A CPU power supply system supporting power redundancy
By introducing redundant design of time and space dimensions into the CPU power system, dynamic scheduling algorithms and intelligent automatic switching methods are used to solve the problem of high failure rate of a single power supply, the rotation and load balancing of the power supply module are realized, and the power supply usage is extended. Lifespan and improve system stability.
Patent Information
- Application Number
- CN202411270382.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-11
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-09-11
AI Technical Summary
The existing CPU power system has increased the failure rate due to the long-term operation of a single power supply for a long time, and lacks multi-dimensional redundant power supply backup, resulting in premature failure of the power module, degraded system performance and equipment damage.
Design a CPU power system that supports power redundancy, adopts time-dimensional and spatial dimension redundant units, manages two 24V power modules through dynamic scheduling algorithms and intelligent automatic switching methods, realizes rotation and load balancing, and uses virtual visual control units to monitor and manage power status.
Extend the service life of the power module, ensure redundancy guarantee in the event of a single power supply failure, avoid system downtime, and improve the continuity and stability of system power supply.
Smart Images

Figure CN118796017B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of power supply redundancy design, in particular to a CPU power supply system supporting power supply redundancy. Background Art
[0002] A CPU power supply system supporting power redundancy is designed to improve the stability and reliability of the power supply system and extend the service life of the power supply module. Through a dynamic scheduling algorithm and an intelligent automatic switching method, the rotation and switching of two 24V power supply modules are controlled to achieve efficient utilization of the power supply modules and rapid fault recovery.
[0003] The existing CPU power supply system is usually a single power supply system, which cannot guarantee the normal operation of the CPU when a single power supply fails. In addition, since a single power supply will run at full load for a long time, the failure rate will increase and there is no multi-dimensional redundant power backup, which will lead to premature failure of the power module, system performance degradation and equipment damage. Therefore, a CPU power supply system that supports power redundancy is designed. Summary of the invention
[0004] The object of the present invention is to provide a CPU power supply system that supports power redundancy, so as to solve the problems raised in the above background technology that the failure rate of a single power supply will increase when it runs at full load for a long time and there is no multi-dimensional redundant power supply backup, which will lead to premature failure of the power supply module, degradation of system performance and damage to the equipment.
[0005] To achieve the above object, the present invention provides a CPU power supply system supporting power redundancy, comprising:
[0006] A time dimension redundancy unit, wherein the time dimension redundancy unit adopts a dynamic scheduling algorithm and an intelligent automatic switching method to regularly manage and switch two power supply modules;
[0007] It also includes a space dimension redundancy unit, which is used for distributed deployment and management of the two power supply modules by distributing the two power supply modules in the physical space;
[0008] It also includes a load management unit, which uses an intelligent load distribution algorithm to optimize the load balance between the two power modules;
[0009] It also includes a virtual visualization control unit, which is used to monitor and manage the status of the two power modules, display the data of the two power modules in real time through a virtual visualization interface, and manually switch the power supply in time and space dimensions and adjust the load distribution operations.
[0010] As a further improvement of the technical solution, the time dimension redundancy unit includes a rotation scheduling module and an automatic switching module;
[0011] The rotation scheduling module uses a dynamic scheduling algorithm and time management mechanism to and power modules Health status at all times , formulate a rotation plan for the two power modules;
[0012] The automatic switching module automatically switches the two power supply modules using the switching control logic and the redundancy guarantee mechanism according to the plan formulated by the rotation scheduling module;
[0013] The dynamic scheduling algorithm and time management mechanism are used to determine the rotation order and time of the two power modules and generate a rotation plan for the two power modules. The specific steps are as follows:
[0014] S1.1.1. Collect the initial health status of the power module , Temperature of each power module at the moment ,Voltage , Current , Initial working time and initial rest time and current load demand ;
[0015] S1.1.2. Apply dynamic scheduling algorithm and time management mechanism to optimize the objective function and determine the rotation order and time of the two power modules;
[0016] S1.1.3. Generate a rotation plan for two power modules based on the optimization solution results;
[0017] The rotation plan of the two power modules in S1.1.3 includes: the working time period, rest time period, switching time point of each power module within the specified time period and the load distribution of the two power modules in different time periods.
[0018] As a further improvement of the technical solution, the dynamic scheduling algorithm and time management mechanism in S1.1.2 optimize the objective function and determine the rotation order and time of the two power modules. The specific method is as follows:
[0019] S1.2.1. Calculate the current health status of each power module and load requirements ;
[0020] S1.2.2. Calculate the cumulative working time of each power module and accumulated rest time ;
[0021] ;
[0022] ;
[0023] in, For the The power modules are The accumulated working time; For the The power modules are The accumulated rest time; For the The power modules are The switch state, Indicates the power module at time In working state; Indicates the power module at time In working state;
[0024] S1.2.3. According to health status And output power , establish the objective function;
[0025] ;
[0026] in, is the total operating cost, and the optimization goal is to minimize this cost. For the The initial health status of each power module, For the Load at time The output power, For the The power modules are health status;
[0027] S1.2.4. Use the gradient descent method to set parameters, optimize the objective function, update the working time and rest time of each power module, and determine the rotation order and time of the two power modules.
[0028] As a further improvement of this technical solution, the current health status of each power module is calculated in S1.2.1 and load requirements , the specific method is as follows:
[0029] ;
[0030] ;
[0031] ;
[0032] in, For the The power modules are health status; For the Initial health status of each power module; For the The power modules are Output power; For the The power modules are Ambient temperature; is the rate function of health status degradation, which represents the impact of power and temperature on health status; For in time Total load demand; For the Load at time Output power;
[0033] As a further improvement of the technical solution, the gradient descent method is used in S1.2.4 to set parameters, optimize the objective function, update the working time and rest time of each power module, and determine the rotation order and time of the two power modules. The specific steps are as follows:
[0034] S1.2.4.1. Initialize the working time and rest time of each power module, set the initial parameters of the gradient descent method, and set the initial working time and initial rest time , set the learning rate , initialize the parameters of the optimization objective function ;
[0035] S1.2.4.2. Calculation of objective function With respect to each parameter The gradient is:
[0036] For each time point , calculate the derivative of the objective function:
[0037] ;
[0038] Using the chain rule, we can calculate the derivative:
[0039] ;
[0040] ;
[0041] S1.2.4.3. Update each time point using gradient descent Parameters :
[0042] ;
[0043] S1.2.4.4. According to the updated parameters , calculate the working time and rest time of each power module:
[0044] ;
[0045] ;
[0046] S1.2.4.5. In each iteration, repeatedly calculate the gradient, update the parameters, and update the work and rest time, and check the objective function Whether it converges or whether the number of iterations reaches the upper limit;
[0047] S1.2.4.6. Analyze the final value, determine the switching status of each power module at each time point, and generate a detailed rotation plan, including the working time, rest time and switching time point of each module.
[0048] As a further improvement of the technical solution, the spatial dimension redundancy unit includes a distributed deployment module and a control switching circuit module;
[0049] Wherein, the distributed deployment module is used to deploy two 24V power modules at different locations in the physical space respectively;
[0050] The control switching circuit module is equipped with a control switching device to monitor in real time whether the two power modules fail and switch the two power modules to each other;
[0051] The control switching circuit module realizes the mutual switching of the two power supplies, and the specific steps are as follows:
[0052] S2.1, collect voltage, current and temperature data of the power module in real time, and control the switching device to determine whether the power module fails according to the collected data;
[0053] S2.2, after detecting a fault, the control switching device sends a switching signal to the CPU;
[0054] S2.3, CPU switches the power module in real time.
[0055] As a further improvement of the technical solution, the load management unit includes a load balancing module and a dynamic adjustment module;
[0056] The load balancing module uses an intelligent load distribution algorithm to dynamically adjust the load distribution of the two power modules and the output power of the power modules according to real-time load demand, health status and current load;
[0057] The dynamic adjustment module monitors the voltage, current and temperature of the power module in real time to obtain the power module data when the load changes, and feeds the power module data back to the load balancing module to check whether the load distribution is completed.
[0058] As a further improvement of this technical solution, the load balancing module uses an intelligent load distribution algorithm according to the real-time total load demand. Dynamically adjust the load distribution of the two power modules. The specific method is as follows:
[0059] ;
[0060] ;
[0061] in, For the system at time Total load demand; For the first 24V power module at time Output power; For the second 24V power module at time Output power; For the The power modules are Output power; For all power modules at time The sum of health conditions.
[0062] As a further improvement of the technical solution, the virtual visualization control unit includes a real-time monitoring module and a visualization interface module;
[0063] The real-time monitoring module displays the data of each power module in real time, including temperature, voltage, current, working time and health status data, and sets the alarm threshold. When an abnormal situation is detected, an alarm is immediately issued and displayed on the visual interface module;
[0064] The visualization interface module displays the collected real-time data, historical data, and all data of the time dimension redundant unit, space dimension redundant unit, and load management unit in a graphical manner, and provides a control panel, and can manually switch the time and space dimension power supply and adjust the load distribution operation.
[0065] Compared with the prior art, the present invention has the following beneficial effects:
[0066] 1. In the CPU power supply system supporting power redundancy, through the power supply redundancy design in the time and space dimensions, using dynamic scheduling algorithms and intelligent automatic switching methods, the two power supply modules can be regularly managed and switched to extend the service life of the modules, and distributed deployment and management can be performed. When a single power supply module fails, the other module can continue to provide redundant power, thereby avoiding system downtime caused by a single failure.
[0067] 2. In the CPU power supply system supporting power redundancy, by using an intelligent load distribution algorithm to optimize the load balance between the two power modules, each power module can operate in the best working state, prevent overload and uneven load, improve the overall efficiency of the system, and ensure the continuity and stability of power supply. BRIEF DESCRIPTION OF THE DRAWINGS
[0068] Figure 1 It is the overall flow chart of the present invention;
[0069] The meaning of each number in the figure is:
[0070] 1. Time dimension redundant unit; 2. Space dimension redundant unit; 3. Load management unit; 4. Virtual visualization control unit. DETAILED DESCRIPTION
[0071] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. Example
[0072] See also Figure 1 As shown, a CPU power supply system supporting power redundancy is provided, including
[0073] The time dimension redundancy unit 1 adopts a dynamic scheduling algorithm and an intelligent automatic switching method to regularly manage and switch two power supply modules, including a rotation scheduling module and an automatic switching module;
[0074] Among them, the rotation scheduling module uses a dynamic scheduling algorithm and time management mechanism to and power modules Health status at all times , formulate a rotation plan for the two power modules;
[0075] Among them, the automatic switching module uses the switching control logic and redundant protection mechanism to automatically switch the two power supply modules according to the plan formulated by the rotation scheduling module.
[0076] Among them, the time management mechanism manages and records the working time and rest time of each power module;
[0077] The time management mechanism is based on the accumulated working time of the power module. and accumulated rest time Dynamically adjust the working and resting states of the modules to ensure that each module will not overheat or be damaged due to excessive use. The specific implementation process is as follows:
[0078] Through the dynamic scheduling algorithm, the time management mechanism tracks the working and resting time of each module, ensuring that each module can switch to the resting state in time after accumulating a certain amount of working time;
[0079] The time management mechanism works together with the dynamic scheduling algorithm to ensure that the two power modules are rotated in a timely manner within the specified time period. According to the current working time and rest time of each power module, the time management mechanism determines when to switch, thereby achieving efficient module rotation;
[0080] The time management mechanism also adjusts the load demand Dynamically adjust the working time and rest time of the power module to ensure balanced distribution of load between the two modules. The specific implementation process is as follows:
[0081] According to the current load demand, the time management mechanism can adjust the power module's on time, distribute the load to the module with better health, reduce the load pressure of another module, and achieve efficient load balancing of the system;
[0082] The dynamic scheduling algorithm and time management mechanism are used to determine the rotation order and time of the two power modules and generate a rotation plan for the two power modules. The specific steps are as follows:
[0083] S1.1.1. Collect the initial health status of the power module , Temperature of each power module at the moment ,Voltage , Current , Initial working time and initial rest time and current load demand ;
[0084] Among them, the initial health status of the power module It refers to the health status of each power module at the beginning of system operation. When calculating and monitoring the changes in the health status of the power modules, a benchmark point is needed to measure. The dynamic scheduling algorithm needs to understand the current health status of the power modules in order to optimize the switching strategy and working time. According to the initial health status and real-time data, the use strategy of the power modules is adjusted to extend their service life. In addition, the initial health status of different power modules may be different, and the load needs to be allocated according to the health status.
[0085] S1.1.2. Apply dynamic scheduling algorithm and time management mechanism to optimize the objective function and determine the rotation order and time of the two power modules;
[0086] S1.1.3. Generate a rotation plan for two power modules based on the optimization solution results;
[0087] Among them, the rotation plan of the two power modules in S1.1.3 includes: the working time period, rest time period, switching time point of each power module within the specified time period and the load distribution of the two power modules in different time periods.
[0088] The dynamic scheduling algorithm and time management mechanism in S1.1.2 optimize the objective function and determine the rotation order and time of the two power modules. The specific method is as follows:
[0089] S1.2.1. Calculate the current health status of each power module and load requirements ;
[0090] S1.2.2. Calculate the cumulative working time of each power module and accumulated rest time ;
[0091] ;
[0092] ;
[0093] in, For the The power modules are The accumulated working time; For the The power modules are The accumulated rest time; For the The power modules are The switch state, Indicates the power module at time In working state; Indicates the power module at time In working state;
[0094] S1.2.3. According to health status And output power , establish the objective function;
[0095] ;
[0096] in, is the total operating cost, and the optimization goal is to minimize this cost. For the The initial health status of each power module, For the Load at time The output power, For the The power modules are health status;
[0097] S1.2.4. Use the gradient descent method to set parameters, optimize the objective function, update the working time and rest time of each power module, and determine the rotation order and time of the two power modules.
[0098] S1.2.1 Calculate the current health status of each power module and load requirements , the specific method is as follows:
[0099] ;
[0100] ;
[0101] ;
[0102] in, For the The power modules are health status; For the Initial health status of each power module; For the The power modules are Output power; For the The power modules are Ambient temperature; is the rate function of health status degradation, which represents the impact of power and temperature on health status; For in time Total load demand; For the Load at time Output power;
[0103] In S1.2.4, the gradient descent method is used to set parameters, optimize the objective function, update the working time and rest time of each power module, and determine the rotation order and time of the two power modules. The specific steps are as follows:
[0104] S1.2.4.1. Initialize the working time and rest time of each power module, set the initial parameters of the gradient descent method, and set the initial working time and initial rest time , set the learning rate , initialize the parameters of the optimization objective function ;
[0105] In this embodiment, , , , Both are 0; learning rate ;
[0106] S1.2.4.2. Calculation of objective function With respect to each parameter The gradient is:
[0107] For each time point , calculate the derivative of the objective function:
[0108] ;
[0109] Using the chain rule, we can calculate the derivative:
[0110] ;
[0111] ;
[0112] S1.2.4.3. Update each time point using gradient descent Parameters :
[0113] ;
[0114] S1.2.4.4. According to the updated parameters , calculate the working time and rest time of each power module:
[0115] ;
[0116] ;
[0117] S1.2.4.5. In each iteration, repeatedly calculate the gradient, update the parameters, and update the work and rest time, and check the objective function Whether it converges or whether the number of iterations reaches the upper limit;
[0118] S1.2.4.6. Analyze the final value, determine the switching status of each power module at each time point, and generate a detailed rotation plan, including the working time, rest time and switching time point of each module.
[0119] Among them, the final analysis The switch status of each power module at each time point is determined, and a detailed rotation plan is generated. In this example, the following is shown:
[0120] ;
[0121] ;
[0122] according to Value, working time period, rest time period and switching time point, generate a detailed rotation plan for each power module:
[0123] Rotation plan for power module 1:
[0124] Working hours: 0:00-4:00, 8:00-12:00, 16:00-20:00
[0125] Rest time: 4:00-8:00, 12:00-16:00, 20:00-24:00
[0126] Switching time: 4:00, 8:00, 12:00, 16:00, 20:00
[0127] Rotation plan for power module 2:
[0128] Working hours: 4:00-8:00, 12:00-16:00, 20:00-24:00
[0129] Rest time: 0:00-4:00, 8:00-12:00, 16:00-20:00
[0130] Switching time points: 4:00, 8:00, 12:00, 16:00, 20:00.
[0131] It also includes a space dimension redundancy unit 2, which is used for distributed deployment and management of two power supply modules by distributing the two power supply modules in physical space, including a distributed deployment module and a control switching circuit module;
[0132] Among them, the distributed deployment module is used to deploy two 24V power modules in different locations in the physical space;
[0133] The control switching circuit module is equipped with a control switching device to monitor in real time whether the two power modules have any faults and switch the two power modules to each other;
[0134] The control switching circuit module realizes the mutual switching of the two power supplies. The specific steps are as follows:
[0135] S2.1, collect voltage, current and temperature data of the power module in real time, and control the switching device to determine whether the power module fails according to the collected data;
[0136] S2.2, after detecting a fault, the control switching device sends a switching signal to the CPU;
[0137] S2.3, CPU switches the power module in real time.
[0138] It also includes a load management unit 3, which uses an intelligent load distribution algorithm to optimize the load balance between the two power modules, including a load balancing module and a dynamic adjustment module;
[0139] The load balancing module uses an intelligent load distribution algorithm to dynamically adjust the load distribution of the two power modules and the output power of the power modules according to real-time load demand, health status and current load;
[0140] The dynamic adjustment module monitors the voltage, current and temperature of the power module in real time to obtain the power module data when the load changes, and feeds the power module data back to the load balancing module to check whether the load distribution is completed;
[0141] The load balancing module uses an intelligent load distribution algorithm based on the real-time total load demand. Dynamically adjust the load distribution of the two power modules. The specific method is as follows:
[0142] ;
[0143] ;
[0144] in, For the system at time Total load demand; For the first 24V power module at time Output power; For the second 24V power module at time Output power; For the The power modules are Output power; For all power modules at time The sum of health conditions.
[0145] It also includes a virtual visualization control unit 4, which is used to monitor and manage the status of the two power modules, display the data of the two power modules in real time through a virtual visualization interface, and manually switch and adjust the load distribution operation of the power supply in time and space dimensions, including a real-time monitoring module and a visualization interface module;
[0146] The real-time monitoring module displays the data of each power module in real time, including temperature, voltage, current, working time and health status data, and sets the alarm threshold. When an abnormal situation is detected, an alarm is immediately issued and displayed in the visual interface module;
[0147] The visualization interface module displays the collected real-time data, historical data, and all data of the time dimension redundant unit 1, space dimension redundant unit 2, and load management unit 3 in a graphical manner, and provides a control panel, and can manually switch the time and space dimension power supply and adjust the load distribution operations.
[0148] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments, and the above embodiments and descriptions are only preferred examples of the present invention, and are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements all fall within the scope of the present invention to be protected.
Claims
1. A CPU power supply system supporting power redundancy, characterized in that: include: A time dimension redundancy unit (1), wherein the time dimension redundancy unit (1) uses a dynamic scheduling algorithm and an intelligent automatic switching method to regularly manage and switch two power supply modules; A space dimension redundancy unit (2), wherein the space dimension redundancy unit (2) is used for distributed deployment and management of the two power supply modules by distributing the two power supply modules in a physical space; A load management unit (3), wherein the load management unit (3) uses an intelligent load distribution algorithm to optimize the load balance between the two power modules; A virtual visualization control unit (4), the virtual visualization control unit (4) being used to monitor and manage the status of the two power modules, to display the data of the two power modules in real time through a virtual visualization interface, and to manually switch the power supplies in time and space dimensions and to adjust load distribution operations; The time dimension redundancy unit (1) comprises a rotation scheduling module and an automatic switching module; The rotation scheduling module uses a dynamic scheduling algorithm and time management mechanism to and power modules Health status at all times , formulate a rotation plan for the two power modules; The automatic switching module automatically switches the two power supply modules using the switching control logic and the redundancy guarantee mechanism according to the plan formulated by the rotation scheduling module; The dynamic scheduling algorithm and time management mechanism are used to determine the rotation order and time of the two power modules and generate a rotation plan for the two power modules. The specific steps are as follows: S1.1.
1. Collect the initial health status of the power module , Temperature of each power module at the moment ,Voltage , Current , Initial working time and initial rest time and current load demand ; S1.1.
2. Apply dynamic scheduling algorithm and time management mechanism to optimize the objective function and determine the rotation order and time of the two power modules; S1.1.
3. Generate a rotation plan for two power modules based on the optimization solution results; The two-way power module rotation plan in S1.1.3 includes: the working time period, rest time period, switching time point of each power module within the specified time period and the load distribution of the two power modules in different time periods; The dynamic scheduling algorithm and time management mechanism in S1.1.2 optimize the objective function and determine the rotation order and time of the two power modules. The specific method is as follows: S1.2.
1. Calculate the current health status of each power module and load requirements ; S1.2.
2. Calculate the cumulative working time of each power module and accumulated rest time ; ; ; in, For the The power modules are The accumulated working time; For the The power modules are The accumulated rest time; For the The power modules are The switch state, Indicates the power module at time In working state; Indicates the power module at time In working state; S1.2.
3. According to health status And output power , establish the objective function; ; in, is the total operating cost, and the optimization goal is to minimize this cost. For the The initial health status of each power module, For the Load at time The output power, For the The power modules are health status; S1.2.
4. Use the gradient descent method to set parameters, optimize the objective function, update the working time and rest time of each power module, and determine the rotation order and time of the two power modules; S1.2.1 calculates the current health status of each power module and load requirements , the specific method is as follows: ; ; ; in, For the The power modules are health status; For the The initial health status of each power module; For the The power modules are Output power; For the The power modules are Ambient temperature; is the rate function of health status degradation, which represents the impact of power and temperature on health status; For in time Total load demand; For the Load at time Output power; In S1.2.4, the gradient descent method is used to set parameters, optimize the objective function, update the working time and rest time of each power module, and determine the rotation order and time of the two power modules. The specific steps are as follows: S1.2.4.
1. Initialize the working time and rest time of each power module, set the initial parameters of the gradient descent method, and set the initial working time and initial rest time , set the learning rate , initialize the parameters of the optimization objective function ; S1.2.4.
2. Calculation of objective function With respect to each parameter The gradient is: For each time point , calculate the derivative of the objective function: ; Using the chain rule, we can calculate the derivative: ; ; S1.2.4.
3. Update each time point using gradient descent Parameters : ; S1.2.4.
4. According to the updated parameters , calculate the working time and rest time of each power module: ; ; S1.2.4.
5. In each iteration, repeatedly calculate the gradient, update the parameters, and update the work and rest time, and check the objective function Whether it converges or whether the number of iterations reaches the upper limit; S1.2.4.
6. Analyze the final value, determine the switching status of each power module at each time point, and generate a detailed rotation plan, including the working time, rest time and switching time point of each module.
2. The CPU power supply system supporting power redundancy according to claim 1, characterized in that: The spatial dimension redundancy unit (2) comprises a distributed deployment module and a control switching circuit module; Wherein, the distributed deployment module is used to deploy two 24V power modules at different locations in the physical space respectively; The control switching circuit module is equipped with a control switching device to monitor in real time whether the two power modules fail and switch the two power modules to each other; The control switching circuit module realizes the mutual switching of the two power supplies, and the specific steps are as follows: S2.1, collect voltage, current and temperature data of the power module in real time, and control the switching device to determine whether the power module fails according to the collected data; S2.2, after detecting a fault, the control switching device sends a switching signal to the CPU; S2.3, CPU switches the power module in real time.
3. The CPU power supply system supporting power redundancy according to claim 1, characterized in that: The load management unit (3) comprises a load balancing module and a dynamic adjustment module; The load balancing module uses an intelligent load distribution algorithm to dynamically adjust the load distribution of the two power modules and the output power of the power modules according to real-time load demand, health status and current load; The dynamic adjustment module monitors the voltage, current and temperature of the power module in real time to obtain the power module data when the load changes, and feeds the power module data back to the load balancing module to check whether the load distribution is completed.
4. The CPU power supply system supporting power redundancy according to claim 3, characterized in that: The load balancing module uses an intelligent load distribution algorithm based on the real-time total load demand Dynamically adjust the load distribution of the two power modules. The specific method is as follows: ; ; in, For the system at time Total load demand; For the first 24V power module at time Output power; For the second 24V power module at time Output power; For the The power modules are Output power; For all power modules at time The sum of health conditions.
5. The CPU power supply system supporting power redundancy according to claim 1, characterized in that: The virtual visualization control unit (4) comprises a real-time monitoring module and a visualization interface module; The real-time monitoring module displays the data of each power module in real time, including temperature, voltage, current, working time and health status data, and sets the alarm threshold. When an abnormal situation is detected, an alarm is immediately issued and displayed on the visual interface module; The visual interface module graphically displays the collected real-time data, historical data, and all data of the time dimension redundant unit (1), the space dimension redundant unit (2), and the load management unit (3), and provides a control panel, and can manually switch the time and space dimension power supplies and adjust the load distribution operations.
