Power supply control, power supply and computing power system and method for data center and related device

Abstract

The invention relates to the technical field of computers, in particular to a power supply control, power supply and computing power system, method and related device of a data center. The power supply control system of the data center comprises a prediction module and a control module. The prediction module is used for inputting a historical data combination and current environment parameters to predict predicted energy supply of the power supply device; wherein the historical data combination comprises historical environment parameters and historical energy supply data of the power supply device; and the control module comprises a prediction input end and a control end, and the prediction input end is connected with the prediction module and used for controlling the power supply device connected with the control end to be connected to the grid or disconnected from the grid based on the predicted energy supply and the required energy of the data center. Through the above mode, the energy supply amount of the power supply device associated with the data center can be predicted, so that reasonable planning of grid connection of the power supply device is facilitated, and the power generation utilization rate of the power supply device can be improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a power supply control system of a data center, a power supply system of a data center, a computing system, a power supply method of a data center, a computer device, and a computer-readable storage medium. Background Art

[0002] With the rapid development of artificial intelligence, the computing power requirements for computers in data centers have also increased, and the power consumption of computers is also huge. Therefore, the use of renewable energy to power computers has emerged. However, the power output of power generation methods such as photovoltaics, wind power, or conventional power generation methods is random and intermittent, making it difficult to maintain a stable power supply. Summary of the invention

[0003] Based on this, it is necessary to provide a data center power supply control system, a data center power supply system, a computing power system, a data center power supply method, computer equipment and a computer-readable storage medium to address the above-mentioned technical problems, which can predict the power supply of the power supply device associated with the data center, thereby facilitating the rational planning of the power supply device grid connection, and further improving the power generation utilization rate of the power supply device.

[0004] On the one hand, a power supply control system of a data center is provided, and the power supply control system of the data center includes: a prediction module and a control module; the prediction module is used to input a historical data combination and current environmental parameters to predict the predicted energy supply of a power supply device; wherein the historical data combination includes historical environmental parameters and historical energy supply data of the power supply device; the control module includes a prediction input terminal and a control terminal, and the prediction input terminal is connected to the prediction module, and is used to control the power supply device connected to the control terminal to be connected to the grid or off the grid based on the predicted energy supply and the energy demand of the data center.

[0005] In one embodiment of the present application, the control module also includes an energy storage input terminal, which is used to connect to the energy storage device to obtain current energy storage information of the energy storage device, so as to combine the current energy storage information, the predicted energy supply and the required energy to determine whether to control the power supply device to be connected to the grid or off the grid.

[0006] In one embodiment of the present application, the control module includes an energy storage control unit, a first switch unit and a first conversion unit; the energy storage control unit is used to determine whether to control the energy storage device to store energy or supply energy based on the current energy storage information, the predicted energy supply and the required energy, and output a charge and discharge signal when it is determined to control the energy storage device to store energy or supply energy; the first switch unit is connected to the energy storage control unit and the first conversion unit, and the first switch unit is also used to connect to the energy storage device to receive the charge and discharge signal to conduct the path between the energy storage device and the first conversion unit; the first conversion unit is also used to connect to the data center and the power supply device to perform power conversion on the electric energy flowing through the first conversion unit.

[0007] In one embodiment of the present application, the control module includes a power supply control unit, a second switch unit and a second conversion unit; the power supply control unit is used to control the power supply device to be connected to the grid and output a grid-connected signal; the second switch unit is connected to the power supply control unit and the second conversion unit, and the second switch unit is also used to connect to the power supply device to receive the grid-connected signal to conduct the path between the power supply device and the second conversion unit; the second conversion unit is also used to connect to the data center to perform power conversion on the output power supply of the power supply device and transmit it to the data center.

[0008] In one embodiment of the present application, the prediction module includes a plurality of prediction units; the plurality of prediction units are used to respectively predict the output energy supply of a plurality of power generation devices of the power supply device.

[0009] On the other hand, a power supply system for a data center is provided, and the power supply system of the data center includes: a power supply device and a power supply control system of the data center; the power supply device is used to input clean energy and convert the clean energy into electrical energy for output; the power supply control system of the data center is connected to the power supply device; the power supply control system includes a prediction module and a control module; the prediction module is used to input a historical data combination and current environmental parameters to predict the predicted energy supply of the power supply device; wherein the historical data combination includes historical environmental parameters and historical energy supply data of the power supply device; the control module includes a prediction input terminal and a control terminal, and the prediction input terminal is connected to the prediction module, and is used to control the power supply device connected to the control terminal to be connected to the grid or off the grid based on the predicted energy supply and the required energy of the data center.

[0010] In one embodiment of the present application, the power supply device includes a plurality of power generation devices.

[0011] In one embodiment of the present application, the power supply system also includes an energy storage device; the energy storage device is connected to the power supply control system; the energy storage device is used to send its current energy storage information to the power supply control system; the power supply control system is used to determine whether to control the energy storage device to store or supply energy based on the predicted energy supply, the current energy storage information and the required energy.

[0012] Another method provides a computing power system, which includes: a data center and a power supply control system of the data center; the power supply control system is connected to the data center; the power supply control system includes a prediction module and a control module; the prediction module is used to input a historical data combination and current environmental parameters to predict the predicted energy supply of the power supply device; wherein the historical data combination includes historical environmental parameters and historical energy supply data of the power supply device; the control module includes a prediction input terminal and a control terminal, the prediction input terminal is connected to the prediction module, and is used to control the power supply device connected to the control terminal to be connected to or off the grid based on the predicted energy supply and the energy demand of the data center.

[0013] In one embodiment of the present application, the computing power system also includes a power supply device; the power supply device is connected to the power supply control system, and is used to output energy when the power supply control system determines to control it to be connected to the grid, so that the data center receives the output energy.

[0014] On the other hand, a power supply method for a data center is provided, the power supply method comprising: obtaining a historical data combination and current environmental parameters; wherein the historical data combination comprises historical environmental parameters and historical energy supply data of a power supply device; using both the historical data combination and the current environmental parameters to predict the predicted energy supply of the power supply device; and based on the predicted energy supply and the energy demand of the data center, controlling whether the power supply device is connected to the grid.

[0015] In one embodiment of the present application, controlling whether the power supply device is connected to the grid based on the predicted energy supply and the required energy of the data center includes: acquiring the current stored energy of the energy storage device; fusing the current stored energy with the predicted energy supply as the predicted total energy; in response to the predicted total energy being higher than the required energy, controlling the power supply device to be connected to the grid to supply energy to the data center.

[0016] In one embodiment of the present application, controlling the power supply device to be connected to the grid to supply energy to the data center includes: comparing the predicted energy supply with the required energy; in response to the predicted energy supply being higher than the required energy, controlling the power supply device to supply energy to the data center and the energy storage device.

[0017] In one embodiment of the present application, controlling the power supply device to be connected to the grid to supply energy to the data center includes: comparing the predicted energy supply with the required energy; in response to the predicted energy supply being lower than the required energy, controlling the power supply device and the energy storage device to supply energy to the data center.

[0018] In one embodiment of the present application, the fusion of the current energy storage and the predicted energy supply as the predicted total energy also includes: in response to the predicted total energy being lower than the required energy, controlling the power supply device to go off-grid; and controlling the power supply device to supply energy to the energy storage device.

[0019] In one embodiment of the present application, environmental parameters include environmental factor parameters and meteorological parameters; wherein the environmental factor represents the environmental factor associated with the type of clean energy input by the power supply device; the use of the combination of historical data and the current environmental parameters to predict the predicted energy supply of the power supply device includes: time serializing the environmental factor parameters of each category to form an environmental factor sequence; time serializing the meteorological parameters to form a meteorological sequence; time serializing the historical energy supply data to form a historical energy supply sequence; and using the environmental factor sequence, the meteorological sequence and the historical energy supply sequence to predict the predicted energy supply of the power supply device within the target time interval.

[0020] In one embodiment of the present application, when the power supply device includes a wind power generation device, the environmental factor parameters include wind speed and wind direction; and / or, when the power supply device includes a photovoltaic power generation device, the environmental factor parameters include light intensity and sun angle.

[0021] In one embodiment of the present application, the power supply method also includes: obtaining real energy supply of the power supply device; wherein the real energy supply represents the actual output energy supply of the power supply device at a time corresponding to the predicted energy supply; utilizing the error between the predicted energy supply and the real energy supply to iteratively update a prediction module; wherein the prediction module is used to predict the predicted energy supply.

[0022] On the other hand, a computer device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program: obtaining a historical data combination and current environmental parameters; wherein the historical data combination comprises historical environmental parameters and historical energy supply data of a power supply device; predicting the predicted energy supply of the power supply device by using both the historical data combination and the current environmental parameters; and controlling whether the power supply device is connected to the grid based on the predicted energy supply and the energy demand of a data center.

[0023] On the other hand, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented: obtaining a historical data combination and current environmental parameters; wherein the historical data combination includes historical environmental parameters and historical energy supply data of a power supply device; using both the historical data combination and the current environmental parameters, predicting the predicted energy supply of the power supply device; and controlling whether the power supply device is connected to the grid based on the predicted energy supply and the energy demand of the data center.

[0024] The power supply control system, power supply system, computing system, power supply method, computer equipment and computer-readable storage medium of the above-mentioned data center predict the possible energy supply of the power supply device to obtain predicted energy supply, and by predicting the energy supply and the energy demand of the data center, control is performed on whether clean energy is connected to the grid to directly supply power to the data center, so as to reasonably plan whether the power supply device is connected to the grid, thereby improving the power generation utilization rate of the power supply device while ensuring that the data center obtains stable energy supply. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a structural schematic diagram of an embodiment of a power supply control system of a data center of the present application; Figure 2 It is a structural schematic diagram of an embodiment of a power supply system of a data center of the present application; Figure 3 It is a structural diagram of an embodiment of a computing system of the present application; Figure 4 It is a flowchart of the working principle of the computing system of this application; Figure 5 It is a flow chart of an embodiment of a power supply method for a data center of the present application; Figure 6 It is a flow chart of another embodiment of the power supply method for a data center of the present application; Figure 7 This is a flow chart of an embodiment of the present application for predicting energy supply; Figure 8 It is a structural diagram of an embodiment of a computer device of the present application. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0027] With the rapid development of artificial intelligence in recent years, the requirements for computer computing power have increased significantly. Computers such as data centers, which carry computing power, often consume a huge amount of electricity, accounting for more than 3% of the total power generation in society. The requirements for energy efficiency of data centers are also gradually increasing. It is usually hoped that the PUE (Power Usage Effectiveness, the ratio of all energy consumed by the data center to the energy consumed by the IT load) of a newly built data center is below 1.3. In order to meet the needs of the above scenarios, the use of new energy sources such as photovoltaics, wind power, and tides for power generation has emerged, and its importance in power supply in data centers has also become more prominent, which plays a certain important role in reducing PUE.

[0028] However, although renewable energy generation has great potential in data center power supply, the power output of photovoltaic, wind power and other power generation methods has the characteristics of randomness and intermittency, and usually cannot meet the requirements for power supply stability.

[0029] In order to solve the technical problem that clean energy is difficult to maintain stable power supply in related technologies, the present application provides a power supply control system for a data center, a power supply system for a data center, a computing system, a power supply method for a data center, a computer device, and a computer-readable storage medium. The power supply control system of the data center includes a prediction module and a control module. The prediction module is used to input a historical data combination and current environmental parameters to predict the predicted energy supply of the power supply device; wherein the historical data combination includes historical environmental parameters and historical energy supply data of the power supply device. The control module includes a prediction input terminal and a control terminal, and the prediction input terminal is connected to the prediction module, and is used to control the power supply device connected to the control terminal to be connected to the grid or off the grid based on the predicted energy supply and the required energy of the data center. The specific architecture and working principle of the present application are described below with examples.

[0030] See also Figure 1 , Figure 1 It is a structural diagram of an embodiment of a power supply control system of a data center of the present application.

[0031] In one embodiment, the power supply control system of the data center may include a prediction module 11 and a control module 12 .

[0032] The prediction module 11 is used to input a historical data combination and current environmental parameters to predict the predicted energy supply of the power supply device 20. The historical data combination includes historical environmental parameters and historical energy supply data of the power supply device 20.

[0033] The control module 12 includes a prediction input terminal and a control terminal. The prediction input terminal is connected to the prediction module 11 and is used to control the power supply device 20 connected to the control terminal to be connected to the grid or off the grid based on the predicted energy supply and the energy demand of the data center 30. For example, the power supply device 20 can be a clean energy device, a conventional power supply device, etc., which is not limited here. The clean energy device refers to a device that can obtain clean energy such as wind energy, light energy, tidal energy, etc., and convert the obtained clean energy into electrical energy; the conventional power supply device can be a device that converts thermal energy, chemical energy, etc. into electrical energy.

[0034] Grid connection means controlling the power supply device 20 to be connected to the current energy supply system of the data center 30 and to provide real-time energy to the data center 30. In other words, off-grid means controlling the power supply device 20 not to be connected to the current energy supply system of the data center 30 and not to provide real-time energy to the data center 30.

[0035] That is to say, the control module 12 can send a control signal to the power supply device according to the predicted energy supply output by the prediction module 11, so as to optimize the use efficiency of the power supply device, and at the same time determine the grid-connected time of the power supply device and even further include the charging and discharging strategy of the energy storage device.

[0036] It can be seen that in this embodiment, the prediction module 11 can be trained in advance, so as to use the trained prediction module 11 to predict the possible energy supply of the power supply device 20 to obtain the predicted energy supply. In this way, the control module 12 can control whether the power supply device 20 is connected to the grid based on the predicted energy supply. That is, the control can quickly control whether the clean energy is connected to the grid and directly supplies power to the data center 30 by predicting the energy supply and the energy demand of the data center 30, so as to reasonably plan whether the power supply device 20 is connected to the grid, thereby ensuring that the data center 30 obtains a stable energy supply while improving the power generation utilization rate of the power supply device.

[0037] Furthermore, when the power supply device 20 is off-grid based on predicted energy supply and demand energy control, the power supply device 20 can be controlled to supply energy to the energy storage device, so that the energy storage device can receive and store the electric energy output by the power supply device 20, thereby being used for the energy storage device to supply energy to the data center in real time or for subsequent energy supply to the data center 30.

[0038] See also Figure 2 , Figure 2 It is a structural diagram of an embodiment of a power supply system of a data center of the present application.

[0039] In one embodiment, the power supply system of the data center may include a power supply device 20 and a power supply control system of the data center.

[0040] The power supply device 20 is used to input clean energy and convert the clean energy into electrical energy for output.

[0041] The power supply control system of the data center is connected to the power supply device 20 .

[0042] The power supply control system of the data center includes a prediction module 11 and a control module 12. The prediction module 11 is used to input a combination of historical data and current environmental parameters to predict the predicted energy supply of the power supply device 20; wherein the combination of historical data includes historical environmental parameters and historical energy supply data of the power supply device 20; the control module 12 includes a prediction input terminal and a control terminal, the prediction input terminal is connected to the prediction module 11, and is used to control the power supply device 20 connected to the control terminal to be connected to the grid or off the grid based on the predicted energy supply and the required energy of the data center 30.

[0043] Furthermore, the power supply device 20 may include multiple power generation devices to increase the total energy supply of the power supply device 20, and is conducive to improving the scalability of the power supply device 20, and the power generation devices can be increased or decreased according to the actual power supply needs of the data center. The types of clean energy input by multiple power generation devices may be overlapping or different, so that the type of clean energy required by the power generation device can be selected according to the actual setting scenario of the power supply device, enriching the energy type selectability of the data center power supply.

[0044] For example, the power generation equipment may be wind power generation equipment, photovoltaic power generation equipment, tidal power generation equipment, etc. Therefore, the power supply device 20 may include one or more of wind power generation equipment, photovoltaic power generation equipment, tidal power generation equipment, etc., and the number of power generation equipment of each category may be at least one.

[0045] See also Figure 3 , Figure 3 It is a structural diagram of an embodiment of the computing system of the present application.

[0046] In one embodiment, the computing power system may include a data center 30 and a power supply control system of the data center.

[0047] The power supply control system of the data center is connected to the data center 30. The power supply control system of the data center includes a prediction module 11 and a control module 12; the prediction module 11 is used to input a combination of historical data and current environmental parameters to predict the predicted energy supply of the power supply device 20; wherein the combination of historical data includes historical environmental parameters and historical energy supply data of the power supply device 20; the control module 12 includes a prediction input terminal and a control terminal, the prediction input terminal is connected to the prediction module 11, and is used to control the power supply device 20 connected to the control terminal to be connected to the grid or off the grid based on the predicted energy supply and the required energy of the data center 30.

[0048] See also Figure 4 , Figure 4 It is a structural diagram of another embodiment of the computing power system of the present application.

[0049] In one embodiment, the computing system may further include a power supply device as described above. The power supply device is connected to the power supply control system of the data center, and is used to output power when the power supply control system determines to control it to be connected to the grid, so that the data center receives the output power.

[0050] Furthermore, the computing power system may also include an energy storage device. The energy storage device is connected to the power supply control system. The energy storage device is used to send its current energy storage information to the power supply control system. The power supply control system is used to determine whether to control the energy storage device to store or supply energy based on the predicted energy supply, current energy storage information and required energy.

[0051] Specifically, the energy storage device of the present application can receive and store the energy output by the power supply device when the energy output of the power supply device is insufficient to stably maintain the energy supply required for the operation of the data center, so as to fully collect the electric energy converted by the power supply device, and in the appropriate energy supply stage planned by the present application, the energy storage device can release the electric energy to supply energy to the data center, thereby improving the power supply reliability of the data center, and can reduce resource waste while maintaining reliable power supply to the data center.

[0052] In an alternative embodiment, the energy storage device may be included in a power supply system, which is not limited herein.

[0053] The following combination Figure 4 The computing system shown in the example explains the power supply principle of the data center.

[0054] The computing power system may include a storage module and a sensing module. The storage module is used to store historical data combinations, and the sensing module is used to sense current environmental parameters.

[0055] The control module may also include an energy storage input terminal for connecting to the energy storage device to obtain current energy storage information of the energy storage device, so as to combine the current energy storage information, predicted energy supply and demand energy to determine whether the power supply device is connected to the grid or off-grid. Therefore, in the process of considering whether the power supply device is connected to the grid, the energy currently stored in the energy storage device is combined with the total supplyable energy that may exist at the future moment or the current moment, thereby improving the grid connection rate of the power supply device and reducing the frequent switching of the power supply device to connect to the grid or off-grid, so as to improve the stability of the energy input from the power supply device, and also facilitate to make full use of the energy stored in the energy storage device when the power supply of the power supply device is insufficient.

[0056] Specifically, the control module may include an energy storage control unit, a first switch unit and a first conversion unit.

[0057] The energy storage control unit is used to determine whether to control the energy storage device to store or supply energy based on the current energy storage information, predicted energy supply and required energy, and output a charge and discharge signal when it is determined that the energy storage device is controlled to store or supply energy.

[0058] The first switch unit is connected to the energy storage control unit and the first conversion unit, and the first switch unit is also used to connect to the energy storage device and receive the charge and discharge signal to conduct the path between the energy storage device and the first conversion unit. Optionally, the first switch unit can be a bidirectional DC / DC conversion unit to convert the power input to the energy storage device and the power output from the energy storage device to the data center.

[0059] For example, the energy storage device may include at least one of a battery, a capacitor, etc., and the number of energy storage devices may be one or more. Optionally, multiple energy storage devices may be used as a whole to supply power to the data center or receive and store energy input from a power supply device; or, multiple energy storage devices may be independently controlled, which is not limited here.

[0060] The first conversion unit is also used to connect to the data center and the power supply device to perform power conversion on the electric energy flowing through the first conversion unit.

[0061] Please continue reading Figure 4 In one embodiment, the control module may further include a power supply control unit, a second switch unit, and a second conversion unit.

[0062] The energy supply control unit is used to control the power supply device to be connected to the grid and output a grid connection signal.

[0063] The second switch unit is connected to the energy supply control unit and the second conversion unit. The second switch unit is also used to connect to the power supply device to receive the grid-connected signal to conduct the path between the power supply device and the second conversion unit.

[0064] The second conversion unit is also used to connect to the data center to convert the output energy of the power supply device and transmit it to the data center.

[0065] Please continue reading Figure 4 In one embodiment, in combination with the above-mentioned implementation that the power supply device may include multiple power generation devices, the prediction module may include multiple prediction units. The multiple prediction units are used to respectively predict the output power supply of the multiple power generation devices of the power supply device; wherein the power supply device includes multiple power generation devices.

[0066] For example, Figure 4 As shown in the example, the power supply device may include a wind power generation device and a photovoltaic power generation device. Based on this, the prediction module may include two prediction units (not shown in the figure), one of which is used to predict the predicted energy supply of the wind power generation device, and the other prediction unit is used to predict the predicted energy supply of the photovoltaic power generation device.

[0067] like Figure 4As shown in the example, the second conversion unit may include an AC / DC conversion unit and a DC / DC conversion unit. The AC / DC conversion unit is connected to the wind power generation equipment through the second switch unit to convert the AC power converted by the wind power generation equipment through wind energy into DC power that can be supplied to the data center. The DC / DC conversion unit is connected to the photovoltaic power generation equipment through the second switch unit to convert the DC power converted by the photovoltaic power generation equipment through light energy into DC power that can be supplied to the data center.

[0068] Alternatively, if Figure 4 As shown in the example, the control module may include multiple second switch units to be connected to the power supply devices respectively, so as to facilitate the control of each power generation equipment, improve the ease of maintainability of updating the control method of each power generation equipment, and improve the scalability of the power supply control system and the computing center.

[0069] That is to say, in this embodiment, the storage module and the sensing module serve as the data input side, and transmit the historical energy supply data of the clean energy device, the historical environmental parameters, and the current environmental parameters sensed by the sensing module, etc., as valid data to the prediction model through the data bus connected to the two. Among them, the valid data may include but is not limited to the output power, wind speed, wind direction, light angle, irradiation intensity, temperature and humidity, meteorological data, etc. with timestamps.

[0070] Then, the prediction phase is entered, and the predicted energy supply can be predicted by the prediction module. The prediction module can also be updated by the subsequent real energy supply. The predicted energy supply predicted by the prediction module is sent to the control module through the control signal bus, so that the control module can adaptively send control signals through signals to schedule the power flow in the high-voltage DC bus.

[0071] Specifically, the predicted energy supply can be used to coordinate and control the control units of the power generation side and the energy storage side. For the control unit on the power generation side (i.e., the energy supply control unit), it can control whether the power supply device is connected to the grid. For example, wind power generation equipment can output high-voltage AC power, which is connected to the high-voltage DC bus through the AC / DC conversion unit. Photovoltaic power generation equipment can output high-voltage DC power, which is connected to the high-voltage DC bus through the DC / DC conversion unit.

[0072] On the energy storage side, the energy storage device can be mounted on the high-voltage DC bus through a bidirectional DC / DC conversion unit to achieve energy storage and supply of the energy storage device.

[0073] For the specific limitations of the power supply control system, power supply system and computing power system of the data center, please refer to the limitations of the power supply method of the data center in the following text, which will not be repeated here. The various modules in the power supply control system, power supply system and computing power system of the above-mentioned data center can be implemented in whole or in part through software, hardware and their combination. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0074] See also Figure 5 , Figure 5 It is a flow chart of an embodiment of a power supply method for a data center of the present application.

[0075] S101: Acquire a historical data combination and current environmental parameters; wherein the historical data combination includes historical environmental parameters and historical energy supply data of a power supply device.

[0076] In this embodiment, the environmental parameters can characterize the environmental information of the environment in which the power supply device is located. The environmental parameters are associated with the electric energy that can be output by the power supply device, that is, different environmental parameters can have an impact on the electric energy that can be output by the power supply device. The environmental parameters at historical moments and the output energy of the power supply device at historical moments can be obtained as a combination of historical data. The environmental information of the current environment of the power supply device can also be obtained.

[0077] S102: Utilizing both the historical data combination and the current environmental parameters, predicting the energy supply of the power supply device.

[0078] In this embodiment, based on the correlation between the historical environmental parameters and the historical energy supply data in the historical data combination, the output electric energy of the power supply device under the current environmental parameters can be predicted as the predicted energy supply.

[0079] Optionally, the output power of the power supply device at the current moment may be predicted; or, the output power of the power supply device within a period of time in the future (ie, a target time interval) may be predicted, which is not limited here.

[0080] S103: Based on the predicted energy supply and the energy demand of the data center, control whether the power supply device is connected to the grid.

[0081] In this embodiment, the energy demand of the data center can be obtained. Optionally, the energy demand of the data center can be predicted; or the energy demand of the data center can be estimated based on the historical energy consumption of the data center to reduce the amount of calculation of the power supply method of the data center, thereby facilitating the stability of the power supply of the data center.

[0082] In this way, whether the power supply device is connected to the grid can be controlled based on the predicted energy supply and the energy demand of the data center. Optionally, when the power supply device is controlled not to be connected to the grid, that is, when the power supply device is controlled to be off the grid, the power supply device can be controlled to supply power to a storage device, etc., so as to fully collect the output power of the power supply device, thereby facilitating the improvement of the utilization rate of the output power of the power supply device. The power stored in the storage device can be used to supply power to the data center or other devices, which is not limited here; or it can be suspended.

[0083] It can be seen that in this embodiment, the possible energy supply of the power supply device is predicted to obtain the predicted energy supply. By predicting the energy supply and the energy demand of the data center, whether the clean energy is connected to the grid to directly supply power to the data center is controlled, so as to reasonably plan whether the power supply device is connected to the grid, thereby ensuring that the data center obtains stable energy supply while improving the utilization rate of clean energy power generation.

[0084] See also Figure 6 , Figure 6 It is a flow chart of another embodiment of the power supply method for a data center of the present application.

[0085] S201: Acquire historical data combinations and current environmental parameters.

[0086] S202: Utilizing both the historical data combination and the current environmental parameters, predicting the energy supply of the power supply device.

[0087] In this embodiment, whether the power supply device is connected to the grid can be controlled based on the predicted energy supply and the energy demand of the data center.

[0088] S203: Integrate the current energy storage and the predicted energy supply as the predicted total energy.

[0089] In this embodiment, the current stored energy of the energy storage device may be obtained to merge the current stored energy with the predicted energy supply as the predicted total energy.

[0090] S204: Compare the predicted total energy with the required energy.

[0091] In this embodiment, when the predicted total energy is higher than the required total energy, it is considered that the power supply device can be controlled to be connected to the grid to supply energy to the data center, and step S205 is executed. When the predicted total energy is lower than the required total energy, it is considered that the power supply device needs to be controlled to be off-grid, and step S208 is executed.

[0092] As designed in this way, in the present embodiment, when considering whether the power supply device is connected to the grid, the energy currently stored in the energy storage device can be combined to comprehensively consider the total available energy that may exist in the future or at the current moment, thereby improving the grid connection rate of the power supply device, reducing the frequent switching of the power supply device to be connected to the grid or off the grid, so as to improve the stability of the energy input from the power supply device, and is also conducive to making full use of the energy stored in the energy storage device when the power supply of the power supply device is insufficient.

[0093] S205: Compare the predicted energy supply with the required energy.

[0094] In this example, when the predicted energy supply is higher than the required energy, step S206 is executed; when the predicted energy supply is lower than the required energy, step S207 is executed.

[0095] S206: Control the power supply device to supply energy to the data center and the energy storage device.

[0096] In this embodiment, in response to the predicted energy supply of the power supply device being higher than the energy demand of the data center, it can be considered that the power supply device can independently provide stable electricity to the data center, and there is a certain amount of surplus electricity, so the power supply device can be controlled to supply energy to the data center and the energy storage device, thereby ensuring reliable power supply to the data center while effectively collecting the surplus electricity generated by the power supply device. This is beneficial when the power independently output by the power supply device is insufficient to maintain the power supply of the data center. It is considered whether the power stored in the energy storage device can maintain the power supply of the data center, thereby reducing the frequent control of the switching of the power supply device on or off the grid, which is beneficial to maintaining the stability of the computing power system.

[0097] S207: Control both the power supply device and the energy storage device to supply energy to the data center.

[0098] In this embodiment, in response to the predicted total energy being higher than the required energy but the predicted energy supply being lower than the required energy, both the power supply device and the energy storage device can be controlled to supply energy to the data center together, which can help extend the grid-connected time of the power supply device and help reduce the risk of the power supply device frequently switching between being connected to the grid or being off the grid, affecting the stability of the power supply device.

[0099] S208: Control the power supply device to be off-grid, and control the power supply device to supply energy to the energy storage device.

[0100] In this embodiment, in response to the prediction that the total energy is lower than the required energy, the power supply device can be controlled to be off-grid and to output electric energy to the energy storage device, so that the energy storage device can receive the electric energy output by the power supply device and store the electric energy for subsequent use in powering the data center. In this embodiment, when the output electric energy of the power supply device is insufficient, the output electric energy of the power supply device can be stored, thereby reducing the waste of electric energy converted by the power supply device, which is beneficial to improving the utilization rate of the electric energy converted by the power supply device.

[0101] Furthermore, taking the example of the power supply device described in the foregoing text as including multiple power generation equipment, the electric energy that can be output by each power generation equipment can be predicted separately to obtain the predicted energy supply of each power generation equipment. When the sum of the predicted energy supplies of each power generation equipment is higher than the required energy, the stability of the power output of each power generation equipment can be predicted based on the environmental parameters, and a first number of power generation equipment with relatively high stability can be selected to supply power to the data center; the predicted energy supply of each power generation equipment can be sorted, and a second number of power generation equipment with high predicted energy supply can be selected to supply power to the data center. Among them, the total predicted energy supply of the selected first number of power generation equipment and the second number of power generation equipment is higher than the required energy. Optionally, the value of the first number can be zero; the value of the second number can be zero.

[0102] See also Figure 7 , Figure 7 It is a flow chart of an embodiment of the present application for predicting energy supply.

[0103] S301: Obtain environmental factor parameters.

[0104] In this embodiment, parameters of environmental factors that affect the power output of the power supply device may be acquired.

[0105] For example, when the power supply device includes a wind power generation device, the environmental factor parameters include wind speed and wind direction. And / or, when the power supply device includes a photovoltaic power generation device, the environmental factor parameters include light intensity and sun angle.

[0106] S302: Preprocessing environmental factor parameters.

[0107] In this embodiment, pre-processing such as data cleaning may be performed on the environmental factor parameters.

[0108] For example, the originally collected environmental factor parameters and other data can be preprocessed, which can include data cleaning, format conversion, outlier removal, etc. For example, the data formats of different sensors may be different, which can be standardized into a unified data format; the missing time period data can be supplemented by interpolation to ensure the integrity of the time series; when it is determined that the data is abnormal due to accidental sensor failure, the faulty sensor can be removed. In this way, the accuracy and consistency of the input data can be improved, which is conducive to ensuring that the prediction module can use data with good quality for prediction.

[0109] S303: Construct a time series.

[0110] In this embodiment, each category of environmental factor parameters may be processed in time series to form an environmental factor sequence.

[0111] Meteorological parameters can also be processed in time series to form a meteorological sequence. Historical energy supply data can also be processed in time series to form a historical energy supply sequence.

[0112] That is to say, in this embodiment, a time series can be constructed for historical environmental parameters, current environmental parameters, and historical energy supply data.

[0113] Taking wind power generation equipment as an example, environmental parameters may include wind speed V wind , wind direction D wind , temperature T wind , historical wind power generation P wind Taking photovoltaic power generation equipment as an example, environmental parameters may include light intensity I solar , sun angle A solar , temperature T solar , historical photovoltaic power generation power P solar As such, in this embodiment, the changing trends of environmental variables such as wind speed, wind direction, and light intensity over time can be captured, and future energy supply such as power generation can be predicted.

[0114] S304: Input the time series into a prediction module to predict the power supply of the power supply device within the target time interval.

[0115] In this embodiment, as described above by way of example, the prediction module is used to predict the predicted energy supply.

[0116] Environmental factor sequences, meteorological sequences and historical energy supply sequences can be used to predict the energy supply of power supply devices within the target time interval.

[0117] Optionally, the prediction module can be in the form of an algorithm, a model, a formula, etc., which is not limited here. Taking the prediction module including a prediction model as an example, the prediction model can include an input layer, a time series network layer such as an LSTM (Long Short-Term Memory) layer, and an output layer, wherein the number of time series network layers can be multiple layers, so as to improve the accuracy of capturing time dependencies and correlations between features. The working principle of the prediction model will be explained with examples later, so it will not be repeated here.

[0118] S305: Acquire the actual energy supply of the power supply device to update the prediction module.

[0119] In this embodiment, the real energy supply refers to the actual output energy supply of the power supply device at the time corresponding to the predicted energy supply.

[0120] The error between the predicted energy supply and the actual energy supply can be used to iteratively update the prediction module. That is, in this embodiment, the prediction module can be corrected by using the actual energy supply, so that the weights and parameters of the prediction model can be dynamically adjusted based on the output feedback of the real-time energy supply, which is conducive to optimizing the weight of the prediction model, thereby driving the predicted energy supply to be as close as possible to the actual energy supply, and improving the prediction accuracy of the prediction module.

[0121] S306: Using the predicted energy supply to control whether the power supply device is connected to the grid.

[0122] The following is an example to illustrate how the prediction model works.

[0123] The input layer is used to input time series data. Its matrix shape is (n, s, d), where n is the number of samples, s is the number of time steps, and d is the number of features in each time step. wind 、Wind direction D wind , Temperature T wind 、Historical wind power generation P wind The input matrix with four eigenvalues ​​as an example can be expressed as: X= =

[0124] Where X represents the input matrix.

[0125] The LSTM layer is used to process time series data. The input is the features of the past t time steps, and the output is the hidden state vector of the last time step. That is, the input can be:

[0126] The output is the hidden state h t .

[0127] The LSTM layer contains a forget gate, an input gate, and an output gate to control the memory state. For each time step, the calculation formula of the LSTM layer can be:

[0128]

[0129]

[0130]

[0131] Among them, f t 、i t , o t are the activation values ​​of the forget gate, input gate, and output gate, respectively. is the candidate cell state, C t is the current cell state, h t is the current hidden state (i.e., predicted energy supply); σ represents the activation function; x t represents the time series input at the current time t; W f 、b f represents the coefficient and bias of the linear relationship associated with the forget gate; h t-1 and C t-1 They represent the hidden state parameters and cell state parameters at time t-1 respectively; W i 、b i The coefficients and bias of the linear relationship associated with the input gate; W o , b o Represents the coefficients and bias of the linear relationship associated with the output gate; tanh() represents the hyperbolic tangent function.

[0132] The output of the LSTM layer can also be connected to a fully connected layer (Dense Layer) to generate the final prediction result through a linear transformation and activation function. The calculation formula can be:

[0133] Among them, W dense is the weight of the fully connected layer, b dense is the bias, output Ppred,t+1 is the predicted power of wind or photovoltaic.

[0134] Furthermore, the difference between the predicted energy supply and the actual energy supply can be measured by using methods such as mean square error. The calculation formula of mean square error can be as follows:

[0135] Where N is the number of samples, Ppred,i is the predicted power of the ith sample, and Pactual,i is the actual power of the ith sample. By minimizing the loss function and other methods, the weight of the prediction model is optimized, so that the predicted energy supply can be as close to the actual energy supply as possible.

[0136] Taking the forecast of wind power generation equipment as an example, the input time series data may include historical wind speed, wind direction, temperature, humidity and historical wind power generation. The expression can be as follows:

[0137] Taking the forecast of photovoltaic power generation equipment as an example, the input time series data may include historical light intensity, sun angle, temperature, humidity and historical photovoltaic power generation. The expression can be as follows:

[0138] In this way, the prediction model responds to the wind power or photovoltaic related data received in the past t time steps, and after processing by several layers of time series network layers and fully connected layers, outputs the wind power generation power P at the next time t+1 wind,t+1 or photovoltaic power generation power P solar,t+1 , the calculation formula can be as follows:

[0139]

[0140] Among them, LSTM means using LSTM to perform time series processing.

[0141] After predicting the wind power generation and photovoltaic power generation respectively through independent prediction models, the total power generation of the two can be calculated. The calculation formula can be as follows:

[0142] Among them, P total,t+1 It represents the sum of wind power generation and photovoltaic power generation at the next moment.

[0143] Therefore, the predicted energy supply that the power supply device may output in the future can be calculated using the following calculation formula:

[0144] Among them, E pre Indicates the predicted energy that the power supply device may output in the future.

[0145] E pre Indicates that in the target time interval T dTaking the predicted energy supply of the internal power supply device as an example, the predicted total energy can be calculated based on the predicted energy supply and the current energy storage of the energy storage device. The calculation formula can be as follows:

[0146] Among them, E out Indicates the total prediction energy; E store represents the current energy storage of the energy storage device; k(T d ) represents the prediction weighting factor, k(T d ) can be an empirical value or obtained through iterative training, which is not limited here.

[0147] In summary, the present application provides an application solution that can combine clean energy generation such as new energy with energy storage to control the power supply of a data center, change the power supply architecture of the data center, overcome the technical problem of low utilization rate of clean energy generation such as wind power generation and photovoltaic power generation in the data center, and improve the coordinated control efficiency of the power generation and energy storage sides. It can also reduce the risk of difficulty in stable power supply due to the lack of output power of clean energy such as wind power and photovoltaics, and realize the organic combination of clean energy power generation and energy storage devices through the energy supply forecast of the power supply device. In addition, information and energy transmission can be realized through the data bus, control signal bus and high-voltage DC bus, and unified and coordinated control of the data center from the data side, power generation side and energy storage side can be performed.

[0148] It should be understood that although Figure 5-Figure 7 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 5-Figure 7 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0149] In one embodiment, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the steps of the power supply method for a data center described above are implemented, which will not be described in detail here.

[0150] See also Figure 8 , Figure 8 It is a structural diagram of an embodiment of a computer device of the present application.

[0151] In one embodiment, the computer device may be a server, and its internal structure diagram may be as follows: Figure 8 As shown in the example.

[0152] The computer device includes a processor, a memory, a network interface and a database connected through a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data such as historical data combinations. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a power supply method for a data center is implemented.

[0153] Those skilled in the art will understand that Figure 8 The structure shown as an example is merely a block diagram of a portion of the structure related to the present application solution, and does not constitute a limitation on the computer device to which the present application solution is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0154] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the following steps may be implemented: S101: Acquire a historical data combination and current environmental parameters; wherein the historical data combination includes historical environmental parameters and historical energy supply data of a power supply device.

[0155] S102: Utilizing both the historical data combination and the current environmental parameters, predicting the energy supply of the power supply device.

[0156] S103: Based on the predicted energy supply and the energy demand of the data center, control whether the power supply device is connected to the grid.

[0157] In one embodiment, when controlling whether the power supply device is connected to the grid based on the predicted energy supply and the energy demand of the data center, the processor can also implement the following steps when executing the computer program: obtaining the current energy storage of the energy storage device; integrating the current energy storage and the predicted energy supply as the predicted total energy; in response to the predicted total energy being higher than the required energy, controlling the power supply device to be connected to the grid to supply energy to the data center.

[0158] In one embodiment, when controlling the power supply device to be connected to the grid to supply energy to the data center, the processor may also implement the following steps when executing the computer program: comparing the predicted energy supply with the required energy; in response to the predicted energy supply being higher than the required energy, controlling the power supply device to supply energy to the data center and the energy storage device.

[0159] In one embodiment, when controlling the power supply device to be connected to the grid to supply energy to the data center, the processor may also implement the following steps when executing the computer program: comparing the predicted energy supply with the required energy; in response to the predicted energy supply being lower than the required energy, controlling the power supply device and the energy storage device to supply energy to the data center.

[0160] In one embodiment, after integrating the current energy storage and the predicted energy supply as the predicted total energy, the processor can also implement the following steps when executing the computer program: in response to the predicted total energy being lower than the required energy, control the power supply device to go off the grid; control the power supply device to supply energy to the energy storage device.

[0161] In one embodiment, the environmental parameters include environmental factor parameters and meteorological parameters; wherein the environmental factor represents the environmental factor associated with the type of clean energy input by the power supply device; when the predicted energy supply of the power supply device is predicted by using the combination of historical data and the current environmental parameters, the processor can also implement the following steps when executing the computer program: time-serializing each category of environmental factor parameters to form an environmental factor sequence; time-serializing the meteorological parameters to form a meteorological sequence; time-serializing the historical energy supply data to form a historical energy supply sequence; using the environmental factor sequence, the meteorological sequence and the historical energy supply sequence to predict the predicted energy supply of the power supply device within the target time interval.

[0162] Further, when the power supply device includes a wind power generation device, the environmental factor parameters include wind speed and wind direction; and / or, when the power supply device includes a photovoltaic power generation device, the environmental factor parameters include light intensity and sun angle.

[0163] In one embodiment, the processor may also implement the following steps when executing a computer program: obtaining the actual energy supply of the power supply device; wherein the actual energy supply represents the actual output energy supply of the power supply device at a time corresponding to the predicted energy supply; utilizing the error between the predicted energy supply and the actual energy supply to iteratively update the prediction module; wherein the prediction module is used to predict the predicted energy supply.

[0164] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps can be implemented: S101: Acquire a historical data combination and current environmental parameters; wherein the historical data combination includes historical environmental parameters and historical energy supply data of a power supply device.

[0165] S102: Utilizing both the historical data combination and the current environmental parameters, predicting the energy supply of the power supply device.

[0166] S103: Based on the predicted energy supply and the energy demand of the data center, control whether the power supply device is connected to the grid.

[0167] In one embodiment, when controlling whether the power supply device is connected to the grid based on the predicted energy supply and the energy demand of the data center, the computer program can also implement the following steps when executed by the processor: obtaining the current energy storage of the energy storage device; integrating the current energy storage and the predicted energy supply as the predicted total energy; in response to the predicted total energy being higher than the required energy, controlling the power supply device to be connected to the grid to supply energy to the data center.

[0168] In one embodiment, when controlling the power supply device to be connected to the grid to supply energy to the data center, the computer program can also implement the following steps when executed by the processor: comparing the predicted energy supply with the required energy; in response to the predicted energy supply being higher than the required energy, controlling the power supply device to supply energy to the data center and the energy storage device.

[0169] In one embodiment, when controlling the power supply device to be connected to the grid to supply energy to the data center, the computer program can also implement the following steps when executed by the processor: comparing the predicted energy supply with the required energy; in response to the predicted energy supply being lower than the required energy, controlling the power supply device and the energy storage device to supply energy to the data center.

[0170] In one embodiment, after integrating the current energy storage and the predicted energy supply as the predicted total energy, the computer program can also implement the following steps when executed by the processor: in response to the predicted total energy being lower than the required energy, control the power supply device to go off the grid; control the power supply device to supply energy to the energy storage device.

[0171] In one embodiment, the environmental parameters include environmental factor parameters and meteorological parameters; wherein the environmental factor represents the environmental factor associated with the type of clean energy input by the power supply device; when the predicted energy supply of the power supply device is predicted by using the combination of historical data and the current environmental parameters, the computer program can also implement the following steps when executed by the processor: time-serialize the environmental factor parameters of each category to form an environmental factor sequence; time-serialize the meteorological parameters to form a meteorological sequence; time-serialize the historical energy supply data to form a historical energy supply sequence; use the environmental factor sequence, meteorological sequence and historical energy supply sequence to predict the predicted energy supply of the power supply device within the target time interval.

[0172] Further, when the power supply device includes a wind power generation device, the environmental factor parameters include wind speed and wind direction; and / or, when the power supply device includes a photovoltaic power generation device, the environmental factor parameters include light intensity and sun angle.

[0173] In one embodiment, when the computer program is executed by a processor, the following steps can also be implemented: obtaining the actual energy supply of the power supply device; wherein the actual energy supply represents the actual output energy supply of the power supply device at the time corresponding to the predicted energy supply; utilizing the error between the predicted energy supply and the actual energy supply to iteratively update the prediction module; wherein the prediction module is used to predict the predicted energy supply.

[0174] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0175] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0176] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application.

Claims

1. A power supply control system for a data center, characterized in that: The power supply control system comprises: A prediction module, used for inputting a historical data combination and current environmental parameters to predict the predicted energy supply of the power supply device; wherein the historical data combination includes historical environmental parameters and historical energy supply data of the power supply device; The control module includes a prediction input terminal and a control terminal. The prediction input terminal is connected to the prediction module and is used to control the power supply device connected to the control terminal to be connected to or off the grid based on the predicted energy supply and the energy demand of the data center.

2. The power supply control system according to claim 1, characterized in that: The control module also includes an energy storage input terminal for connecting to an energy storage device to obtain current energy storage information of the energy storage device, so as to determine whether to connect the power supply device to the grid or go off the grid in combination with the current energy storage information, the predicted energy supply and the required energy.

3. The power supply control system according to claim 2, characterized in that: The control module includes an energy storage control unit, a first switch unit and a first conversion unit; The energy storage control unit is used to determine whether to control the energy storage device to store energy or supply energy based on the current energy storage information, the predicted energy supply and the required energy, and output a charge and discharge signal when it is determined that the energy storage device is controlled to store energy or supply energy; The first switch unit is connected to the energy storage control unit and the first conversion unit, and the first switch unit is also used to connect to the energy storage device and receive the charge and discharge signal to conduct the path between the energy storage device and the first conversion unit; The first conversion unit is also used to connect to the data center and the power supply device to perform power conversion on the electric energy flowing through the first conversion unit.

4. The power supply control system according to claim 1, characterized in that: The control module includes an energy supply control unit, a second switch unit and a second conversion unit; The energy supply control unit is used to control the power supply device to be connected to the grid and output a grid connection signal; The second switch unit is connected to the energy supply control unit and the second conversion unit, and the second switch unit is also used to connect to the power supply device to receive the grid-connected signal to conduct the path between the power supply device and the second conversion unit; The second conversion unit is also used to connect to the data center to perform power conversion on the output energy of the power supply device and transmit it to the data center.

5. The power supply control system according to claim 1, characterized in that: The prediction module includes a plurality of prediction units; The multiple prediction units are used to respectively predict the output energy supply of the multiple power generation devices included in the power supply device.

6. A power supply system for a data center, characterized in that: The power supply system comprises: A power supply device, used to input clean energy and convert the clean energy into electrical energy for output; The power supply control system of the data center is connected to the power supply device; the power supply control system includes a prediction module and a control module; the prediction module is used to input a historical data combination and current environmental parameters to predict the predicted energy supply of the power supply device; wherein the historical data combination includes historical environmental parameters and historical energy supply data of the power supply device; the control module includes a prediction input terminal and a control terminal, the prediction input terminal is connected to the prediction module, and is used to control the power supply device connected to the control terminal to be connected to or off the grid based on the predicted energy supply and the energy demand of the data center.

7. The power supply system according to claim 6, characterized in that: The power supply device includes a plurality of power generation devices.

8. The power supply system according to claim 6, characterized in that: The power supply system also includes an energy storage device; The energy storage device is connected to the power supply control system; the energy storage device is used to send its current energy storage information to the power supply control system; The power supply control system is used to determine whether to control the energy storage device to store energy or supply energy based on the predicted energy supply, the current energy storage information and the required energy.

9. A computing power system, characterized in that: The computing power system comprises: Data centers; A power supply control system of a data center is connected to the data center; the power supply control system includes a prediction module and a control module; the prediction module is used to input a historical data combination and current environmental parameters to predict the predicted energy supply of the power supply device; wherein the historical data combination includes historical environmental parameters and historical energy supply data of the power supply device; the control module includes a prediction input terminal and a control terminal, the prediction input terminal is connected to the prediction module, and is used to control the power supply device connected to the control terminal to be connected to or off the grid based on the predicted energy supply and the energy demand of the data center.

10. The computing system according to claim 9, characterized in that: The computing power system also includes a power supply device; The power supply device is connected to the power supply control system, and is used for outputting energy when the power supply control system determines to control it to be connected to the grid, so that the data center receives the output energy.

11. A power supply method for a data center, characterized in that: The power supply method comprises: Acquire a historical data combination and current environmental parameters; wherein the historical data combination includes historical environmental parameters and historical energy supply data of the power supply device; Using both the combination of the historical data and the current environmental parameters, predicting the predicted energy supply of the power supply device; Based on the predicted energy supply and the energy demand of the data center, control is performed on whether the power supply device is connected to the grid.

12. The power supply method according to claim 11, characterized in that: The controlling whether the power supply device is connected to the grid based on the predicted energy supply and the energy demand of the data center includes: Obtaining the current stored energy of the energy storage device; fusing the current energy storage and the predicted energy supply as predicted total energy; In response to the predicted total energy being higher than the required energy, the power supply device is controlled to be connected to the grid to supply energy to the data center.

13. The power supply method according to claim 12, characterized in that: The controlling the power supply device to be connected to the grid to supply energy to the data center comprises: comparing the predicted energy supply with the required energy; In response to the predicted energy supply being higher than the required energy, the power supply device is controlled to supply energy to the data center and the energy storage device.

14. The power supply method according to claim 12, characterized in that: The controlling the power supply device to be connected to the grid to supply energy to the data center comprises: comparing the predicted energy supply with the required energy; In response to the predicted energy supply being lower than the required energy, the power supply device and the energy storage device are controlled to supply energy to the data center.

15. The power supply method according to claim 12, characterized in that: The fusing of the current energy storage and the predicted energy supply as the predicted total energy further includes: In response to the predicted total energy being lower than the required energy, controlling the power supply device to be off-grid; Control the power supply device to supply energy to the energy storage device.

16. The power supply method according to claim 11, characterized in that: Environmental parameters include environmental factor parameters and meteorological parameters; wherein the environmental factor represents an environmental factor associated with the type of clean energy input by the power supply device; The predicting of the energy supply of the power supply device by using the combination of the historical data and the current environmental parameters includes: Performing time series processing on the environmental factor parameters of each category to form an environmental factor sequence; performing time series processing on the meteorological parameters to form a meteorological sequence; performing time series processing on the historical energy supply data to form a historical energy supply sequence; The predicted energy supply of the power supply device within a target time interval is predicted by using the environmental factor sequence, the meteorological sequence and the historical energy supply sequence.

17. The power supply method according to claim 16, characterized in that: When the power supply device includes a wind power generation device, the environmental factor parameters include wind speed and wind direction; and / or, When the power supply device includes a photovoltaic power generation device, the environmental factor parameters include light intensity and sun angle.

18. The power supply method according to claim 11, characterized in that: The power supply method further comprises: Acquire the real energy supply of the power supply device; wherein the real energy supply represents the actual output energy supply of the power supply device at the time corresponding to the predicted energy supply; The prediction module is iteratively updated using the error between the predicted energy supply and the actual energy supply; wherein the prediction module is used to predict the predicted energy supply.

19. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the power supply method for a data center according to any one of claims 11 to 18 are implemented.

20. 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 method for a data center described in any one of claims 11 to 18 are implemented.

Images