A control method and device for a supercomputing center power supply system

Abstract

The application provides a control method and device for a power supply system of an intelligent computing center, relates to the field of power control, and solves the technical problem that providing a backup power supply for all loads of the intelligent computing center may result in less use of the backup power supply, complete power failure before the power is restored, stop of the core load of the intelligent computing center, and influence on efficiency. The method comprises the following steps: obtaining use data of loads of the intelligent computing center; controlling the power supply system to supply power to the loads based on the use data; obtaining real-time operation data of the power supply system; controlling power supply of the power supply system based on the real-time operation data; obtaining environment data of a power market; and dynamically adjusting power supply measures of the power supply system based on the environment data. The application is used in the control process of the power supply system.

Description

Technical Field

[0001] This application relates to the field of power control, and in particular to a control method and device for a power supply system of an intelligent computing center. Background Technology

[0002] A power supply system for an intelligent computing center refers to a complete infrastructure and solution specifically designed, constructed, and operated for intelligent computing centers. It provides a continuous, stable, reliable, and efficient supply of power to all internal IT equipment and auxiliary facilities. Its core purpose is to provide uninterrupted power for high-density, high-performance, and high-reliability intelligent computing loads while maximizing energy efficiency. Controlling the power supply system of an intelligent computing center can significantly improve power reliability and energy efficiency, reduce operating costs, and enhance system maintainability and scalability, while also meeting green computing and compliance requirements. With the rapid development of technologies such as artificial intelligence and big data, the power demand of intelligent computing centers will continue to grow, making intelligent power supply control a key technology for ensuring their stable operation and sustainable development.

[0003] In existing technologies, power generation control of the power supply system ensures that the generated power can guarantee the normal operation of the intelligent computing center's load. In the event of a sudden power outage, a backup power source is directly provided to supply power to all loads of the intelligent computing center, enabling them to operate normally. However, the existing technology provides backup power to all loads of the intelligent computing center, which may result in the backup power source being used for a short period of time. If a complete power outage occurs before power is restored, the core loads of the intelligent computing center will stop operating, affecting efficiency. Summary of the Invention

[0004] This application provides a control method and device for a power supply system of an intelligent computing center, which solves the technical problem that the existing technology provides backup power to all loads of the intelligent computing center, but the backup power may not be available for long. If a complete power outage occurs before the power is restored, the core loads of the intelligent computing center will stop operating, thus affecting efficiency.

[0005] To achieve the above objectives, this application adopts the following technical solution:

[0006] Firstly, a control method for a power supply system of an intelligent computing center is provided, including:

[0007] Obtain usage data of the intelligent computing center's load; whereby the usage data is used to characterize the utilization of the intelligent computing center's load.

[0008] Power is supplied to the load based on the use of data-controlled power supply systems;

[0009] Obtain real-time operating data of the power supply system;

[0010] Control the power supply of the power supply system based on real-time operating data;

[0011] Obtain environmental data for the electricity market; whereby the environmental data is used to characterize the environmental factors of the electricity market.

[0012] The power supply measures of the power supply system are dynamically adjusted based on environmental data.

[0013] Based on the above technical solutions, the control method for the power supply system of the intelligent computing center provided in this application supplies power to the load according to the acquired load usage data, which can prioritize supplying power to a portion of the load in the event of a sudden power outage; it controls the power supply according to the real-time operation data of the power supply system, which can promptly detect and handle problems when the power supply system malfunctions, ensuring the normal operation of the power supply system; and it adjusts the power supply measures according to the environmental data of the electricity market, which can adopt different power supply methods for different periods, which is conducive to saving the cost of the power supply system and improving the energy efficiency.

[0014] In conjunction with the first aspect above, in one possible implementation, the provision of power to the load based on a data-controlled power supply system includes:

[0015] Retrieve usage data of the intelligent computing center's load; the usage data includes: business importance score, usage frequency, energy consumption, and interruption tolerance; analyze the priority coefficient of the intelligent computing center's load based on the usage data;

[0016] The loads of the intelligent computing center are sorted in descending order of priority coefficient to obtain a power supply sorting table; power is then supplied to the loads of the intelligent computing center in sequence according to the power supply sorting table.

[0017] In conjunction with the first aspect above, in one possible implementation, the step of using the priority coefficient of the intelligent computing center load based on data analysis includes:

[0018] The business importance score, usage frequency, energy consumption, and outage tolerance in the usage data are labeled as YP, SP, NL, and ZR, respectively; using the formula... Calculate the priority coefficient YX of the intelligent computing center load;

[0019] Where SPmax represents the maximum usage frequency of the intelligent computing center load; BNL represents the standard energy consumption of the intelligent computing center load; NLmax and NLmin represent the maximum and minimum actual energy consumption of the intelligent computing load, respectively; ZPmax represents the maximum interruption tolerance of the intelligent computing center load; α represents the power function; and β is a proportionality coefficient greater than 0.

[0020] In conjunction with the first aspect above, in one possible implementation, controlling the power supply of the power supply system based on real-time operational data includes:

[0021] Retrieve real-time operating data of the power supply system; the real-time operating data includes: real-time voltage, real-time current and real-time temperature;

[0022] Real-time operating data is compared with corresponding fault thresholds. When any real-time operating data exceeds the fault threshold, an alarm signal is generated. The power supply system is processed and controlled based on the alarm signal. Otherwise, the operating data within a set time period is predicted. The fault thresholds include voltage threshold, current threshold, and temperature threshold.

[0023] In conjunction with the first aspect above, in one possible implementation, the method for obtaining the fault threshold includes:

[0024] Acquire historical operating data; the data type of historical operating data is the same as that of real-time operating data; extract the critical values ​​at the time of power supply system failure from the historical operating data;

[0025] The critical values ​​are sorted chronologically and divided into several data segments; the average value PJi of each data segment is calculated; and then the formula is used to... The baseline threshold of the power supply system is calculated; where Wi represents the weighting coefficient of the i-th average value; PJi represents the average value of the i-th data group; λ represents the aging attenuation coefficient; and A represents the aging acceleration.

[0026] Through formula The dynamic threshold of the power supply system is calculated; where η represents the margin coefficient; and FX represents the risk score of the power supply system. ; Indicates the baseline failure rate; Indicates the load impact coefficient; The aging effect coefficient is represented by Z1 = current load / rated load; Z2 = ln(cumulative operating hours / 10000); the baseline threshold and dynamic threshold are integrated into the fault threshold.

[0027] In conjunction with the first aspect above, in one possible implementation, the step of processing and controlling the power supply system based on the alarm signal includes:

[0028] Retrieve alarm signals; the alarm signals include: level one alarm or level two alarm; when the alarm signal is level one alarm, determine the standby time of the power supply system; when the time difference between the standby time and the current time is less than the time threshold, the power supply system is inspected during the standby time; otherwise, the power supply system is inspected when the time threshold is reached.

[0029] When the alarm signal is a level two alarm, the operation of the power supply system shall be stopped immediately and the power supply system shall be inspected and repaired; according to the priority of the intelligent computing center load, the energy storage power shall be used to supply power to the intelligent computing center load in sequence.

[0030] In conjunction with the first aspect above, in one possible implementation, the prediction of operational data within a set time period includes:

[0031] Extract real-time operational data within a specified time period; fit the real-time operational data within the specified time period into operational curves according to data type; integrate the data types and operational curves into an operational prediction sequence;

[0032] The prediction model is invoked; the prediction sequence is input into the prediction model to obtain the prediction curve within the set time period; the prediction curve is compared with the baseline threshold line respectively; if there is a value in the prediction curve that is greater than the baseline threshold line, an early warning signal is generated; the prediction model is built based on an artificial intelligence model.

[0033] In conjunction with the first aspect above, in one possible implementation, the operational prediction model is constructed based on an artificial intelligence model, including:

[0034] Select appropriate models and deep learning frameworks from the artificial intelligence library; build the models based on the deep learning frameworks to obtain the constructed models;

[0035] Obtain the standard dataset; the standard dataset includes standard input data consistent with the content attributes of the predicted sequence and standard output data consistent with the content attributes of the predicted curve.

[0036] The standard dataset is divided into a training set, a validation set, and a test set according to a preset ratio; the model is trained using the training set; the internal parameters of the model are adjusted using the validation set; and the trained model is tested using the test set to obtain test metrics.

[0037] If all test metrics are greater than the test threshold, the corresponding built model is marked as a running prediction model; otherwise, the running prediction model is rebuilt.

[0038] In conjunction with the first aspect above, in one possible implementation, the dynamic adjustment of the power supply measures of the power supply system based on environmental data includes:

[0039] Retrieve environmental data from the electricity market, including electricity prices and carbon prices; when electricity prices are in off-peak or peak conditions, utilize grid power to power the intelligent computing center load; and charge the energy storage of the power supply system.

[0040] When the electricity price is at parity and the carbon price is within the standard range, the power supply system is adjusted according to the adjustment measures in the processing database; when the electricity price is at its peak, the power supply system is adjusted according to the range of the carbon price.

[0041] It should be noted that the technical staff integrates the historical adjustments to obtain the final processing library; the technical staff can update the processing library according to the adjustments of the power supply system.

[0042] Secondly, a control device for a power supply system of an intelligent computing center is provided, comprising: a communication unit and a processing unit; the communication unit is used to acquire usage data of the intelligent computing center load, real-time operation data of the power supply system, and environmental data of the electricity market; wherein, the usage data is used to characterize the usage of the intelligent computing center load; and the environmental data is used to characterize environmental factors of the electricity market.

[0043] The processing unit is used to control the power supply system to supply power to the load based on usage data; to control the power supply of the power supply system based on real-time operating data; and to dynamically adjust the power supply measures of the power supply system based on environmental data.

[0044] Thirdly, this application provides a control device for a power supply system of an intelligent computing center, comprising: a processor and a storage medium; the storage medium includes instructions, and the processor is used to execute the instructions to implement the methods described in the first aspect and any possible implementation thereof. The control device for the power supply system of the intelligent computing center can be an electronic device or a chip within an electronic device.

[0045] Fourthly, this application provides a control system for a power supply system of an intelligent computing center, comprising: a data acquisition module and a processing module; wherein the data acquisition module is used to acquire usage data of the intelligent computing center load, real-time operation data of the power supply system, and environmental data of the electricity market; wherein the usage data is used to characterize the usage of the intelligent computing center load; the environmental data is used to characterize environmental factors of the electricity market; the processing module is used to control the power supply system to supply power to the load based on the usage data; to control the power supply of the power supply system based on the real-time operation data; and to dynamically adjust the power supply measures of the power supply system based on the environmental data.

[0046] Fifthly, this application provides a computer-readable storage medium storing instructions that, when executed on a control device of a power supply system for an intelligent computing center, cause the control device to perform the method described in the first aspect and any possible implementation thereof.

[0047] Sixthly, this application provides a computer program product containing instructions that, when the computer program product is run on the control device of the intelligent computing center power supply system, cause the control device of the intelligent computing center power supply system to perform the method described in the first aspect and any possible implementation thereof.

[0048] This application provides a control method and device for a power supply system of an intelligent computing center, which can comprehensively analyze the priority of the load based on the load usage data and supply power to the loads sequentially according to the load priority; it can comprehensively analyze the priority of the load from multiple perspectives, making the calculation of the priority coefficient more realistic, and is conducive to ensuring the operation of core loads and ensuring operational efficiency.

[0049] By analyzing historical operational data to determine baseline and dynamic thresholds, and by analyzing real-time operational data of the power supply system, optimal measures can be taken to maintain the power supply system while ensuring its normal operation. This ensures the operation of the intelligent computing center's load. Furthermore, by predicting operational data for a set time period, the power supply system can be maintained in advance based on the prediction results, which is beneficial to ensuring the continuity of the power supply system's operation.

[0050] It should be understood that the descriptions of technical features, technical solutions, beneficial effects, or similar language in this application do not imply that all features and advantages can be achieved in any single embodiment. Rather, it is understood that the description of a feature or beneficial effect means that a specific technical feature, technical solution, or beneficial effect is included in at least one embodiment. Therefore, the descriptions of technical features, technical solutions, or beneficial effects in this specification do not necessarily refer to the same embodiment. Furthermore, the technical features, technical solutions, and beneficial effects described in this embodiment can be combined in any suitable manner. Those skilled in the art will understand that embodiments can be implemented without one or more specific technical features, technical solutions, or beneficial effects of a particular embodiment. In other embodiments, additional technical features and beneficial effects may be identified in specific embodiments that do not embody all embodiments. Attached Figure Description

[0051] Figure 1 A system architecture diagram of a power supply system for a smart computing center provided in this application embodiment;

[0052] Figure 2 A flowchart illustrating a control method for a power supply system of a smart computing center, provided as an embodiment of this application;

[0053] Figure 3 A flowchart illustrating another control method for a power supply system of a smart computing center provided in an embodiment of this application;

[0054] Figure 4A flowchart illustrating another control method for a power supply system of a smart computing center provided in an embodiment of this application;

[0055] Figure 5 A schematic diagram of the structure of a control device for a power supply system of a smart computing center provided in an embodiment of this application;

[0056] Figure 6 This is a schematic diagram of the hardware structure of a control device for a power supply system of a smart computing center, provided in an embodiment of this application. Detailed Implementation

[0057] In the description of this application, unless otherwise stated, " / " means "or," for example, A / B can mean A or B. The "and / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, and B alone. Furthermore, "at least one" means one or more, and "multiple" means two or more. The terms "first," "second," etc., do not limit the quantity or order of execution, and "first," "second," etc., do not necessarily imply differences.

[0058] It should be noted that, in this application, the terms "exemplary" or "for example" are used to indicate that something is being described as an example, illustration, or illustration. Any embodiment or design described as "exemplary" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or design solutions. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0059] The control method for the power supply system of the intelligent computing center provided in this application embodiment can be applied to, for example... Figure 1 In the server system 100 shown, such as Figure 1 As shown, the server system includes: acquisition device 101 and processing device 102.

[0060] Among them, the data acquisition device 101 is used to acquire the usage data of the intelligent computing center load, the real-time operation data of the power supply system, and the environmental data of the electricity market.

[0061] The processing device 102 is used to control the power supply system to supply power to the load based on usage data; to control the power supply of the power supply system based on real-time operating data; and to dynamically adjust the power supply measures of the power supply system based on environmental data.

[0062] To address the technical problem in existing technologies where backup power is provided to all loads in a smart computing center, which may result in short backup power usage time and a complete power outage before power is restored, causing the core loads of the smart computing center to stop operating and affecting efficiency, this application provides a control method for a power supply system of a smart computing center. The method includes: acquiring usage data of the smart computing center loads; wherein the usage data is used to characterize the usage status of the smart computing center loads.

[0063] Power is supplied to the load based on the use of data-controlled power supply systems;

[0064] Obtain real-time operating data of the power supply system;

[0065] Control the power supply of the power supply system based on real-time operating data;

[0066] Obtain environmental data for the electricity market; whereby the environmental data is used to characterize the environmental factors of the electricity market.

[0067] Based on environmental data, the power supply measures of the power supply system are dynamically adjusted. Based on this, power is supplied to the load according to the acquired load usage data, and in the event of a sudden power outage, power can be prioritized to a portion of the load. Power supply is controlled according to the real-time operation data of the power supply system, and when the power supply system malfunctions, problems can be detected and dealt with in a timely manner to ensure the normal operation of the power supply system. Adjusting power supply measures according to the environmental data of the electricity market allows for different power supply methods to be adopted for different periods, which helps to save power supply system costs and improve energy efficiency.

[0068] like Figure 2 As shown in the embodiment of this application, the control method for the power supply system of the intelligent computing center includes:

[0069] S201. Obtain the usage data of the intelligent computing center load; control the power supply system to supply power to the load based on the usage data.

[0070] The data used is used to characterize the load of the intelligent computing center, including: business importance score, usage frequency, energy consumption, and outage tolerance.

[0071] In some implementations, business importance scores can be obtained based on expert ratings, usage frequency and energy consumption rates can be statistically analyzed based on historical data of the intelligent computing center's load, and interruption tolerance can be obtained through simulation experiments.

[0072] S202. Obtain real-time operating data of the power supply system; control the power supply of the power supply system based on the real-time operating data.

[0073] Among them, real-time operating data is used to characterize the operating parameters of the power supply system during operation, including: real-time voltage, real-time current, and real-time temperature.

[0074] In some implementations, real-time voltage, real-time current, and real-time temperature can all be acquired through corresponding data sensors. When setting up the power supply system, the corresponding data sensors are placed in the corresponding positions, and the operating data of the power supply system is collected by the data sensors at the set acquisition frequency, thereby obtaining the real-time operating data of the power supply system.

[0075] S203. Obtain environmental data of the electricity market; dynamically adjust the power supply measures of the power supply system based on the environmental data.

[0076] Among them, environmental data refers to environmental factors in the electricity market, including electricity prices and carbon prices.

[0077] In some implementations, environmental data from the electricity market is retrieved, including electricity prices and carbon prices. When electricity prices are at their lowest or highest levels, the grid power is used to power the intelligent computing center load and to charge the energy storage of the power supply system. When electricity prices are at parity and carbon prices are within the standard range, the power supply system is adjusted according to the adjustment measures in the processing database. When electricity prices are at their highest levels, the power supply of the power supply system is adjusted according to the range of carbon prices.

[0078] It should be noted that the technical personnel integrate the historical adjustments to obtain the final processing library; the technical personnel can update the processing library according to the adjustments of the power supply system.

[0079] It should be noted that electricity prices in the environmental data of the electricity market will affect the electricity costs of the intelligent computing center load. When all electricity is supplied from the grid and the electricity price increases, the electricity costs of the corresponding intelligent computing center load will also increase accordingly. Carbon prices in the environmental data of the electricity market will affect the power generation costs of the power supply system. When all power generation is supplied from the power supply system and the carbon price increases, the power generation costs of the corresponding power supply system will also increase. Therefore, it is very important to consider the environmental factors of the electricity market.

[0080] For example, when electricity prices are at off-peak or grid parity, refer to the following table:

[0081]

[0082] Table 1: Electricity Price Adjustment Measures

[0083] When the SOC is less than 50%, the generator will be used to generate electricity.

[0084] When the electricity price reaches its peak and exceeds 0.8, the power supply system will be adjusted according to the carbon price.

[0085] The carbon price adjustment strategy is as follows: In high carbon price scenarios (carbon price > 100 yuan / ton): power generation accounts for 20% or energy storage discharge is used for power supply; In low carbon price scenarios (carbon price < 50 yuan / ton): power generation accounts for 60%.

[0086] Based on the above technical solutions, the control method for a power supply system of a smart computing center provided in this application supplies power to the load according to the acquired load usage data, which can prioritize supplying power to a portion of the load in the event of a sudden power outage; it controls the power supply according to the real-time operation data of the power supply system, which can promptly detect and handle problems when the power supply system malfunctions, ensuring the normal operation of the power supply system; and it adjusts the power supply measures according to the environmental data of the electricity market, which can adopt different power supply methods for different periods, which is conducive to saving the cost of the power supply system and improving the energy efficiency.

[0087] In one possible implementation of the embodiments of this application, combined with Figure 2 ,like Figure 3 As shown, the above S201 can be implemented through the following S301, S302 and S303, which are explained in detail below:

[0088] S301. Retrieve the usage data of the intelligent computing center load; label the business importance score, usage frequency, energy consumption and interruption tolerance in the usage data as YP, SP, NL and ZR respectively.

[0089] The data used includes: business importance score, usage frequency, energy consumption, and outage tolerance.

[0090] S302, through formula Calculate the priority coefficient YX of the intelligent computing center load.

[0091] Where SPmax represents the maximum usage frequency of the intelligent computing center load; BNL represents the standard energy consumption of the intelligent computing center load; NLmax and NLmin represent the maximum and minimum actual energy consumption of the intelligent computing load, respectively; ZPmax represents the maximum interruption tolerance of the intelligent computing center load; α represents the power function; and β is a proportionality coefficient greater than 0.

[0092] For example, suppose the intelligent computing center has 3 loads. We retrieve their usage data (step S301) and label the variables: Load A: YP=80, SP=40, NL=150, ZR=80; Load B: YP=70, SP=60, NL=250, ZR=50; Load C: YP=85, SP=55, NL=180, ZR=90.

[0093] From this data, the global parameters are extracted as follows: SPmax=max(40, 60, 55)=60; NLmax=max(150, 250, 180)=250; NLmin=min(150, 250, 180)=150; ZRmax=max(80, 50, 90)=90; α=0.5; β=1.0; γ=1.0; The calculated priorities of the intelligent computing center loads are as follows: the priority of load A is 2.118; the priority of load B is 1.037; and the priority of load C is 2.035.

[0094] S303. Sort the intelligent computing center loads in descending order of priority coefficient to obtain a power supply sorting table; supply power to the intelligent computing center loads in sequence according to the power supply sorting table.

[0095] For example, the loads of the intelligent computing center are sorted according to the priority calculated in the above steps as follows: load A, load C, load B; power is supplied to loads A, C, and B in sequence; in the event of a power outage or power failure, power is supplied to load A first based on the amount of stored energy.

[0096] Based on the above technical solution, the priority of the load is comprehensively analyzed according to the load usage data, and the load is powered sequentially according to the load priority. This can comprehensively analyze the load priority from multiple perspectives, making the calculation of the priority coefficient more realistic, and is conducive to ensuring the operation of the core load and ensuring the efficiency of operation.

[0097] In one possible implementation of the embodiments of this application, combined with Figure 2 ,like Figure 4 As shown, the above S202 can be specifically implemented through the following S401, S402 and S403, which are explained in detail below:

[0098] S401. Retrieve real-time operating data of the power supply system; compare the real-time operating data with the corresponding fault thresholds; generate an alarm signal when any item of the real-time operating data exceeds the fault threshold; process and control the power supply system according to the alarm signal; otherwise, predict the operating data within the set time period.

[0099] The real-time operating data includes: real-time voltage, real-time current, and real-time temperature; the fault thresholds include: voltage threshold, current threshold, and temperature threshold.

[0100] In some implementations, the fault threshold is obtained in the following ways:

[0101] Acquire historical operating data; the data type of historical operating data is the same as that of real-time operating data; extract the critical values ​​at the time of power supply system failure from the historical operating data;

[0102] The critical values ​​are sorted chronologically and divided into several data segments; the average value PJi of each data segment is calculated; and then the formula is used to... The baseline threshold of the power supply system is calculated; where Wi represents the weighting coefficient of the i-th average value; PJi represents the average value of the i-th data group; λ represents the aging attenuation coefficient; and A represents the aging acceleration.

[0103] Through formula The dynamic threshold of the power supply system is calculated; where η represents the margin coefficient; and FX represents the risk score of the power supply system. ; Indicates the baseline failure rate; Indicates the load impact coefficient; The aging effect coefficient is represented by Z1 = current load / rated load; Z2 = ln(cumulative operating hours / 10000); the baseline threshold and dynamic threshold are integrated into the fault threshold.

[0104] It should be noted that the aging attenuation coefficient, aging acceleration, baseline failure rate, load influence coefficient, and aging influence coefficient are all set by technical personnel based on practical experience. Generally, the aging attenuation coefficient is set to 0.02, the aging acceleration is set to 15, the baseline failure rate is set to 0.1, the load influence coefficient is set to 1.2, and the aging influence coefficient is set to 0.8.

[0105] It should be noted that the method for integrating the baseline threshold and dynamic threshold into the fault threshold is as follows: the two are directly concatenated, fault threshold = [baseline threshold, dynamic threshold]; when alarm warning is issued, if the real-time operating data is greater than the baseline threshold, the generated alarm signal is a level one alarm signal; if the real-time operating data is greater than the dynamic threshold, the generated alarm signal is a level two alarm signal; when the dynamic threshold is greater than the baseline threshold, historical operating data is reselected, the dynamic threshold is recalculated, and the dynamic threshold is updated according to the set period; the baseline thresholds include: voltage baseline threshold, current baseline threshold, and temperature baseline threshold; the dynamic thresholds include: voltage dynamic threshold, current dynamic threshold, and temperature dynamic threshold.

[0106] For example, taking current as an example, let's analyze the fault threshold of real-time current; as shown in the table below:

[0107]

[0108] Table 2 Ranking of Current Critical Values

[0109] Segment 1 (early stage): 102, 105 → PJ1 = 103.5; Segment 2 (mid-stage): 108, 110 → PJ2 = 109.0; Segment 3 (recent stage): 115 → PJ3 = 115;

[0110] Weighting (recent weights are higher): W1=0.2, W2=0.3, W3=0.5 (satisfying ∑W_i=1); Aging parameters: λ=0.02, A=15; then the baseline threshold Y=82.13 is calculated using the formula.

[0111] Analysis of the dynamic threshold: Current load: 85kW; Rated load: 100kW → Z1=0.85; Cumulative operating hours: 45,000 hours → Z2=1.504; H0(t)=0.1; β1=1.2; β2=0.8; The dynamic threshold calculated by the formula is JY=74.90.

[0112] 402. Retrieve alarm signals; the alarm signals include: Level 1 alarm or Level 2 alarm; when the alarm signal is Level 1 alarm, determine the standby time of the power supply system. If the time difference between the standby time and the current time is less than the time threshold, the power supply system is inspected during the standby time; otherwise, the power supply system is inspected when the time threshold is reached; when the alarm signal is Level 2 alarm, immediately stop the operation of the power supply system and inspect the power supply system; according to the priority of the intelligent computing center load, use the energy storage power to supply power to the intelligent computing center load in sequence.

[0113] S403. Extract real-time operating data within a set time period; fit the real-time operating data within the set time period into operating curves according to data type; integrate the data type and operating curves into an operating prediction sequence; call the operating prediction model; input the operating prediction sequence into the operating prediction model to obtain the prediction curve within the set time period; compare the prediction curve with the baseline threshold line respectively; if there is a value in the prediction curve that is greater than the baseline threshold line, generate an early warning signal.

[0114] It should be noted that the predictive model is built upon an artificial intelligence model, including:

[0115] Select models and deep learning frameworks from the artificial intelligence library; build models based on deep learning frameworks to obtain the constructed models;

[0116] Obtain the standard dataset; the standard dataset includes standard input data consistent with the content attributes of the predicted sequence and standard output data consistent with the content attributes of the predicted curve.

[0117] The standard dataset is divided into a training set, a validation set, and a test set according to a preset ratio; the model is trained using the training set; the internal parameters of the model are adjusted using the validation set; and the trained model is tested using the test set to obtain test metrics.

[0118] If all test metrics are greater than the test threshold, the corresponding built model is marked as a running prediction model; otherwise, the running prediction model is rebuilt.

[0119] In some implementations, the model needs to be iteratively optimized during training until the model converges to a preset standard, indicating that the model training is complete. When building the model, it can be trained based on a known model framework, or it can be built according to requirements. The specific model is selected based on the actual situation.

[0120] It should be noted that the model's test metrics include: accuracy, stability, F1 score, and recall; the standard dataset is generally divided in an 8:1:1 ratio, and the metric thresholds are set based on historical experience; when the model needs to be retrained, the standard dataset's division ratio can be adjusted, or the model framework can be adjusted.

[0121] Based on the above technical solution, by analyzing the baseline and dynamic thresholds according to historical operating data and analyzing the real-time operating data of the power supply system, the optimal measures can be taken to maintain the power supply system while ensuring its normal operation, thus ensuring the operation of the intelligent computing center load. Furthermore, by predicting the operating data for a set time period, the power supply system can be maintained in advance based on the prediction results, which is conducive to ensuring the continuity of the power supply system operation.

[0122] The above primarily describes the solutions of the embodiments of this application from the perspective of device implementation. It is understood that each device, such as the control device of a power supply system for an intelligent computing center, includes at least one of the hardware structures and software modules corresponding to each function in order to achieve the above-mentioned functions. Those skilled in the art should readily recognize that, based on the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0123] This application embodiment can divide the control device of the intelligent computing center power supply system into functional units according to the above method example. For example, each function can be divided into a separate functional unit, or two or more functions can be integrated into one processing unit. The integrated unit can be implemented in hardware or software functional units. It should be noted that the unit division in this application embodiment is illustrative and only represents one logical functional division. In actual implementation, there may be other division methods.

[0124] When using integrated units, Figure 5 A possible structural schematic diagram of the control device (referred to as the control device 50 of the intelligent computing center power supply system) involved in the above embodiments is shown. The control device 50 of the intelligent computing center power supply system includes a processing unit 501 and a communication unit 502, and may also include a storage unit 503. Figure 5 The structural diagram shown can be used to illustrate the structure of the control device of the intelligent computing center power supply system involved in the above embodiments.

[0125] when Figure 5 The structural diagram shown is used to illustrate the structure of the control device of the intelligent computing center power supply system involved in the above embodiments. The processing unit 501 is used to control and manage the operation of the control device of the intelligent computing center power supply system, the communication unit 502 is used for the control device of the intelligent computing center power supply system to communicate with other devices, and the storage unit 503 is used to store the program code and data of the control device of the intelligent computing center power supply system.

[0126] For example, communication unit 502 is used to acquire usage data of the intelligent computing center load, real-time operation data of the power supply system, and environmental data of the electricity market; wherein, usage data is used to characterize the usage of the intelligent computing center load; and environmental data is used to characterize environmental factors of the electricity market.

[0127] The processing unit 501 is also used to control the power supply system to supply power to the load based on usage data; to control the power supply of the power supply system based on real-time operating data; and to dynamically adjust the power supply measures of the power supply system based on environmental data.

[0128] The processing unit 501 can be a processor or a controller, and the communication unit 502 can be a communication interface, transceiver, transceiver circuit, transceiver device, etc. The term "communication interface" is a general term and may include one or more interfaces. The storage unit 503 can be a memory. When the control device 50 of the intelligent computing center power supply system is a chip, the processing unit 501 can be a processor or a controller, and the communication unit 502 can be an input interface and / or an output interface, pins, or circuits, etc. The storage unit 503 can be a storage unit within the chip (e.g., a register, cache, etc.) or a storage unit located outside the chip (e.g., read-only memory (ROM), random access memory (RAM, etc.)).

[0129] The communication unit can also be called a transceiver unit. The antenna and control circuit with transceiver functions in the control device 50 of the intelligent computing center power supply system can be considered as the communication unit 502 of the control device 50, and the processor with processing functions can be considered as the processing unit 501 of the control device 50. Optionally, the device in the communication unit 502 used to implement the receiving function can be considered as the communication unit, which is used to execute the receiving steps in the embodiments of this application. The communication unit can be a receiver, a receiver circuit, etc. The device in the communication unit 502 used to implement the transmitting function can be considered as the transmitting unit, which is used to execute the transmitting steps in the embodiments of this application. The transmitting unit can be a transmitter, a transmitter, a transmitting circuit, etc.

[0130] Figure 5 If the integrated units in the process are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this application, in essence, or the parts that contribute to the prior art, or all or part of the technical solutions, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the methods described in the various embodiments of this application. Storage media for storing computer software products include various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory, random access memory, magnetic disks, or optical disks.

[0131] Figure 5 The units in the process can also be called modules; for example, a processing unit can be called a processing module.

[0132] This application embodiment also provides a hardware structure diagram of a control device for a power supply system of an intelligent computing center (referred to as the control device 60 of the power supply system of the intelligent computing center), see [link to relevant documentation]. Figure 6 The control device 60 of the power supply system of the intelligent computing center includes a processor 601, and optionally, a memory 602 connected to the processor 601.

[0133] In the first possible implementation, see Figure 6 The control device 60 of the intelligent computing center power supply system also includes a transceiver 603. The processor 601, memory 602, and transceiver 603 are connected via a bus. The transceiver 603 is used to communicate with other devices or communication networks. Optionally, the transceiver 603 may include a transmitter and a receiver. The device in the transceiver 603 that implements the receiving function can be considered as a receiver, which is used to perform the receiving steps in the embodiments of this application. The device in the transceiver 603 that implements the transmitting function can be considered as a transmitter, which is used to perform the transmitting steps in the embodiments of this application.

[0134] Based on the first possible implementation method Figure 6 The structural diagram shown can be used to illustrate the structure of the control device of the intelligent computing center power supply system involved in the above embodiments.

[0135] in, Figure 6 The diagram can also illustrate the system chip in the control device of the power supply system for the intelligent computing center. In this case, the actions performed by the control device of the power supply system for the intelligent computing center can be implemented by the system chip. For details of the actions performed, please refer to the above text, which will not be repeated here.

[0136] In implementation, each step of the method provided in this embodiment can be completed by integrated logic circuits in the processor or by instructions in software form. The steps of the method disclosed in the embodiments of this application can be directly manifested as being executed by a hardware processor, or being executed by a combination of hardware and software modules in the processor.

[0137] The processor in this application may include, but is not limited to, at least one of the following: a central processing unit (CPU), a microprocessor, a digital signal processor (DSP), a microcontroller unit (MCU), or an artificial intelligence processor, etc., which are various computing devices that run software. Each computing device may include one or more cores for executing software instructions to perform calculations or processing. The processor may be a separate semiconductor chip or integrated with other circuits into a single semiconductor chip. For example, it may be integrated with other circuits (such as encoding / decoding circuits, hardware acceleration circuits, or various bus and interface circuits) to form a SoC (System-on-a-Chip), or it may be integrated as a built-in processor within an ASIC. The ASIC with the integrated processor may be packaged separately or together with other circuits. In addition to the cores for executing software instructions to perform calculations or processing, the processor may further include necessary hardware accelerators, such as field-programmable gate arrays (FPGAs), PLDs (programmable logic devices), or logic circuits that implement dedicated logic operations.

[0138] The memory in the embodiments of this application may include at least one of the following types: read-only memory (ROM) or other types of static storage devices capable of storing static information and instructions; random access memory (RAM) or other types of dynamic storage devices capable of storing information and instructions; or electrically erasable programmable-only memory (EEPROM). In some scenarios, the memory may also be a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media, or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures that can be accessed by a computer, but is not limited thereto.

[0139] This application also provides a computer-readable storage medium including instructions that, when run on a computer, cause the computer to perform any of the methods described above.

[0140] This application also provides a computer program product containing instructions that, when run on a computer, cause the computer to perform any of the methods described above.

[0141] This application also provides a chip including a processor and an interface circuit. The interface circuit is coupled to the processor. The processor is used to run computer programs or instructions to implement the above-described method. The interface circuit is used to communicate with other modules outside the chip.

[0142] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented using software programs, implementation can be, in whole or in part, in the form of a computer program product. This computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device containing one or more servers, data centers, etc., that can be integrated with the medium. The available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid-state disks (SSDs)).

Claims

1. A control method for a power supply system of an intelligent computing center, characterized in that, include: Obtain usage data of the intelligent computing center's load; whereby the usage data is used to characterize the utilization of the intelligent computing center's load. Power is supplied to the load based on the use of data-controlled power supply systems; Obtain real-time operating data of the power supply system; Control the power supply of the power supply system based on real-time operating data; Obtain environmental data for the electricity market; whereby the environmental data is used to characterize the environmental factors of the electricity market. Dynamically adjust power supply measures of the power supply system based on environmental data; The method of supplying power to the load based on the data-controlled power supply system includes: Retrieve usage data of the intelligent computing center's load; the usage data includes: business importance score, usage frequency, energy consumption, and interruption tolerance; analyze the priority coefficient of the intelligent computing center's load based on the usage data; The loads of the intelligent computing center are sorted in descending order of priority coefficient to obtain a power supply sorting table; power is supplied to the loads of the intelligent computing center in sequence according to the power supply sorting table. The control of power supply to the power supply system based on real-time operating data includes: Retrieve real-time operating data of the power supply system; the real-time operating data includes: real-time voltage, real-time current and real-time temperature; Real-time operating data is compared with corresponding fault thresholds; when any real-time operating data exceeds the fault threshold, an alarm signal is generated; the power supply system is processed and controlled based on the alarm signal; otherwise, the operating data within a set time period is predicted; the fault thresholds include: voltage threshold, current threshold, and temperature threshold. The process of controlling the power supply system based on the alarm signal includes: Retrieve alarm signals; the alarm signals include: level one alarm or level two alarm; when the alarm signal is level one alarm, determine the standby time of the power supply system; when the time difference between the standby time and the current time is less than the time threshold, the power supply system is inspected during the standby time; otherwise, the power supply system is inspected when the time threshold is reached. When the alarm signal is a level two alarm, the operation of the power supply system shall be stopped immediately and the power supply system shall be inspected and repaired; according to the priority of the intelligent computing center load, the energy storage power shall be used to supply power to the intelligent computing center load in sequence. The prediction of operational data within a set time period includes: Extract real-time operational data within a specified time period; fit the real-time operational data within the specified time period into operational curves according to data type; integrate the data types and operational curves into an operational prediction sequence; The prediction model is invoked; the prediction sequence is input into the prediction model to obtain the prediction curve within the set time period; the prediction curve is compared with the baseline threshold line respectively; if there is a value in the prediction curve that is greater than the baseline threshold line, an early warning signal is generated; the prediction model is built based on an artificial intelligence model.

2. The control method for a power supply system of an intelligent computing center according to claim 1, characterized in that, The priority coefficient for analyzing the load of the intelligent computing center based on usage data includes: The business importance score, usage frequency, energy consumption, and outage tolerance in the usage data are labeled as YP, SP, NL, and ZR, respectively; using the formula... Calculate the priority coefficient YX of the intelligent computing center load; Where SPmax represents the maximum usage frequency of the intelligent computing center load; BNL represents the standard energy consumption of the intelligent computing center load; NLmax and NLmin represent the maximum and minimum actual energy consumption of the intelligent computing load, respectively; ZRmax represents the maximum interruption tolerance of the intelligent computing center load; α represents the power function; and β is a proportionality coefficient greater than 0.

3. The control method for a power supply system of an intelligent computing center according to claim 1, characterized in that, The method for obtaining the fault threshold includes: Acquire historical operating data; the data type of historical operating data is the same as that of real-time operating data; extract the critical values ​​at the time of power supply system failure from the historical operating data; The critical values ​​are sorted chronologically and divided into several data segments; the average value PJi of each data segment is calculated; and then the formula is used to... The baseline threshold of the power supply system is calculated; where Wi represents the weighting coefficient of the i-th average value; PJi represents the average value of the i-th data group; λ represents the aging attenuation coefficient; and A represents the aging acceleration. Through formula The dynamic threshold of the power supply system is calculated; where η represents the margin coefficient; and FX represents the risk score of the power supply system. ; Indicates the baseline failure rate; Indicates the load impact coefficient; The aging effect coefficient is represented by Z1 = current load / rated load; Z2 = ln(cumulative operating hours / 10000); the baseline threshold and dynamic threshold are integrated into the fault threshold.

4. The control method for a power supply system of an intelligent computing center according to claim 1, characterized in that, The operational prediction model is built based on an artificial intelligence model and includes: Select models and deep learning frameworks from the artificial intelligence library; build models based on deep learning frameworks to obtain the constructed models; Obtain the standard dataset; the standard dataset includes standard input data consistent with the content attributes of the predicted sequence and standard output data consistent with the content attributes of the predicted curve. The standard dataset is divided into a training set, a validation set, and a test set according to a preset ratio; the training set is used to train the model; the validation set is used to adjust the internal parameters of the model; and the test set is used to test the trained model and obtain test metrics. If all test metrics are greater than the test threshold, the corresponding built model is marked as a running prediction model; otherwise, the running prediction model is rebuilt.

5. The control method for a power supply system of an intelligent computing center according to claim 1, characterized in that, The dynamic adjustment of power supply measures based on environmental data includes: Retrieve environmental data from the electricity market, including electricity prices and carbon prices; when electricity prices are in off-peak or peak conditions, utilize grid power to power the intelligent computing center load; and charge the energy storage of the power supply system. When the electricity price is at parity and the carbon price is within the standard range, the power supply system is adjusted according to the adjustment measures in the processing database; when the electricity price is at its peak, the power supply system is adjusted according to the range of the carbon price.

6. A control device for a power supply system of an intelligent computing center, applied to the control method for a power supply system of an intelligent computing center as described in any one of claims 1-5, characterized in that, include: Communication unit and processing unit; The communication unit is used to acquire usage data of the intelligent computing center load, real-time operation data of the power supply system, and environmental data of the electricity market; wherein, the usage data is used to characterize the usage of the intelligent computing center load; and the environmental data is used to characterize environmental factors of the electricity market. The processing unit is used to control the power supply system to supply power to the load based on the data used; The power supply system is controlled based on real-time operational data; the power supply measures of the power supply system are dynamically adjusted based on environmental data.

Images