Power supply management method, device and system and computer storage medium
Through the uninterruptible power supply system of the mother-child type and dynamic power distribution technology, the problem of power imbalance in the traditional power supply mode is solved, the power supply balance of various components of complex equipment is ensured, and the stability of the equipment and system reliability are improved.
Patent Information
- Application Number
- CN202411121438.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-15
- Publication Date
- 2024-11-29
AI Technical Summary
The traditional power supply model cannot effectively ensure the power supply balance of various components of complex and large equipment when power demand peaks, resulting in the inability to receive priority power support for important components, affecting the normal operation of the equipment.
The mother-child uninterruptible power system is adopted to detect the status of the main power and the main backup power supply, dynamically adjust the power supply, and distribute power according to the power demand weight of each component to ensure that key components are given priority power supply.
The equipment is operated stably under various power situations, which improves the reliability and stability of the system, and avoids failures or performance degradation caused by power imbalance.
Smart Images

Figure CN119051235A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of power system management, and in particular to a power supply management method, device, system and computer storage medium. Background Art
[0002] With the rapid development and widespread application of information technology, uninterruptible power supply (UPS) technology plays an increasingly important role in modern society. The main function of UPS technology is to immediately take over the power supply when the power supply is interrupted or abnormal, ensuring that key equipment such as computers and communication equipment continue to operate normally, thereby avoiding serious consequences such as data loss and equipment damage.
[0003] For complex and large-scale equipment, its power supply requirements are usually more complicated, and the power and power shortage weights of different components are different. In order to ensure the normal operation of the equipment, some users usually choose to power each component of the equipment independently, which requires multiple UPS devices to ensure that each component has a stable power supply. In order to save costs, some users will choose to use a UPS device to power multiple components of multiple devices at the same time. Although the cost is saved, the power requirements of different components are different. If only one UPS device is used to power multiple components, some components may not get enough power when the power demand is peak, causing the problem of unbalanced power supply. In particular, the disadvantage weights of some components are high, that is, the overall operation of the equipment is greatly affected in the case of disadvantages, but they cannot get priority power supply guarantee when the power is tight. Therefore, the traditional simple power supply mode cannot achieve power balance, and important components cannot get enough power support, which affects the normal operation of the equipment. Summary of the invention
[0004] In view of the above, the present application provides a power supply management method, device, system and computer storage medium to perform power distribution according to power demand weights to solve the technical problem of unbalanced power supply in the prior art.
[0005] A first aspect of the present application provides a power supply management method, which is applied to a power supply management system, wherein the power supply management system includes a parent-child uninterruptible power supply system, wherein the parent-child uninterruptible power supply system includes a main power supply and at least one backup power supply, and the power supply management method includes: When it is detected that the mains power is in an abnormal state, the power supply source of the target AI device is switched from the mains power to the main power supply; when it is detected that the main power supply is in an abnormal state, the power supply source is switched from the main power supply to the backup power supply; when it is detected that the backup power supply is in a power shortage state, the power demand weight of each component of the target AI device is determined; The backup power supply is controlled to supply power to the plurality of components according to the power demand weights.
[0006] In an optional implementation, determining the power demand weight of each component of the target AI device includes: Acquiring operating parameters of a target component, wherein the target component is any one of the multiple components, and the operating parameters include operating load, operating temperature, and power consumption; The operating load, the operating temperature and the power consumption are standardized to obtain a first standard score corresponding to the operating load, a second standard score corresponding to the operating temperature and a third standard score corresponding to the power consumption; a first weight factor corresponding to the operating load, a second weight factor corresponding to the operating temperature and a third weight factor corresponding to the power consumption are determined; The power demand weight of the target component is determined according to the first standard score, the second standard score, the third standard score, the first weight factor, the second weight factor, and the third weight factor.
[0007] In an optional implementation, controlling the backup power supply to supply power to the multiple components according to the power demand weights includes: Determine a correlation index between each of the multiple components and the business goal of the target AI device, and determine an importance level corresponding to each component according to the correlation index; Prioritizing the multiple components according to the importance levels and the power demand weights to obtain a power supply priority corresponding to each component; An operating state of the backup power supply is determined, and the backup power supply is controlled to supply power to the plurality of components according to the operating state and the power supply priority.
[0008] In an optional implementation, controlling the backup power supply to supply power to the multiple components according to the operating state and the power supply priority includes: The target power requirement of the i-th component among the multiple components is determined by the following calculation formula: Among them, Ei is the target power demand, P i is the power supply priority of the i-th component, S i is the maximum power provided by the backup power supply to the i-th component at the current moment, ∑P j a sum of power supply priorities for each of the plurality of components; The backup power supply is controlled to supply power to the i-th component according to the target power demand.
[0009] In an optional embodiment, the method further comprises: When the number of power supplies of the backup power supply is greater than 3, when it is detected that the main power supply is in an abnormal state, the power supply source is switched from the main power supply to the first-level backup power supply in the backup power supply; When the first-level backup power supply is in an abnormal state, switching the power supply source from the first-level backup power supply to the second-level backup power supply; When the second-level backup power source is in an abnormal state, the power supply source is switched from the second-level backup power source to a third-level backup power source until each of the backup power sources is in an abnormal state.
[0010] In an optional embodiment, the method further comprises: Acquire the power status of the target AI device according to a preset time period; Predicting the target AI device according to the power status, and outputting a power problem of the target AI device; A processing strategy is formulated according to the power problem, and the power problem and the processing strategy are sent to a preset terminal.
[0011] A second aspect of the present application provides a power supply management device, the power supply management device comprising: A first detection switching module, configured to switch the power supply source of the target AI device from the mains to the main power supply when detecting that the mains power is in an abnormal state; A second detection switching module, configured to switch the power supply source from the main power supply to the backup power supply when detecting that the main power supply is in an abnormal state; A determination module, configured to determine a power demand weight of each component of the target AI device when detecting that the backup power supply is in a power shortage state; A control module is used to control the backup power supply to supply power to the multiple components according to the power demand weights.
[0012] A third aspect of the present application provides a power supply management system, the power supply management system comprising a parent-child uninterruptible power supply system, the parent-child uninterruptible power supply system comprising: A primary power source and at least one backup power source; The main power supply is used to replace the mains power to provide power supply to the target AI device when it is detected that the mains power is in an abnormal state; The backup power supply is used to replace the main power supply to provide power supply to the target AI device when it is detected that the main power supply is in an abnormal state.
[0013] In an optional embodiment, the power supply management system also includes a power supply control device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the steps of the power supply management method when executing the computer program.
[0014] A fourth aspect of the present application provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the above-mentioned power supply management method when executed by a processor.
[0015] In summary, the power supply management method, device, system and computer storage medium provided by the present application, by using a parent-child uninterruptible power supply system, automatically switches the power supply source of the target AI device from the mains to the main power supply when an abnormality in the mains is detected, thereby ensuring continuous power supply to the device. If the main power supply is abnormal, it will automatically switch to the backup power supply again to ensure the continuity of the power supply. In addition, when the backup power supply is in a state of power shortage, the power supply is dynamically adjusted according to the power demand weights of each component of the equipment to ensure that each component can get appropriate power supply, thereby avoiding equipment failure or performance degradation caused by power imbalance. In general, through intelligent power supply management, the stable operation of the equipment under various power conditions is ensured, and the reliability and stability of the system are improved. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] Figure 1 is a structural schematic diagram of a power supply management system shown in an embodiment of the present application; Figure 2 is a flow chart of a power supply management method shown in an embodiment of the present application; Figure 3 is a functional module diagram of a power supply management device shown in an embodiment of the present application; Figure 4 It is a structural schematic diagram of a power supply control device shown in an embodiment of the present application.
[0017] Description of Reference Numerals 1. Power supply management system; 10. Parent-child uninterruptible power supply system; 101. Main power supply; 102. Backup power supply; 12. Power supply control device; 121. Memory; 122. Processor; 123. Communication bus. DETAILED DESCRIPTION
[0018] The terms used in the following embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to be used as limitations to the present application. As used in the specification of the present application, the singular expressions "one", "a kind of", "said", "above", "the" and "this" are intended to also include plural expressions, unless there is a clear indication to the contrary in the context. It should also be understood that the term "and / or" used in the present application refers to any or all possible combinations comprising one or more of the listed items.
[0019] In the following, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as suggesting or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features, and in the description of the embodiments of the present application, unless otherwise specified, "plurality" means two or more.
[0020] Reference Figure 1 , which is a schematic diagram of the structure of a power supply management system shown in an embodiment of the present application.
[0021] The power supply management system 1 includes a parent-child uninterruptible power supply system 10 and a power supply control device 12. The parent-child uninterruptible power supply system 10 includes a main power supply 101 and at least one backup power supply 102. The main power supply 101 is used as a parent machine to replace the main power supply to provide power supply to the target AI device when the main power is detected to be in an abnormal state. The backup power supply 102 is used as a child machine to replace the main power supply to provide power supply to the target AI device when the main power is detected to be in an abnormal state.
[0022] In some embodiments, the main power supply 101, as the core of the parent-child uninterruptible power supply system 10, can be responsible for the power supply management and power supply distribution of the parent-child uninterruptible power supply system 10. The main power supply 101 has a high-capacity battery pack, which can quickly switch to the battery power supply mode when the mains power is in an abnormal state (for example, the mains power is interrupted or the mains power is unstable, etc.), ensuring a stable and reliable power supply for the target AI device.
[0023] In some embodiments, the backup power supply 102 is usually small in size and weight so that it can be installed inside or near the target AI device within an allowable range. The backup power supply 102 has the ability to respond quickly and precisely control, and can quickly switch to the battery power supply mode when the main power supply 101 is in an abnormal state (for example, the main power supply fails or the main power supply is insufficient, etc.), ensuring a stable and reliable power supply for the target AI device.
[0024] The power supply control device 12 is used to implement the power supply management method.
[0025] Reference Figure 2 , which is a flow chart of a power supply management method shown in an embodiment of the present application, and the power supply management method includes the following steps.
[0026] S21, when it is detected that the mains power is in an abnormal state, switching the power supply source of the target AI device from the mains power to the main power supply.
[0027] Among them, the mains refers to the power source mainly used by residents in the city, and the target AI device refers to the subject for which power supply management is executed. The application of artificial intelligence (AI) devices is becoming more and more extensive, but with the continuous improvement of the complexity and functionality of AI devices, the requirements for power supply are also getting higher and higher. Under normal circumstances, the mains is used to supply power to the target AI device. When it is detected that the mains is in an abnormal state, for example, the mains is interrupted or the mains is unstable, the power supply management device can switch the power supply source of the target AI device from the mains to the main power supply, and then use the main power supply to supply power to the target AI device.
[0028] In some embodiments, the power supply management device can monitor the voltage, current, frequency and other operating parameters of the mains in real time through smart meters, sensors and other devices, and generate corresponding waveforms to identify whether the mains has fluctuations and unstable behavior. Secondly, the power supply management device can also monitor the power status of the target AI device in real time. When the power status of the target AI device is in the off state, it can be determined that the target AI device is in a power-off state, that is, the mains is in an interrupted state. In other embodiments, the power supply management device can collect and organize the historical operating parameters of the mains, and establish a machine learning model, such as a random forest, a support vector machine (SVM) or a neural network, and obtain an abnormal recognition model based on the historical operating parameters. The operating parameters of the mains are detected in real time and input into the abnormal recognition model for prediction to output the result of whether the mains is in an abnormal state.
[0029] Through the above optional implementation method, by real-time detection of the mains power status, a rapid response to mains power anomalies and stable power supply of the target AI device are achieved, which effectively improves the reliability and stability of the AI device and ensures its normal operation, thereby improving the power supply efficiency.
[0030] S22, when it is detected that the main power supply is in an abnormal state, switching the power supply source from the main power supply to the backup power supply.
[0031] When the main power supply is used to supply power to the target AI device, according to the same embodiment as step S21, the power supply management device can detect whether the main power supply is in an abnormal state. Assuming that the main power supply is detected to be in an abnormal state, for example, the main power supply fails or the main power supply is insufficient, the power supply management device can switch the power supply source of the target AI device from the main power supply to the backup power supply.
[0032] Through the above optional implementation, by quickly responding to the abnormality of the main power supply and switching to the backup power supply in time, the continuous power supply of the target AI device in the case of abnormal main power supply is ensured, reducing the risk of downtime and data loss of the target AI device caused by power outage or shortage, and improving the reliability and stability of the target AI device. Secondly, it can effectively protect the target AI device from power fluctuations or power failures, extend the service life of the device, and reduce the possibility of equipment damage.
[0033] S23, when it is detected that the backup power supply is in a power shortage state, determine the power demand weight of each component of the multiple components of the target AI device.
[0034] According to the embodiment of detecting whether the mains power is in an abnormal state in step S21, the power supply management device can detect whether the backup power supply is in a power shortage state.
[0035] The target AI device is usually composed of multiple components, some of which are more critical to the normal operation of the device. When it is detected that the backup power supply is in a power shortage state, the power supply management device can determine the power demand weight of each of the multiple components of the target AI device to ensure that when the backup power supply is in a power shortage, the components that are critical to the function of the device are given priority to maintain the basic function of the device.
[0036] In an optional implementation, determining the power demand weight of each component of the target AI device includes: Acquiring operating parameters of a target component, wherein the target component is any one of the multiple components, and the operating parameters include operating load, operating temperature, and power consumption; The operating load, the operating temperature and the power consumption are standardized to obtain a first standard score corresponding to the operating load, a second standard score corresponding to the operating temperature and a third standard score corresponding to the power consumption; a first weight factor corresponding to the operating load, a second weight factor corresponding to the operating temperature and a third weight factor corresponding to the power consumption are determined; The power demand weight of the target component is determined according to the first standard score, the second standard score, the third standard score, the first weight factor, the second weight factor, and the third weight factor.
[0037] In some embodiments, when the power supply management device needs to determine the power demand weight corresponding to each of multiple components of the target AI device, it arbitrarily selects one component from the multiple components as the target component and preferentially calculates the power demand weight corresponding to the target component.
[0038] The power supply management device can detect the operating parameters of each component in the target AI device in real time, and record the operating parameter table in the database, or generate a corresponding log file according to the operating parameters. Therefore, the power supply management device can obtain the operating parameters corresponding to the target component from the database or log file. The operating parameters may include operating load, operating temperature, power consumption, etc.
[0039] Since different operating parameters have different dimensions and ranges, it is necessary to further standardize the operating parameters so that the operating parameters have the same comparison basis. Specifically, the power supply management device can use the Z-score algorithm to standardize the operating parameters to obtain a first standard score corresponding to the operating load, a second standard score corresponding to the operating temperature, and a third standard score for power consumption.
[0040] The first standard score corresponding to the operating load is determined by the following formula: Among them, Z1 is the first standard score, X1 is the operating load, μ1 is the load mean corresponding to the historical operating load, σ1 is the first standard deviation of the operating load, and the first standard deviation refers to the degree of discreteness of the load data set composed of the operating load and the historical operating load.
[0041] The second standard score corresponding to the operating temperature is determined by the following formula: Among them, Z2 is the second standard score, X2 is the operating temperature, μ2 is the temperature mean corresponding to the historical operating temperature, σ2 is the second standard deviation of the operating temperature, and the second standard deviation refers to the degree of discreteness of the temperature data set composed of the operating temperature and the historical operating temperature.
[0042] The third standard score for power consumption is determined by the following formula: Among them, Z3 is the third standard score, X3 is the power consumption, μ3 is the mean power consumption corresponding to the historical power consumption, σ3 is the third standard deviation of the power consumption, and the third standard deviation refers to the degree of discreteness of the power consumption data set composed of power consumption and historical power consumption.
[0043] After determining the first standard score, the second standard score and the third standard score, the power supply management device can assign weights to the operating load, operating temperature and power consumption in the operating parameters based on the Analytic Hierarchy Process (AHP), and the weight corresponding to the operating load is called the first weight factor, the weight corresponding to the operating temperature is called the second weight factor, and the weight corresponding to the power consumption is called the third weight factor. Specifically, step 1, establish a hierarchical model, first determine the target, the target is to determine the power demand weight; secondly, divide the levels into the target layer and the criterion layer, respectively, the target layer is to determine the power demand weight, and the criterion layer includes three factors: operating load, operating temperature, and power consumption. For the criterion layer, a 3×3 judgment matrix is constructed to determine the relative importance of the operating load, operating temperature, and power consumption, and the operating load, operating temperature, and power consumption are compared in pairs, and the weight vector of the judgment matrix is calculated according to the eigenvector. Among them, the weight vector represents the relative importance of each factor relative to other factors. Therefore, the calculated weight vector can be called the weight of each factor. For example, assuming that the calculation is performed according to the above steps, the obtained weight vector is [0.5, 0.3, 0.2], which means that among the three factors of operating load, operating temperature and power consumption, the weight of operating load is 0.5, the weight of operating temperature is 0.3, and the weight of power consumption is 0.2.
[0044] Once the standard score and weight factor for each of the operating parameters are determined, the power demand weight of the target component is determined by multiplying the standard score of each factor by its weight factor and then adding the results for all operating parameters.
[0045] For component j, the power demand weight W corresponding to component j can be determined by the following formula: j : Among them, z ij is the standard score of component j, w i is the weight factor of component j, and n is the number of parameters of the operating parameters.
[0046] In the embodiment of the present application, when determining the first standard score and the first weight factor corresponding to the operating load, the second standard score and the second weight factor corresponding to the operating temperature, and the third standard score and the third weight factor corresponding to the power consumption, the power demand weight of the target component is determined according to the first standard score, the second standard score, the third standard score, the first weight factor, the second weight factor, and the third weight factor. Specifically, the power demand weight = the first standard score × the first weight factor + the second standard score × the second weight factor + the third standard score × the third weight factor.
[0047] When determining the power demand weight corresponding to the target component, according to the same embodiment, the power supply management device may determine the power demand weight corresponding to each component.
[0048] Through the above optional implementation, the Z-score algorithm is used for standardization, and the hierarchical analysis method is used for weight allocation, so that the power supply management device can accurately evaluate the power demand weights of each component in the target AI device, thereby more effectively allocating and managing power resources. Secondly, by considering parameters such as the operating load, operating temperature, and power consumption of different components, and weighing them according to the relative importance of the operating parameters, it helps to optimize the performance and efficiency of the power supply management system, improve the system's adaptability to changes in power demand, and thus improve the system's reliability and stability.
[0049] S24, controlling the backup power supply to supply power to the multiple components according to the power demand weights.
[0050] After determining the power demand weight corresponding to each component, the power supply management device can accurately control the backup power supply to distribute power to each component according to the power demand weight.
[0051] In an optional implementation, controlling the backup power supply to supply power to the multiple components according to the power demand weights includes: Determine a correlation index between each of the multiple components and the business goal of the target AI device, and determine an importance level corresponding to each component according to the correlation index; Prioritizing the multiple components according to the importance levels and the power demand weights to obtain a power supply priority corresponding to each component; An operating state of the backup power supply is determined, and the backup power supply is controlled to supply power to the plurality of components according to the operating state and the power supply priority.
[0052] In some embodiments, the power supply management device first needs to clarify the business objectives of the target AI device, such as providing efficient data processing, supporting real-time decision-making, etc. Then, by analyzing the functions of each component for the target AI device, such as data processing, storage, communication, etc., and determining the functional dependencies between each component, identify which components are the basis for the normal operation of other components, and finally simulate component failure conditions to analyze the impact of components on the target AI device and business objectives, so as to identify which components are directly or indirectly related to the business objectives, that is, determine the correlation index of each component related to the business objectives of the target AI device. For example, for the core processing unit of an AI device, its correlation index is higher, because the normal operation of this component directly determines whether the AI device can achieve its business objectives.
[0053] Furthermore, the importance level of each component corresponding to the target AI device is evaluated according to the relevance index. The higher the relevance index, the higher its importance level. For example, the components are divided into different importance levels, where the importance level can be discrete, for example, high, medium, and low. The importance level can also be continuous, for example, 1 to 10, to distinguish the priority of different components in power supply. Then, according to the importance level and power demand weight of each component, priority is sorted. Specifically, components with high importance levels and large power demand weights are placed in front, and components with low importance levels and small power demand weights are placed in the back.
[0054] In addition, the power supply management device also needs to monitor the operating status of the backup power supply in real time, including key parameters such as voltage, current, remaining power, temperature, etc. According to the operating status of the backup power supply, the current power supply capacity and health status of the backup power supply are evaluated, so that the power supply management device can control the backup power supply to supply power to each of the multiple components according to its own operating status and the power supply priority corresponding to each component. Among them, according to the priority order, the power supply of the backup power supply first meets the power demand of the component with the highest priority.
[0055] In some embodiments, the maximum power that the backup power supply can provide for the i-th component at the current moment is determined according to the operating state of the backup power supply. Specifically, the total capacity of the backup power supply is determined from the factory specification of the backup power supply, and the maximum power S that can be provided for the i-th component at the current moment is calculated based on the total capacity and operating state of the backup power supply. i The maximum power S is determined by the following formula: i : Si =(Total_Capacity-Current_Load)×Adjustment_Factor In the calculation, Total_Capacity is the total capacity of the backup power supply, Current_Load is the total load size of each component at the current moment, and Adjustment_Factor is the adjustment factor obtained based on the operating status of the backup power supply. Among them, the adjustment factor is set according to the actual application situation. For example, if the power is lower than a certain threshold (such as 20%), the adjustment factor can be set to a smaller value to limit the power supply. If the temperature is too high, the adjustment factor can also be reduced. If the backup power supply is in good health, the adjustment factor can be close to 1.
[0056] Furthermore, the power supply management device can determine the target power demand E of the i-th component by the following calculation formula: i : Among them, P i is the power supply priority of the i-th component, S i is the maximum power provided by the backup power supply to the i-th component at the current moment, ∑P j The power supply priority of each component in the plurality of components is a sum of the power supply priorities of each component in the plurality of components.
[0057] After determining the target power demand of the i-th component, the power supply management device can control the backup power supply to supply power to the i-th component according to the target power demand corresponding to the i-th component. Therefore, according to the same embodiment, the power supply management device can determine the target power demand corresponding to each component and realize power supply for each of the multiple components.
[0058] Through the above optional implementation, by weighing the business goal relevance and power demand weight of each component, the power demand of the component that has the greatest impact on the business goal is given priority, thereby improving the stability and performance of the power supply system. Secondly, the target power demand is determined according to the priority and power demand weight of the component, realizing personalized and precise power supply for each component, maximizing the working requirements of the power supply system and improving the overall energy efficiency. That is, while meeting the needs of each component, it can give priority to components with higher importance levels and power demand weights, thereby achieving a balance in power supply.
[0059] In some embodiments, after the backup power supply is controlled in real time by the power supply management device to provide power to each component, feedback data in the power distribution process can be collected in real time, such as the remaining power of the backup power supply, the power consumption and operating status of each component, etc., so that the power supply management device can adjust the power demand weight corresponding to each component according to the feedback data to improve power utilization efficiency and the overall performance of the AI device. By dynamically adjusting the power demand weights of each component, power resources can be allocated more accurately and power utilization efficiency can be improved.
[0060] In an optional embodiment, the method further comprises: When the number of power supplies of the backup power supply is greater than 3, when it is detected that the main power supply is in an abnormal state, the power supply source is switched from the main power supply to the first-level backup power supply in the backup power supply; When the first-level backup power supply is in an abnormal state, switching the power supply source from the first-level backup power supply to the second-level backup power supply; When the second-level backup power source is in an abnormal state, the power supply source is switched from the second-level backup power source to a third-level backup power source until each of the backup power sources is in an abnormal state.
[0061] In some embodiments, when the main power supply is in an abnormal state, when it is necessary to switch the power supply source of the target AI device from the main power supply to the backup power supply, the number of power supplies of the backup power supply can be determined so that the power supply management device can switch the power supply source of the target AI device from the main power supply to which of the backup power supplies. Specifically, when the number of power supplies of the backup power supply is 2, when the power supply management device determines to switch the power supply source of the target AI device from the main power supply to the first-level backup power supply, it is necessary to detect the operating status of the first-level backup power supply in real time to determine whether the first-level backup power supply is in an abnormal state. When the first-level backup power supply is in an abnormal state, the power supply management device can switch the power supply source of the target AI device from the first-level backup power supply to the second-level backup power supply, and detect the operating status of the second-level backup power supply in real time to determine whether the second-level backup power supply is in an abnormal state. When the second-level backup power supply is in an abnormal state, the power supply management device records the detailed information of each abnormal backup power supply (e.g., timestamp, abnormal type, abnormal level, etc.), and reports to the administrator or operator through some communication channels (e.g., monitoring interface, alarm system, email, etc.). Secondly, for abnormalities that can be handled by itself (such as temporary power fluctuations or short circuits), the power supply management device can try to automatically retry or restore the backup power supply. For example, restart the backup power supply or switch to a different power line.
[0062] When the number of backup power sources is 3, when the second-level backup power source is in an abnormal state, the power supply management device can switch the power supply source of the target AI device from the second-level backup power source to the third-level backup power source, and detect the operating state of the third-level backup power source in real time to determine whether the third-level backup power source is in an abnormal state. When the third-level backup power source is in an abnormal state, it is processed according to the same embodiment as above.
[0063] When the number of backup power sources is 4, when the third-level backup power source is in an abnormal state, the power supply management device may switch the power supply source of the target AI device from the third-level backup power source to the fourth-level backup power source, and detect the operating state of the fourth-level backup power source in real time to determine whether the fourth-level backup power source is in an abnormal state. This process is repeated until each of the backup power sources is in an abnormal state.
[0064] Through the above optional implementation, a multi-level backup power supply, that is, a redundant backup power supply, is set up to improve the stability and reliability of the power supply system. Even if one of the levels of the backup power supply is abnormal, the system can still maintain power supply. Secondly, the power supply management device monitors the status of the backup power supply in real time. Once an abnormality is detected, it will automatically switch immediately to ensure the continuous power supply of the AI device, reduce the need for manual intervention, and effectively improve the stability, reliability and automation of the power supply system, providing an efficient and reliable power supply management solution for the target AI device.
[0065] In an optional embodiment, the method further comprises: Acquire the power status of the target AI device according to a preset time period; Predicting the target AI device according to the power status, and outputting a power problem of the target AI device; A processing strategy is formulated according to the power problem, and the power problem and the processing strategy are sent to a preset terminal.
[0066] In some embodiments, the power supply management device sets a sliding window to detect the power status of the target AI device according to a preset time period, such as operating parameters such as voltage, current, power, temperature, etc. The power status is input into a prediction model pre-trained based on a machine learning algorithm (e.g., LSTM or RNN, etc.) to predict the power problems that may occur in the target AI device within the next preset time week, such as power drop rate, voltage fluctuation, battery life exhaustion, and power failure. According to the predicted power problems, the power supply management device can formulate corresponding processing strategies. For example, for power problems with insufficient power, preventive measures such as charging or replacing batteries can be taken in advance. For power problems with power failures, measures can be taken in advance to repair or prepare to replace the device. In addition, the power supply management device can also use visualization tools to display power problems and their corresponding processing strategies, and send power problems and their corresponding processing strategies to a preset terminal (e.g., a user or administrator's mobile phone) so that the user or administrator can know the power status of the target AI device and how to deal with the power problem in real time.
[0067] Reference Figure 3 , which is a functional module diagram of a power supply management device shown in an embodiment of the present application.
[0068] In some embodiments, the power supply management device 30 may include a plurality of functional modules composed of computer program segments. The computer programs of the various program segments of the power supply management device 30 may be stored in a memory of an electronic device and executed by at least one processor to perform (see Figure 2 Description) Functionality of power supply management.
[0069] In this embodiment, the power supply management device 30 can be divided into multiple functional modules according to the functions it performs. The functional modules may include: a first detection switching module 301, a second detection switching module 302, a determination module 303, a control module 304, a switching module 305 and a prediction module 306. The module referred to in this application refers to a series of computer program segments that can be executed by at least one processor and can complete fixed functions, which are stored in a memory. In this embodiment, the functions of each module will be described in detail in subsequent embodiments.
[0070] The first detection switching module 301 is used to switch the power supply source of the target AI device from the mains to the main power supply when it is detected that the mains is in an abnormal state.
[0071] The second detection switching module 302 is used to switch the power supply source from the main power supply to the backup power supply when it is detected that the main power supply is in an abnormal state.
[0072] The determination module 303 is used to determine the power demand weight of each component of the multiple components of the target AI device when it is detected that the backup power supply is in a power shortage state.
[0073] The determination module 303 is also specifically used to: obtain operating parameters of a target component, wherein the target component is any one of the multiple components, and the operating parameters include operating load, operating temperature and power consumption; standardize the operating load, the operating temperature and the power consumption to obtain a first standard score corresponding to the operating load, a second standard score corresponding to the operating temperature and a third standard score corresponding to the power consumption; determine a first weight factor corresponding to the operating load, a second weight factor corresponding to the operating temperature and a third weight factor corresponding to the power consumption; determine the power demand weight of the target component based on the first standard score, the second standard score, the third standard score, the first weight factor, the second weight factor and the third weight factor.
[0074] The control module 304 is used to control the backup power supply to supply power to the multiple components according to the power demand weights.
[0075] The control module 304 is further specifically used to: determine a correlation index between each of the multiple components and the business objectives of the target AI device, and determine the importance level corresponding to each component based on the correlation index; prioritize the multiple components according to the importance level and the power demand weight to obtain the power supply priority corresponding to each component; determine the operating status of the backup power supply, and control the backup power supply to supply power to the multiple components based on the operating status and the power supply priority.
[0076] The control module 304 is further specifically used for: The target power requirement of the i-th component among the multiple components is determined by the following calculation formula: Among them, E i is the target power demand, P i is the power supply priority of the i-th component, S i is the maximum power provided by the backup power supply to the i-th component at the current moment, ∑P j a sum of power supply priorities for each of the plurality of components; The backup power supply is controlled to supply power to the i-th component according to the target power demand.
[0077] The switching module 305 is used to: when the number of power supplies of the backup power supply is greater than 3, when it is detected that the main power supply is in an abnormal state, switch the power supply source from the main power supply to the first-level backup power supply among the backup power supplies; when the first-level backup power supply is in an abnormal state, switch the power supply source from the first-level backup power supply to the second-level backup power supply; when the second-level backup power supply is in an abnormal state, switch the power supply source from the second-level backup power supply to the third-level backup power supply, until each of the backup power supplies is in an abnormal state.
[0078] The prediction module 306 is used to: obtain the power status of the target AI device according to a preset time period; predict the target AI device according to the power status and output the power problem of the target AI device; formulate a processing strategy according to the power problem, and send the power problem and the processing strategy to a preset terminal.
[0079] It should be understood that the various variations and specific embodiments of the power supply management method provided in the above embodiments are also applicable to the power supply management device of this embodiment. Through the above detailed description of the power supply management method, those skilled in the art can clearly know the implementation method of the power supply management device in this embodiment. For the sake of brevity of the specification, it will not be described in detail here.
[0080] An embodiment of the present application also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, all or part of the steps of the power supply management method are implemented.
[0081] Reference Figure 4 , which is a schematic diagram of the structure of a power supply control device according to an embodiment of the present application. In a preferred embodiment of the present application, the power supply control device 12 includes a memory 121 , at least one processor 122 and at least one communication bus 123 .
[0082] Those skilled in the art should understand that Figure 4 The structure of the power supply control device shown does not constitute a limitation of the embodiments of the present application, and can be either a bus structure or a star structure. The power supply control device 12 can also include more or less other hardware or software than shown in the figure, or a different component arrangement.
[0083] In some embodiments, the power supply control device 12 is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application-specific integrated circuits, programmable gate arrays, digital processors, and embedded devices. The power supply control device 12 may also include user equipment, which includes but is not limited to any electronic product that can interact with a user through a keyboard, mouse, remote control, touchpad, or voice control device, such as a personal computer, tablet computer, smart phone, digital camera, etc.
[0084] It should be noted that the power supply control device 12 is only an example, and other existing or future electronic products that are suitable for the present application should also be included in the protection scope of the present application and included here by reference.
[0085] In some embodiments, the memory 121 stores a computer program, and when the computer program is executed by the at least one processor 122, all or part of the steps in the power supply management method described above are implemented. The memory 121 includes a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), a one-time programmable read-only memory (OTPROM), an electronically erasable rewritable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc memory, disk memory, tape memory, or any other computer-readable medium that can be used to carry or store data. Further, the computer-readable storage medium may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function, etc.; the data storage area may store data created according to the use of the blockchain node, etc. The blockchain referred to in this application is a new application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanism, encryption algorithm, etc. Blockchain is essentially a decentralized database, a string of data blocks generated by cryptographic methods. Each data block contains a batch of network transaction information, which is used to verify the validity of its information (anti-counterfeiting) and generate the next block. Blockchain can include the underlying blockchain platform, platform product service layer, and application service layer.
[0086] In some embodiments, the at least one processor 122 is the control core (Control Unit) of the power supply control device 12, and uses various interfaces and lines to connect the various components of the entire power supply control device 12, and executes or executes the program or module stored in the memory 121, and calls the data stored in the memory 121 to execute various functions and process data of the power supply control device 12. For example, when the at least one processor 122 executes the computer program stored in the memory, it implements all or part of the steps of the power supply management method described above in the embodiment of the present application; or implements all or part of the functions of the power supply management device. The at least one processor 122 can be composed of an integrated circuit, for example, it can be composed of a single packaged integrated circuit, or it can be composed of multiple integrated circuits with the same function or different functions, including one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. Combination, etc.
[0087] In some embodiments, the at least one communication bus 123 is configured to realize connection and communication between the memory 121 and the at least one processor 122, etc. Although not shown, the power supply control device 12 may also include a power supply (such as a battery) for powering each component. Preferably, the power supply may be logically connected to the at least one processor 122 through a power management device, so that the power management device can realize functions such as management of charging, discharging, and power consumption management. The power supply may also include any components such as one or more DC or AC power supplies, recharging devices, power failure detection circuits, power converters or inverters, power status indicators, etc. The power supply control device 12 may also include a variety of sensors, Bluetooth modules, Wi-Fi modules, etc., which are not repeated here.
[0088] The above-mentioned integrated unit implemented in the form of a software function module can be stored in a computer-readable storage medium. The above-mentioned software function module is stored in a storage medium, including a number of instructions for enabling an electronic device (which can be a personal computer, electronic device, or network device, etc.) or a processor to execute part of the method described in each embodiment of the present application.
[0089] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only illustrative, for example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.
[0090] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0091] The above are all preferred embodiments of the present application, and the protection scope of the present application is not limited thereto. Therefore, any equivalent changes made according to the structure, shape, and principle of the present application should be included in the protection scope of the present application.
Claims
1. A power supply management method, applied to a power supply management system, characterized in that: The power supply management system comprises a parent-child uninterruptible power supply system, wherein the parent-child uninterruptible power supply system comprises a main power supply and at least one backup power supply, and the power supply management method comprises: When detecting that the mains power is in an abnormal state, switching the power supply source of the target AI device from the mains power to the main power supply; When it is detected that the main power supply is in an abnormal state, the power supply source is switched from the main power supply to the backup power supply; when it is detected that the backup power supply is in a power shortage state, the power demand weight of each component of the multiple components of the target AI device is determined; The backup power supply is controlled to supply power to the plurality of components according to the power demand weights.
2. The power supply management method according to claim 1, characterized in that: Determining the power requirement weight of each component of the target AI device includes: Acquiring operating parameters of a target component, wherein the target component is any one of the multiple components, and the operating parameters include operating load, operating temperature, and power consumption; Standardizing the operating load, the operating temperature, and the power consumption to obtain a first standard score corresponding to the operating load, a second standard score corresponding to the operating temperature, and a third standard score corresponding to the power consumption; Determining a first weight factor corresponding to the operating load, a second weight factor corresponding to the operating temperature, and a third weight factor corresponding to the power consumption; The power demand weight of the target component is determined according to the first standard score, the second standard score, the third standard score, the first weight factor, the second weight factor, and the third weight factor.
3. The power supply management method according to claim 1, characterized in that: The controlling the backup power supply to supply power to the plurality of components according to the power demand weights comprises: Determine a correlation index between each of the multiple components and the business goal of the target AI device, and determine an importance level corresponding to each component according to the correlation index; Prioritizing the multiple components according to the importance levels and the power demand weights to obtain a power supply priority corresponding to each component; An operating state of the backup power supply is determined, and the backup power supply is controlled to supply power to the plurality of components according to the operating state and the power supply priority.
4. The power supply management method according to claim 3, characterized in that: The controlling the backup power supply to supply power to the plurality of components according to the operating state and the power supply priority comprises: The target power requirement of the i-th component among the multiple components is determined by the following calculation formula: Among them, E i is the target power demand, P i is the power supply priority of the i-th component, S i is the maximum power provided by the backup power supply to the i-th component at the current moment, ∑P j a sum of power supply priorities for each of the plurality of components; The backup power supply is controlled to supply power to the i-th component according to the target power demand.
5. The power supply management method according to any one of claims 1 to 4, characterized in that: The method further comprises: When the number of power supplies of the backup power supply is greater than 3, when it is detected that the main power supply is in an abnormal state, the power supply source is switched from the main power supply to the first-level backup power supply in the backup power supply; When the first-level backup power supply is in an abnormal state, switching the power supply source from the first-level backup power supply to the second-level backup power supply; When the second-level backup power source is in an abnormal state, the power supply source is switched from the second-level backup power source to a third-level backup power source until each of the backup power sources is in an abnormal state.
6. The power supply management method according to claim 5, characterized in that: The method further comprises: Acquire the power status of the target AI device according to a preset time period; Predicting the target AI device according to the power status, and outputting a power problem of the target AI device; A processing strategy is formulated according to the power problem, and the power problem and the processing strategy are sent to a preset terminal.
7. A power supply management device, arranged in a power supply management system, characterized in that: The power supply management system comprises a parent-child uninterruptible power supply system, wherein the parent-child uninterruptible power supply system comprises a main power supply and at least one backup power supply, and the power supply management device comprises: A first detection switching module, configured to switch the power supply source of the target AI device from the mains to the main power supply when detecting that the mains power is in an abnormal state; A second detection switching module, configured to switch the power supply source from the main power supply to the backup power supply when detecting that the main power supply is in an abnormal state; A determination module, configured to determine a power demand weight of each component of the target AI device when detecting that the backup power supply is in a power shortage state; A control module is used to control the backup power supply to supply power to the multiple components according to the power demand weights.
8. A power supply management system, characterized in that: The power supply management system includes a parent-child uninterruptible power supply system, and the parent-child uninterruptible power supply system includes: A primary power source and at least one backup power source; The main power supply is used to replace the mains power to provide power supply to the target AI device when it is detected that the mains power is in an abnormal state; The backup power supply is used to replace the main power supply to provide power supply to the target AI device when it is detected that the main power supply is in an abnormal state.
9. The power supply management system according to claim 8, characterized in that: The power supply management system also includes a power supply control device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, the steps of the power supply management method described in any one of claims 1 to 6 are implemented.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the power supply management method according to any one of claims 1 to 6 are implemented.
