipoint thermal management device, characterized by performing the power supply multipoint thermal management method according to any one of claims 1 to 3, the power supply multipoint thermal management device comprising: The temperature sensing module is used for acquiring temperature data of each electric core board in real time through a distributed temperature sensing network; the single board control module is used for executing single board-level self-adaptive cooling control based on the temperature data of each electric core board; The feedforward compensation module is used for executing feedforward compensation control based on the real-time power load of the electric core plate; the mode management module is used for executing mode self-adaptive control based on the working state of the system; and the regional cooperative module is used for executing regional cooperative heat dissipation control based on the running states of the turbofans of the adjacent electric core plate groups.
- 5. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the power multipoint thermal management method according to any of claims 1-3 when executing the program.
- 6. A computer readable storage medium, on which a computer program is stored, characterized in that the program, when executed by a processor, implements the power supply multipoint thermal management method according to any of claims 1-3.
Description
Power supply multipoint thermal management method and device, computer equipment and storage medium Technical Field The present invention relates to the field of power supply thermal management technologies, and in particular, to a power supply multipoint thermal management method, a device, a computer device, and a storage medium. Background In modern high-power and high-energy-density electronic equipment and energy storage systems, such as large server clusters, battery energy storage stations and high-end computing equipment, a core module of the system adopts a high-density array integration mode generally, and the structure of the system improves the performance of the system and simultaneously causes the outstanding problems of concentrated heating and uneven heat distribution. At present, the traditional thermal management scheme mainly adopts a global unified air cooling technology, and cannot accurately sense and respond to the temperature change of a local area by implementing a unified cooling strategy on the whole system level, so that the dilemma of coexistence of 'overcooling' and 'undercooling' is often faced in actual operation, unnecessary energy waste is caused in a low-load area, hot spot formation in a high-load area cannot be effectively restrained, the energy efficiency of the system is low, and the temperature control effect is not ideal. Therefore, there is an urgent need in the art for an intelligent thermal management method that can achieve both accurate temperature control, high energy utilization and system reliability, so as to solve the heat dissipation problem faced by the high-density battery cell group in practical application. Disclosure of Invention The invention mainly aims to provide a power supply multipoint thermal management method, a device, computer equipment and a storage medium, wherein each electric core board realizes local temperature sensing through 3 sensors at the bottom layer, and independently controls a turbofan to finish accurate single-point self-adjustment. On the upper layer, the 4 regional big fans dynamically adjust the air quantity according to the average rotating speed of the adjacent board fans, so as to form efficient regional cooperative heat dissipation. The two-stage framework jointly ensures the uniformity and stability of the temperature of 18 electric core plates in a unified environment, and realizes the balance of accuracy, high efficiency and energy conservation. To achieve the above object, the present invention provides a power supply multipoint thermal management method applied to a high density battery cell group including a plurality of battery core boards, each battery core board being configured with a temperature sensor and a turbofan, the system further including a plurality of regional large fans, the method comprising the steps of: Acquiring temperature data of each electric core board in real time through a distributed temperature sensing network, wherein the temperature sensing network is composed of a plurality of temperature sensors arranged on each electric core board; Based on the temperature data of each electric core board, executing single board-level self-adaptive cooling control, including temperature data fusion processing, over-temperature protection judgment, temperature offset correction and sectional PWM calculation, and independently controlling the rotating speed of the corresponding turbofan; Based on the real-time power load of the electric core plate, executing feedforward compensation control, including collecting electric parameters through a voltage-current sensor, calculating output power, and determining a compensation value based on a power PWM mapping relation; based on the system working state, executing mode self-adaptive control, including identification of a calibration mode and a small current mode and corresponding fan control; And executing regional cooperative heat dissipation control based on the fan running states of the adjacent electric core plates, wherein the regional cooperative heat dissipation control comprises real-time acquisition of PWM duty ratio values, data processing based on preset rules and rotation speed adjustment of regional large fans, the regional cooperative heat dissipation control adopts a hierarchical linkage mechanism, and the rotation speeds of the regional large fans are dynamically adjusted according to the turbine fan running states of the adjacent electric core plates to form a distributed cooperative heat dissipation network. Further, the step of executing the veneer-level adaptive cooling control specifically includes: if the over-temperature protection condition is triggered, executing an over-temperature treatment process; when the fan is not started, adopting a compensation algorithm to carry out offset correction on the current temperature; when the plate temperature is lower than a first temperature threshold value, setting PWM as a preset minimum