A high-voltage direct-current power supply system for a green data center

Abstract

This invention discloses a high-voltage direct current (HVDC) power supply system for green data centers, comprising an execution module, a calculation module, and an acquisition module. The calculation module calculates the expected power of each power supply side based on load forecast curves, photovoltaic output forecast curves, and real-time electricity prices, and generates a first operating instruction. It also determines the expected voltage value based on the load forecast curves and generates a second operating instruction. The execution module operates each power supply side according to the first operating instruction and the HVDC bus according to the second operating instruction, acquiring real-time power and voltage values. Based on the expected power and real-time power, it determines first correction instructions and superimposes them onto each power supply side for execution. Based on the expected voltage and real-time voltage values, it determines a second correction instruction and loads it onto the HVDC bus for execution. This invention effectively utilizes green energy and ensures optimal efficiency of the HVDC power supply system under all operating conditions.

Description

Technical Field

[0001] This invention relates to the field of high-voltage DC power supply technology for data centers, and particularly to a high-voltage DC power supply system for green data centers. Background Technology

[0002] With the rapid development of cloud computing, big data, and artificial intelligence technologies, data centers are continuously expanding in scale, and their power consumption has become a significant component of global energy consumption. Traditional data center power supply systems generally adopt AC UPS (Uninterruptible Power Supply) solutions. The power needs to undergo multiple stages of transformation, including AC mains power, rectification (AC / DC), inversion (DC / AC), and server power supply rectification (AC / DC). Each stage of transformation generates energy loss, making it difficult for the overall system efficiency to exceed 92%.

[0003] High-voltage direct current (HVDC) power supply technology converts AC grid power into DC power at 240V, 336V or higher voltage levels, directly powering data center IT equipment. This reduces intermediate conversion steps such as rectification and inversion, significantly improving power supply efficiency.

[0004] Photovoltaics and energy storage are developing rapidly. However, existing high-voltage direct current (HVDC) power supply technologies lack a dynamic power distribution mechanism based on photovoltaics, energy storage, and general grid power (such as thermal power and hydropower), resulting in a low utilization rate of green energy by HVDC power supply technologies. Summary of the Invention

[0005] The main objective of this invention is to provide a high-voltage direct current (HVDC) power supply system for green data centers, aiming to solve the problem that current HVDC power supply technologies lack a dynamic power allocation mechanism based on photovoltaics, energy storage, and general grid power (such as thermal power and hydropower), resulting in low utilization of green energy.

[0006] To achieve the above objectives, this invention proposes a high-voltage DC power supply system for green data centers, comprising an execution module, a computing module, and an acquisition module; The acquisition module is used to acquire the load prediction curve, photovoltaic output prediction curve, real-time electricity price, first preset value, second preset value, and optimization objective function, respectively. The calculation module is used to calculate the expected power of each of the energy supply sides in the future time period through the optimization objective function based on the load prediction curve, the photovoltaic output prediction curve and the real-time electricity price, and generate the first operation instruction based on each expected power; determine the expected voltage value of the high-voltage DC bus based on the load prediction curve, and generate the second operation instruction based on the expected voltage value; wherein, each of the energy supply sides includes the grid power side, the photovoltaic side and the energy storage side; The execution module is used to run each of the power supply sides according to the first running instruction and the second running instruction to run the high-voltage DC bus, to obtain the real-time power of each of the power supply sides and the real-time voltage value of the high-voltage DC bus; to determine each of the first correction instructions according to each of the expected power and each of the real-time power, and to superimpose each of the first correction instructions onto each of the power supply sides for execution; to determine a second correction instruction according to the expected voltage value and the real-time voltage value, and to load the second correction instruction onto the high-voltage DC bus for execution.

[0007] Preferably, the step of determining each first correction command based on each expected power and each real-time power, and superimposing each first correction command onto each power supply side for execution, includes: Based on the expected power and the real-time power, determine whether the difference between the actual output power of each power supply side and the real-time power of each power supply side is greater than the first preset value. When the difference between the expected power of each energy supply side and the real-time power of each energy supply side is less than or equal to the first preset value, at the next preset time point, the step of calculating the expected power of each energy supply side in the future time period through the optimization objective function based on the load prediction curve, the photovoltaic output prediction curve and the real-time electricity price, and generating the first operation instruction based on each expected power is executed. When the difference between the expected power of the power supply side and the real-time power of each power supply side is greater than the first preset value, the correction amount is calculated by incremental proportional calculation, the first correction instruction is determined, and it is superimposed on the operation of each power supply side until each expected power is equal to each real-time power.

[0008] Preferably, after determining the second correction command based on the expected voltage value and the real-time voltage value, and loading the second correction command onto the high-voltage DC bus for operation, the step includes: Based on the expected voltage value and the real-time voltage value, determine whether the difference between the expected voltage value and the real-time voltage value is greater than the second preset value; When the difference between the expected voltage value and the real-time voltage value is less than or equal to the second preset value, at the next preset time point, the steps of determining the expected voltage value of the high-voltage DC bus based on the load prediction curve and generating a second operating instruction based on the expected voltage value are executed. When the difference between the expected voltage value and the real-time voltage value is greater than the second preset value, the correction amount is calculated by incremental proportional calculation, the second correction command is determined, and it is superimposed on the operation of the high-voltage DC bus until the expected voltage value is equal to the real-time voltage value.

[0009] Preferably, the formula for calculating the load prediction curve is as follows: ; in, The load forecast value for the k-th future time period; This represents the historical load value for the same period in the k-th future time interval; Let be the power change of the i-th business change in the k-th future time period; This refers to the current time point; This represents the total number of business change projects. This refers to the sequence number of a future time period.

[0010] Preferably, the load prediction curve simultaneously satisfies the following constraints: ; as well as, ; in, This represents the maximum permissible change for a single correction.

[0011] Preferably, the calculation formula for the photovoltaic output prediction curve is as follows: ; in, This represents the predicted photovoltaic power output for the k-th future time period. This represents the current installed capacity of photovoltaic power. This refers to historical installed capacity. Historical photovoltaic power output curves for the same period; This is a correction factor.

[0012] Preferably, the calculation formula for the optimization objective function is as follows: ; in, Power drawn from the mains; Photovoltaic power utilization; Energy storage charging and discharging power; For real-time electricity prices; The conversion factor for photovoltaic power generation; This is the cost factor for energy storage lifespan. This is the weighting coefficient for the cost of mains electricity; This is the weighting coefficient for photovoltaic revenue; This is the weighting coefficient for energy storage costs; This represents the total number of future time periods.

[0013] Preferably, the step of determining the expected voltage value of the high-voltage DC bus based on the load prediction curve and generating a second operating command based on the expected voltage value includes: Determine the future load rate based on the load forecast curve; The expected voltage value of the high-voltage DC bus is dynamically calculated based on the piecewise function, and a second operating command is generated based on the expected voltage value.

[0014] Preferably, the formula for calculating the future load rate is as follows: ; in, Let be the load rate for the k-th future time period; This is the rated power.

[0015] Preferably, the calculation formula for the expected voltage value of the high-voltage DC bus based on the piecewise function is as follows: ; in, Let be the expected voltage value for the k-th future time period; This is the lower limit of the voltage of the high-voltage DC bus. This is the upper limit of the voltage of the high-voltage DC bus; This is the optimal voltage for the high-voltage DC bus operation; Minimum load rate; Maximum load rate; and All are set values; and All are adjustment coefficients.

[0016] Compared with the prior art, the present invention has at least the following beneficial effects: By introducing the grid side, photovoltaic side, and energy storage side, and based on the load forecast curve, photovoltaic output forecast curve, and implemented electricity price, the optimal operating strategy for the future time period is determined (i.e., generating the first operating instruction based on each expected power and the second operating instruction based on the expected voltage value). The power supply side and the high-voltage DC bus are operated with the optimal operating strategy. The first correction instruction and the second correction instruction are generated based on the actual operating conditions (i.e., real-time power and real-time voltage value) until the operating conditions of each power supply side and the high-voltage DC bus meet the optimal operating strategy, effectively utilizing green energy and ensuring the optimal efficiency of the high-voltage DC power supply system under all operating conditions. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of an embodiment of a high-voltage DC power supply system for green data centers.

[0019] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0020] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0021] The following description, with reference to the accompanying drawings, illustrates an embodiment of a high-voltage DC power supply system for green data centers.

[0022] Figure 1 This is a schematic diagram of an embodiment of a high-voltage DC power supply system for green data centers.

[0023] Please see Figure 1 To achieve the above objectives, the first embodiment of the present invention provides a high-voltage DC power supply system for green data centers, including an execution module, a computing module, and an acquisition module; The acquisition module is used to acquire the load forecast curve, photovoltaic output forecast curve, real-time electricity price, first preset value, second preset value, and optimization objective function respectively; The calculation module is used to calculate the expected power of each energy supply side in the future time period by optimizing the objective function based on the load forecast curve, the photovoltaic output forecast curve and the real-time electricity price, and generate the first operation command based on each expected power; it determines the expected voltage value of the high voltage DC bus based on the load forecast curve, and generates the second operation command based on the expected voltage value; wherein, each energy supply side includes the grid side, the photovoltaic side and the energy storage side. The execution module is used to run each power supply side according to the first running instruction and the second running instruction to run the high voltage DC bus, to obtain the real-time power of each power supply side and the real-time voltage value of the high voltage DC bus; to determine each first correction instruction according to each expected power and each real-time power, and to superimpose each first correction instruction onto each power supply side for execution; to determine the second correction instruction according to the expected voltage value and the real-time voltage value, and to load the second correction instruction onto the high voltage DC bus for execution.

[0024] By introducing the grid side, photovoltaic side, and energy storage side, and based on the load forecast curve, photovoltaic output forecast curve, and implemented electricity price, the optimal operating strategy for the future time period is determined (i.e., generating the first operating instruction based on each expected power and the second operating instruction based on the expected voltage value). The power supply side and the high-voltage DC bus are operated with the optimal operating strategy. The first correction instruction and the second correction instruction are generated based on the actual operating conditions (i.e., real-time power and real-time voltage value) until the operating conditions of each power supply side and the high-voltage DC bus meet the optimal operating strategy, effectively utilizing green energy and ensuring the optimal efficiency of the high-voltage DC power supply system under all operating conditions.

[0025] Specifically, both the first and second operating instructions are control instructions; the first correction instruction is a control instruction based on the first operating instruction; and the second correction instruction is a control instruction based on the second operating instruction. The mains power side is a mains rectifier, whose input is connected to the AC power grid, and whose output is connected to the high-voltage DC bus, used to convert AC power into high-voltage DC power. The photovoltaic side is a photovoltaic power generation device, whose output is connected to the high-voltage DC bus, used to voltage-match the DC power output from the photovoltaic power generation before feeding it into the high-voltage DC bus. The energy storage side is an energy storage device, connected to the high-voltage DC bus via a bidirectional DC / DC converter, used for storing and releasing electrical energy.

[0026] In the second embodiment of the present invention, based on the first embodiment, the step of determining each first correction command according to each expected power and each real-time power, and superimposing each first correction command onto each power supply side for execution includes: Step S10: Based on each expected power and each real-time power, determine whether the difference between the actual output power of each power supply side and the real-time power of each power supply side is greater than the first preset value. Step S11: When the difference between the expected power of each energy supply side and the real-time power of each energy supply side is less than or equal to the first preset value, at the next preset time point, the step of calculating the expected power of each energy supply side in the future time period by optimizing the objective function based on the load prediction curve, the photovoltaic output prediction curve and the real-time electricity price, and generating the first operation instruction based on each expected power is executed. Step S12: When the difference between the expected power of the power supply side and the real-time power of each power supply side is greater than the first preset value, the correction amount is calculated by incremental proportional calculation, the first correction command is determined, and it is superimposed on the operation of each power supply side until each expected power is equal to each real-time power.

[0027] A first preset value is set as the dead zone threshold for power deviation. When the deviation is small, no correction is triggered, avoiding frequent system adjustments due to minor fluctuations and ensuring stable system operation. An incremental proportional calculation of the correction amount is used (instead of directly outputting a new absolute command value). Each time, only the correction amount that needs to be added or subtracted from the original command is output, resulting in a smooth execution process without shock switching. After the correction amount is added to the power supply side, continuous monitoring continues. If the deviation still exists, correction continues until the expected power equals the real-time power, achieving complete elimination of the deviation. When the deviation is small, the step of calculating the expected power is re-executed at the next preset time point, ensuring that the system is always scheduled based on the latest forecast and electricity price data.

[0028] In the third embodiment of the present invention, based on the first embodiment, after determining a second correction command according to the expected voltage value and the real-time voltage value, and loading the second correction command onto the high-voltage DC bus for operation, the method includes: Step S13: Based on the expected voltage value and the real-time voltage value, determine whether the difference between the expected voltage value and the real-time voltage value is greater than the second preset value; Step S14: When the difference between the expected voltage value and the real-time voltage value is less than or equal to the second preset value, at the next preset time point, the step of determining the expected voltage value of the high-voltage DC bus based on the load prediction curve and generating the second operating command based on the expected voltage value is executed. Step S15: When the difference between the expected voltage value and the real-time voltage value is greater than the second preset value, the correction amount is calculated by incremental proportional calculation, the second correction command is determined, and it is superimposed on the operation of the high-voltage DC bus until the expected voltage value is equal to the real-time voltage value.

[0029] Operating independently in parallel with power correction, the voltage channel is dedicated to stabilizing the bus voltage. When a sudden load change causes a drop in bus voltage, the voltage correction channel responds independently, quickly pulling the voltage back to the expected value. Correction is not triggered when the voltage is within the allowable fluctuation range, avoiding frequent adjustments to the bus voltage due to minor fluctuations and ensuring power quality. Incremental proportional calculation of the voltage correction amount ensures smooth bus voltage regulation, preventing voltage overshoot or drops. When the voltage deviation is small, the step of determining the expected voltage value is re-executed at the next preset time point, ensuring that the bus voltage setpoint is always dynamically adjusted based on the latest load forecast.

[0030] Specifically, the formula for calculating the incremental proportional correction is as follows: ; in, This represents the correction amount for the k-th future time period; This represents the deviation for the k-th future time period; This represents the deviation in the (k-1)th future time period; This is the proportionality coefficient; is the integral coefficient.

[0031] The incremental proportional calculation of corrections relies solely on the current deviation and the previous deviation to calculate the correction, without accumulating all historical deviation data. It has advantages such as low computational cost, seamless switching, and resistance to integral saturation.

[0032] For example, if the expected voltage is 800V and the real-time voltage is 785V, the voltage deviation is +15V. The second preset value is ±11V, exceeding the second preset value by 4V. If the expected power on the photovoltaic side is 100kW and the real-time power on the photovoltaic side is 60kW, the deviation on the photovoltaic side is ±5kW. The first preset value is ±15kW, lower than the second preset value by 35kW. Incremental proportional calculation is initiated, with the voltage correction increasing by 10V and the photovoltaic side correction increasing by 35kW (supplemented from the mains power). The voltage correction is mapped to a second correction command and superimposed into the second operating command to generate the next corrected second operating command. Similarly, the photovoltaic side correction is mapped to a first correction command and superimposed into the first operating command to generate the next corrected first operating command. The green energy utilization rate on a typical workday reaches over 85%, and the average operating efficiency under all operating conditions reaches 97.3%, saving approximately 5.8% energy compared to traditional AC UPS solutions (efficiency approximately 92%).

[0033] In the fourth embodiment of the present invention, based on any one of the first to third embodiments, the calculation formula for the load prediction curve is as follows: ; in, The load forecast value for the k-th future time period. The final output load prediction curve, in kW; This represents the historical load value for the same period in the k-th future time period. The average value is obtained by aligning historical actual load data by time slot, and the unit is kW. This represents the power change for the i-th business change in the k-th future time period. Positive values ​​indicate increased load, and negative values ​​indicate decreased load, in kW (e.g., planning to add 100 x 64GB memory modules, each with a rated power of 15W, i.e., an increase of 1.5kW; planning to add 2 servers, each with a rated power of 500W, i.e., an increase of 1.0kW). This refers to the current time point, i.e., the starting time of the prediction; This represents the total number of business change projects. This is the sequence number of a future time period, typically set to 15 minutes per future time period.

[0034] In the fifth embodiment of the present invention, based on the fourth and third embodiments, the load prediction curve simultaneously satisfies the following constraints: ; as well as, ; in, This represents the maximum permissible change for a single correction.

[0035] In the sixth embodiment of the present invention, based on any one of the first to third embodiments, the calculation formula for the photovoltaic power output prediction curve is as follows: ; in, This represents the predicted photovoltaic power output for the k-th future time period. This represents the current installed capacity of photovoltaic power. This refers to historical installed capacity. Historical photovoltaic power output curves for the same period; This is a correction factor.

[0036] Specifically, the photovoltaic power output prediction curve satisfies the following constraints: .

[0037] In the seventh embodiment of the present invention, based on any one of the first to third embodiments, the calculation formula for the optimized objective function is as follows: ; in, Power drawn from the mains; Photovoltaic power utilization; Energy storage charging and discharging power; For real-time electricity prices; The conversion factor for photovoltaic power generation; This is the cost factor for energy storage lifespan. This is the weighting coefficient for the cost of mains electricity; This is the weighting coefficient for photovoltaic revenue; This is the weighting coefficient for energy storage costs; This represents the total number of future time periods.

[0038] The grid power consumption, photovoltaic power utilization, and energy storage charging / discharging power are all variable values. The expected power on the grid side is the optimal grid power consumption; the expected power on the photovoltaic side is the optimal photovoltaic power utilization; and the expected power on the energy storage side is the optimal energy storage charging / discharging power.

[0039] For example, during periods of abundant photovoltaic power (such as 12:00-14:00), the expected power on the photovoltaic side is high, so photovoltaic power is used first, and excess energy is stored on the energy storage side; during periods of insufficient photovoltaic power and peak electricity prices (such as 19:00-21:00), the expected power on the energy storage side is positive (discharging), supplementing the power supply gap; during periods of low electricity prices (such as 0:00-6:00), the expected power on the energy storage side is negative (charging), and the energy storage device is charged by the grid power.

[0040] Specifically, the objective function should satisfy the following constraints: ; in, The maximum charging power on the energy storage side is a negative value (e.g., as shown in the image). 100kW indicates that the maximum charging power is 100kW. This represents the maximum discharge power on the energy storage side, and is a positive value (e.g., 100kW means the maximum discharge power is 100kW).

[0041] In the eighth embodiment of the present invention, the step of determining the expected voltage value of the high-voltage DC bus based on any one of the first to third embodiments, according to the load prediction curve, and generating a second operating command based on the expected voltage value, includes: Determine the future load factor based on the load forecast curve; The expected voltage value of the high-voltage DC bus is dynamically calculated based on the piecewise function, and a second operating command is generated based on the expected voltage value.

[0042] In the ninth embodiment of the present invention, based on the eighth embodiment, the formula for calculating the future load rate is as follows: ; in, Let be the load rate for the k-th future time period; This is the rated power.

[0043] In the tenth embodiment of the present invention, based on the ninth embodiment, the calculation formula for the expected voltage value of the high-voltage DC bus dynamically calculated according to the piecewise function is as follows: ; in, Let be the expected voltage value for the k-th future time period; This is the lower limit of the voltage of the high-voltage DC bus. This is the upper limit of the voltage of the high-voltage DC bus; This is the optimal voltage for the high-voltage DC bus operation; Minimum load rate; Maximum load rate; and All are set values; and All are adjustment coefficients.

[0044] During light load periods (such as late at night, with a load rate of 15%), reduce the voltage of the high-voltage DC bus to [value missing]. To reduce converter switching losses, increase bus voltage appropriately during heavy load periods (such as peak daytime business hours, with a load rate of 85%) to reduce line losses.

[0045] In the description of this specification, references are made to the terms "one embodiment", "another embodiment", "other embodiments" Descriptions such as "example" or "first embodiment to Xth embodiment" refer to descriptions in conjunction with that embodiment or example. Specific features, structures, materials, or characteristics are included in at least one embodiment or example of the present invention.

[0046] In this specification, the illustrative expressions of the terms used above do not necessarily refer to the same embodiments or examples.

[0047] Furthermore, the specific features, structures, materials, method steps, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0048] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to... This encompasses non-exclusivity inclusion, thereby allowing a process, method, article, or device to include a range of elements. The setting includes not only those elements, but also other elements not explicitly listed, or may also include... Elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising a..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0049] The sequence numbers of the above embodiments of the present invention are merely for descriptive purposes and do not represent the superiority or inferiority of the embodiments. Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases, the former is a better implementation method. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0050] The embodiments of the present invention have been described above with reference to the accompanying drawings, but the present invention is not limited thereto. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art, guided by the teachings of this invention, will apply the principles and claims of this invention without departing from its spirit and scope. Within the scope of protection, many other forms can be made, all of which fall within the protection scope of this invention.

Claims

1. A high-voltage DC power supply system for green data centers, characterized in that, It includes an execution module, a calculation module, and an acquisition module; The acquisition module is used to acquire the load prediction curve, photovoltaic output prediction curve, real-time electricity price, first preset value, second preset value, and optimization objective function, respectively. The calculation module is used to calculate the expected power of each of the energy supply sides in the future time period through the optimization objective function based on the load prediction curve, the photovoltaic output prediction curve and the real-time electricity price, and generate the first operation instruction based on each expected power; determine the expected voltage value of the high-voltage DC bus based on the load prediction curve, and generate the second operation instruction based on the expected voltage value; wherein, each of the energy supply sides includes the grid power side, the photovoltaic side and the energy storage side; The execution module is used to run each of the power supply sides according to the first running instruction and the second running instruction to run the high-voltage DC bus, to obtain the real-time power of each of the power supply sides and the real-time voltage value of the high-voltage DC bus; to determine each of the first correction instructions according to each of the expected power and each of the real-time power, and to superimpose each of the first correction instructions onto each of the power supply sides for execution; to determine a second correction instruction according to the expected voltage value and the real-time voltage value, and to load the second correction instruction onto the high-voltage DC bus for execution.

2. The high-voltage DC power supply system for green data centers as described in claim 1, characterized in that, The step of determining each first correction command based on each expected power and each real-time power, and superimposing each first correction command onto each power supply side for execution, includes: Based on the expected power and the real-time power, determine whether the difference between the actual output power of each power supply side and the real-time power of each power supply side is greater than the first preset value. When the difference between the expected power of each energy supply side and the real-time power of each energy supply side is less than or equal to the first preset value, at the next preset time point, the step of calculating the expected power of each energy supply side in the future time period through the optimization objective function based on the load prediction curve, the photovoltaic output prediction curve and the real-time electricity price, and generating the first operation instruction based on each expected power is executed. When the difference between the expected power of the power supply side and the real-time power of each power supply side is greater than the first preset value, the correction amount is calculated by incremental proportional calculation, the first correction instruction is determined, and it is superimposed on the operation of each power supply side until each expected power is equal to each real-time power.

3. The high-voltage DC power supply system for green data centers as described in claim 1, characterized in that, After determining the second correction command based on the expected voltage value and the real-time voltage value, and loading the second correction command onto the high-voltage DC bus for operation, the following steps are included: Based on the expected voltage value and the real-time voltage value, determine whether the difference between the expected voltage value and the real-time voltage value is greater than the second preset value; When the difference between the expected voltage value and the real-time voltage value is less than or equal to the second preset value, at the next preset time point, the steps of determining the expected voltage value of the high-voltage DC bus based on the load prediction curve and generating a second operating instruction based on the expected voltage value are executed. When the difference between the expected voltage value and the real-time voltage value is greater than the second preset value, the correction amount is calculated by incremental proportional calculation, the second correction command is determined, and it is superimposed on the operation of the high-voltage DC bus until the expected voltage value is equal to the real-time voltage value.

4. A high-voltage DC power supply system for green data centers as described in any one of claims 1-3, characterized in that, The formula for calculating the load prediction curve is as follows: ； in, The load forecast value for the k-th future time period; This represents the historical load value for the same period in the k-th future time interval; Let be the power change of the i-th business change in the k-th future time period; This refers to the current time point; This represents the total number of business change projects. This refers to the sequence number of a future time period.

5. A high-voltage DC power supply system for green data centers as described in claim 4, characterized in that, The load prediction curve simultaneously satisfies the following constraints: ； as well as, ; in, This represents the maximum permissible change for a single correction.

6. A high-voltage DC power supply system for green data centers as described in any one of claims 1-3, characterized in that, The formula for calculating the photovoltaic power output prediction curve is as follows: ； in, This represents the predicted photovoltaic power output for the k-th future time period. This represents the current installed capacity of photovoltaic power. This refers to historical installed capacity. Historical photovoltaic power output curves for the same period; This is a correction factor.

7. A high-voltage DC power supply system for green data centers as described in any one of claims 1-3, characterized in that, The formula for calculating the objective function is as follows: ； in, Power drawn from the mains; Photovoltaic power utilization; Energy storage charging and discharging power; For real-time electricity prices; The conversion factor for photovoltaic power generation; This is the cost factor for energy storage lifespan. This is the weighting coefficient for the cost of mains electricity; This is the weighting coefficient for photovoltaic revenue; This is the weighting coefficient for energy storage costs; This represents the total number of future time periods.

8. A high-voltage DC power supply system for green data centers as described in any one of claims 1-3, characterized in that, The step of determining the expected voltage value of the high-voltage DC bus based on the load prediction curve and generating a second operating command based on the expected voltage value includes: Determine the future load rate based on the load forecast curve; The expected voltage value of the high-voltage DC bus is dynamically calculated based on the piecewise function, and a second operating command is generated based on the expected voltage value.

9. A high-voltage DC power supply system for green data centers as described in claim 8, characterized in that, The formula for calculating the future load rate is as follows: ； in, Let be the load rate for the k-th future time period; This is the rated power.

10. A high-voltage DC power supply system for green data centers as described in claim 9, characterized in that, The formula for calculating the expected voltage value of the high-voltage DC bus based on the piecewise function is as follows: ； in, Let be the expected voltage value for the k-th future time period; This is the lower limit of the voltage of the high-voltage DC bus. This is the upper limit of the voltage of the high-voltage DC bus; This is the optimal voltage for the high-voltage DC bus operation; Minimum load rate; Maximum load rate; and All are set values; and All are adjustment coefficients.

Images