Method and device for determining power distribution strategy and machine room energy management system

Abstract

The invention discloses a method and device for determining a power distribution strategy and a machine room energy management system. The method comprises the following steps: acquiring state data of a target machine room in a preset time period; determining an initial power distribution strategy of the target machine room in a preset time period according to the state data; dividing the preset time period into a plurality of electricity price time periods according to the actual electricity price value; and in each electricity price period, determining a control strategy corresponding to the electricity price period, and adjusting the initial power distribution strategy according to the control strategy to obtain a target power distribution strategy. The technical problem that energy is wasted or cost is increased due to the fact that a machine room energy management system adopted in the related technology depends on a preset rule or a simple optimization model and cannot cope with electricity price fluctuation and the like is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data center energy management, in particular to a power distribution strategy determination method and device and a machine room energy management system. BACKGROUND

[0002] With the rapid development of information technology and the continuous expansion of data center scale, the energy consumption problem of machine room as the core facility of data storage and processing is increasingly prominent. The load of the machine room usually remains stable, but as the business volume rises, the power consumption also increases sharply, leading to a continuous rise in the electricity cost of the machine room. However, the machine room energy management system adopted by the related technology relies on preset rules or simple optimization models, and cannot cope with situations such as electricity price fluctuations and load mutations, resulting in energy waste or cost increase.

[0003] For the above problems, no effective solution has been proposed so far. SUMMARY

[0004] The embodiments of the present application provide a power distribution strategy determination method, device and machine room energy management system to at least solve the technical problem that the machine room energy management system adopted by the related technology relies on preset rules or simple optimization models and cannot cope with situations such as electricity price fluctuations, resulting in energy waste or cost increase.

[0005] According to an aspect of an embodiment of the present application, a power distribution strategy determination method is provided, comprising: obtaining state data of a target machine room in a preset time period, wherein the target machine room includes a plurality of power supply modules, and the state data includes at least one of the following: an electricity price prediction value of a power grid supplying power to the target machine room, a load prediction value of the target machine room, and a power prediction value corresponding to each of the plurality of power supply modules; determining an initial power distribution strategy of the target machine room in the preset time period according to the state data, wherein the initial power distribution strategy is used to reflect the output power distribution of the plurality of power supply modules; dividing the preset time period into a plurality of electricity price periods according to actual electricity price values; in each electricity price period, determining a control strategy corresponding to the electricity price period, and adjusting the initial power distribution strategy according to the control strategy to obtain a target power distribution strategy, wherein the control strategy is used to reflect the power supply priority of the plurality of power supply modules in the electricity price period.

[0006] In some embodiments of the present application, determining the initial power distribution strategy of the target machine room in the preset time period according to the state data comprises: taking the minimization of the electricity cost of the target machine room as an optimization target, and determining a target function corresponding to the optimization target; determining a constraint condition corresponding to the optimization target, wherein the constraint condition includes a power balance constraint that the output power of the plurality of power supply modules is equal to the input power of the target machine room; and determining the power distribution strategy that satisfies the constraint condition and the target function as the initial power distribution strategy.

[0007] In some embodiments of the present application, the plurality of power supply modules includes an energy storage device module, wherein the energy storage device module supplies power to the target machine room using an energy storage battery; determining the constraint condition corresponding to the optimization target includes: obtaining a state of charge interval corresponding to the energy storage battery, and determining a state of charge constraint according to the state of charge interval; obtaining a charging power interval corresponding to the energy storage battery, and determining a charging power constraint according to the charging power interval; obtaining a discharging power interval corresponding to the energy storage battery, and determining a discharging power constraint according to the discharging power interval; determining the constraint condition corresponding to the optimization target according to the state of charge constraint, the charging power constraint and the discharging power constraint.

[0008] In some embodiments of the present application, the power allocation strategy that satisfies the constraint condition and the objective function is determined as the initial power allocation strategy, including: dividing a preset time period into a plurality of sub-time periods, and discretizing the state of charge of the energy storage battery within the range of the state of charge constraint according to a preset step size to obtain a plurality of state variables, wherein each state variable corresponds to a sub-time period; determining a target electricity cost corresponding to each state variable, wherein the target electricity cost includes a price cost and a discharging cost of the energy storage battery; determining a state transition equation according to the charging power constraint and the discharging power constraint, wherein the state transition equation is used to determine the state of charge of a second sub-time period in a first sub-time period, and the second sub-time period is the next sub-time period of the first sub-time period; determining the initial power allocation strategy according to the state transition equation, the target electricity cost, the state variable and the objective function.

[0009] In some embodiments of the present application, in each electricity price period, a control strategy corresponding to the electricity price period is determined, including: in the case that the electricity price period is a first electricity price period, determining a first power supply module that uses grid power supply as the highest power supply priority; in the case that the electricity price period is a second electricity price period, obtaining load information of the target machine room and power generation power of a second power supply module that uses photovoltaic power supply, and determining the power supply priority of the first power supply module and the second power supply module according to the load information and the power generation power, wherein the minimum electricity price of the second electricity price period is greater than the maximum electricity price of the first electricity price period; in the case that the electricity price period is a third electricity price period, determining the second power supply module as the highest power supply priority, wherein the minimum electricity price of the third electricity price period is greater than the maximum electricity price of the second electricity price period.

[0010] In some embodiments of the present application, the plurality of power supply modules includes a third power supply module that uses an energy storage battery for power supply; the method further includes: in the case that the electricity price period is a first electricity price period, charging the third power supply module; in the case that the electricity price period is a third electricity price period, the power supply priority of the third power supply module is greater than the power supply priority of the first power supply module.

[0011] In some embodiments of the present application, further comprising: in the case of powering the target machine room by the second power supply module, comparing the power generation of the second power supply module with the demand load of the target machine room to obtain a first comparison result; in the case that the first comparison result indicates that the power generation is greater than the demand load, storing the excess photovoltaic power corresponding to the power generation to the third power supply module; in the case that the first comparison result indicates that the power generation is less than or equal to the demand load, powering the target machine room by discharging of the third power supply module.

[0012] In some embodiments of the present application, before determining the initial power distribution strategy of the target machine room in the preset time period according to the state data, further comprising: determining the prediction error corresponding to the electricity price prediction value, the load prediction value and the power prediction value; comparing the prediction error with the first preset threshold and the second preset threshold respectively to obtain a second comparison result, wherein the second preset threshold is greater than the first preset threshold; in the case that the second comparison result indicates that the prediction error is greater than the first preset threshold and less than or equal to the second preset threshold, determining to continue to generate the initial power distribution strategy in the preset time period; in the case that the second comparison result indicates that the prediction error is greater than the second preset threshold, adjusting the real-time power distribution strategy corresponding to the current time.

[0013] According to another aspect of the embodiments of the present application, a machine room energy management system is further provided, comprising: a power supply module and a control module, wherein the control module is connected with the power supply module, and is configured to acquire state data of a target machine room in a preset time period, wherein the target machine room comprises a plurality of power supply modules, and the state data comprises at least one of the following: an electricity price prediction value of a power grid powering the target machine room, a load prediction value of the target machine room, and power prediction values corresponding to the plurality of power supply modules respectively; determine an initial power distribution strategy of the target machine room in the preset time period according to the state data, wherein the initial power distribution strategy is used to reflect the output power distribution of the plurality of power supply modules; divide the preset time period into a plurality of electricity price periods according to the actual electricity price value; in each electricity price period, determine a control strategy corresponding to the electricity price period, and adjust the initial power distribution strategy according to the control strategy to obtain a target power distribution strategy, wherein the control strategy is used to reflect the power supply priority of the plurality of power supply modules in the electricity price period; the power supply module is connected with the control module, and is configured to adjust the output power according to the target power distribution strategy.

[0014] In some embodiments of the present application, the power supply module includes a commercial power supply module, a photovoltaic power generation module, and an energy storage device module, wherein the commercial power supply module is connected to the control module and is configured to supply power to the target computer room using grid power; the photovoltaic power generation module is connected to the control module and is configured to convert solar energy into electrical energy and supply power to the target computer room using the electrical energy; and the energy storage device module is connected to the photovoltaic power generation module, the commercial power supply module, and the control module, and is configured to store electrical energy generated by the photovoltaic power generation module and / or charged by the commercial power supply module, and supply power to the target computer room through discharging.

[0015] In some embodiments of the present application, a terminal device module is further included, wherein the terminal device module is connected to the control module and is configured to display state information of the target computer room and receive optimization target setting information of the target object, wherein the state information includes energy usage and electricity cost of the target computer room, and the optimization target setting information is used to determine an initial power distribution strategy.

[0016] According to another aspect of the embodiments of the present application, a power distribution strategy determination device is further provided, which includes: an acquisition module configured to acquire state data of a target computer room in a preset time period, wherein the target computer room includes a plurality of power supply modules, and the state data includes at least one of the following: a price prediction value of a power grid supplying power to the target computer room, a load prediction value of the target computer room, and a power prediction value corresponding to each of the plurality of power supply modules; a distribution module configured to determine an initial power distribution strategy of the target computer room in the preset time period according to the state data, wherein the initial power distribution strategy is used to reflect output power distribution of the plurality of power supply modules; a division module configured to divide the preset time period into a plurality of price periods according to a price actual value; and a determination module configured to determine a control strategy corresponding to each of the price periods in each of the price periods, and adjust the initial power distribution strategy according to the control strategy to obtain a target power distribution strategy, wherein the control strategy is used to reflect power supply priorities of the plurality of power supply modules in the price periods.

[0017] According to still another aspect of the embodiments of the present application, an electronic device is further provided, which includes a memory and a processor, the memory is configured to store program instructions, and the processor is connected to the memory and is configured to execute the above-mentioned power distribution strategy determination method.

[0018] According to still another aspect of the embodiments of the present application, a non-volatile storage medium is further provided, which includes a stored computer program, wherein a device in which the non-volatile storage medium is located executes the above-mentioned power distribution strategy determination method by running the computer program.

[0019] According to still another aspect of the embodiments of the present application, a computer program product is provided, which comprises computer instructions, and the computer instructions, when executed by a processor, implement the method for determining a power distribution strategy.

[0020] In the embodiments of the present application, a data-driven manner is adopted, multi-source state data of a target machine room is integrated, and time periods are divided in combination with actual fluctuations of electricity prices. In different electricity price periods, a power distribution strategy is dynamically adjusted according to real-time power supply priorities, the purpose of intelligent optimization of energy configuration is achieved, and technical effects of minimization of electricity cost and improvement of energy utilization efficiency are achieved. Thus, the technical problem that a machine room energy management system in the related art relies on preset rules or a simple optimization model and cannot cope with electricity price fluctuations, leading to energy waste or cost increase, is solved. BRIEF DESCRIPTION OF DRAWINGS

[0021] The accompanying drawings, which are included to provide a further understanding of the present application, form a part of the present application and help explain the technical schemes of the illustrative embodiments of the present application and do not constitute improper limitations on the present application. In the drawings:

[0022] Figure 1 FIG. 1 is a hardware structure block diagram of a computer terminal for a method for determining a power distribution strategy according to an embodiment of the present application;

[0023] Figure 2 FIG. 2 is a flowchart of a method for determining a power distribution strategy according to an embodiment of the present application;

[0024] Figure 3 FIG. 3 is a schematic flowchart of an overall flow of a method for determining a power distribution strategy according to an embodiment of the present application;

[0025] Figure 4 FIG. 4 is a structure schematic diagram of a machine room energy management system according to an embodiment of the present application;

[0026] Figure 5 FIG. 5 is a structure schematic diagram of a device for determining a power distribution strategy according to an embodiment of the present application. DETAILED DESCRIPTION

[0027] In order to enable persons skilled in the art to better understand the present application, the technical schemes in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by persons skilled in the art without creative labor should fall within the scope of protection of the present application.

[0028] It should be noted that the terms "first", "second" and the like in the description and claims of the application and the above drawings are used to distinguish similar objects, and do not necessarily indicate a specific order or sequence. It should be understood that the data thus used can be interchanged, where appropriate, so that the embodiments of the application described herein can be carried out in other than the order shown or described herein. Furthermore, the terms "comprise" and "have", and any variations thereof, are intended to cover non-exclusive inclusion, for example, processes, methods, systems, products, or devices that include a series of steps or units without being limited to those clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products, or devices.

[0029] In order to save power resources, the machine room usually considers using renewable energy to reduce energy consumption and electricity cost at the initial stage of construction. Distributed photovoltaic power generation devices become the first choice for energy supplement of machine rooms due to their characteristics of being clean, renewable and easy to install. However, the energy consumption of the machine room has its particularity: on the one hand, the load of the machine room is relatively stable, but the photovoltaic power generation power is greatly affected by weather, season and other factors; on the other hand, the different time periods of the power grid price further increase the complexity of energy scheduling. At the initial stage of the construction of the machine room, due to the low load, the peak power of photovoltaic power generation may exceed the actual demand of the machine room, resulting in that the excess photovoltaic power cannot be effectively utilized, and even wasted.

[0030] The machine room energy management system adopted by the related art mostly uses a single algorithm or a fixed strategy, and has problems of insufficient dynamic adaptability, extensive energy storage life management, limited prediction accuracy and the like. First, the traditional system relies on preset rules or simple optimization models, and cannot cope with price fluctuations, photovoltaic output uncertainty and load mutations, resulting in energy waste or cost increase; second, it cannot intelligently schedule according to real-time information such as machine room load, photovoltaic power generation power and power grid price period; and third, it does not consider the influence of battery charge and discharge depth and frequency on the life, and long-term operation easily accelerates battery attenuation; in addition, it lacks the cooperative optimization of multi-dimensional prediction models, and the prediction error of price, load and photovoltaic power is large, which affects the accuracy of the optimization strategy. As can be seen, the energy management method adopted by the related art not only cannot fully develop the potential of renewable energy, but also easily leads to an increase in electricity cost and energy waste.

[0031] In order to solve the above technical problems, the embodiments of the present application provide corresponding solutions, which are described in detail below.

[0032] The power distribution strategy determination method embodiment provided by the embodiments of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Figure 1 A hardware structure block diagram of a computer terminal for implementing the power distribution strategy determination method is shown. As shown inFigure 1 As shown, the computer terminal 10 can include one or more processors (which can include, but are not limited to, processing devices such as microprocessors (MCU) or programmable logic devices (FPGA)), a memory 104 for storing data, and a transmission module 106 for communication function connected through wired and / or wireless network. In addition, it can also include a display, a keyboard, a cursor control device, an input / output interface (I / O interface), a universal serial bus (USB) port (which can be included as one of the ports of the I / O interface), a network interface, a BUS bus. Those skilled in the art can understand that Figure 1 The structure shown is only a schematic, which does not limit the structure of the above-mentioned electronic device. For example, the computer terminal 10 can also include more or less components than those shown in the figure, or have a different configuration from that shown in the figure. Figure 1 For example, the computer terminal 10 can also include more or less components than those shown in the figure, or have a different configuration from that shown in the figure. Figure 1 For example, the computer terminal 10 can also include more or less components than those shown in the figure, or have a different configuration from that shown in the figure.

[0033] It should be noted that the above-mentioned one or more processors and / or other data processing circuits can be referred to herein as "data processing circuits" in general. The data processing circuit can be embodied in whole or in part as software, hardware, firmware or any other combination. In addition, the data processing circuit can be a single independent processing module, or all or part of any one of the other elements combined into the computer terminal 10. As referred to in the embodiments of the present application, the data processing circuit serves as a processor control (for example, selection of a variable resistance terminal path connected to an interface).

[0034] The memory 104 can be used to store software programs and modules of application software, such as program instructions / data storage devices corresponding to the power allocation strategy determination method in the embodiments of the present application. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, implements the power allocation strategy determination method described above. The memory 104 can include a high-speed random access memory, and can also include a non-volatile memory such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 can further include a memory remotely disposed with respect to the processor, which can be connected to the computer terminal 10 through a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0035] The transmission module 106 is configured to receive or send data via a network. The network can include a wireless network provided by a communication provider of the computer terminal 10. In one example, the transmission module 106 includes a network interface controller (NIC) that can connect to other network devices through a base station to communicate with the Internet. In one example, the transmission module 106 can be a radio frequency (RF) module that is configured to communicate with the Internet wirelessly.

[0036] The display can be a liquid crystal display (LCD) that is touch screen, for example, which can enable a user to interact with a user interface of the computer terminal 10.

[0037] It is noted that in some alternative embodiments, the above Figure 1 The computer terminal can include hardware elements (including circuitry), software elements (including computer code stored on a computer readable media), or a combination of both hardware and software elements. It should be noted that in some embodiments, the computer terminal can be comprised of a combination of physical and virtual elements. Figure 1 is merely one example of a particular implementation and is intended to illustrate the types of components that can be present in the computer terminal.

[0038] In the above operating environment, the embodiment of the present application provides a method for determining a power distribution strategy. It is noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described herein can be executed in an order different from that shown herein.

[0039] Figure 2 is a flowchart of a method for determining a power distribution strategy according to an embodiment of the present application, as shown in the figure, the method includes the following steps: Figure 2

[0040] In step S202, state data of the target machine room in a preset time period is obtained, wherein the target machine room includes a plurality of power supply modules, and the state data includes at least one of the following: a power price prediction value of a power grid supplying power to the target machine room, a load prediction value of the target machine room, and a power prediction value corresponding to each of the plurality of power supply modules.

[0041] In step S202, the preset time period refers to a specific time interval for the system to make and optimize the power distribution strategy, such as a working day (24 hours), a week, or a custom time length. The state data is used to reflect the real-time and predicted running state information of the target machine room, including but not limited to the power price prediction value of the power grid, the load prediction value of the machine room, and the power prediction value of the plurality of power supply modules.​

[0042] In some embodiments of the present application, the power supply module can include a commercial power supply module, a photovoltaic power generation module, and an energy storage device module, which collectively provide energy support for the machine room.

[0043] In order to quickly respond to price fluctuations and load changes, the data acquisition unit in the intelligent control module (referred to as control module) can be used to monitor and record the state data of the power grid price, the machine room load, and the power supply module power in real time. In addition, the prediction unit can also use historical data, weather forecasts, business demand, and other information to generate predicted values of the power grid price, the machine room load, and the power supply module power using prediction models such as LSTM, Transformer, and Prophet.

[0044] Step S204, determining an initial power allocation strategy for the target machine room in a preset time period based on the state data, wherein the initial power allocation strategy is used to reflect the output power allocation of the plurality of power supply modules.

[0045] In the above step S204, the initial power allocation strategy is a power allocation plan prepared in advance for the energy demand of the machine room in the preset time period based on the current state data and prediction information, which reflects the power output ratio of the plurality of power supply modules (such as commercial power, photovoltaic, and energy storage) at different time points, and provides a preliminary framework for intelligent energy scheduling.

[0046] In some embodiments of the present application, the initial power allocation strategy for the target machine room in the preset time period can be determined by: taking minimizing the electricity cost of the target machine room as an optimization objective, and determining a target function corresponding to the optimization objective; determining a constraint condition corresponding to the optimization objective, wherein the constraint condition includes a power balance constraint that the output power of the plurality of power supply modules is equal to the input power of the target machine room; and determining a power allocation strategy that satisfies the constraint condition and the target function as the initial power allocation strategy.

[0047] Specifically, the optimization objective can be set as minimizing the electricity cost of the target machine room, that is, by intelligently scheduling the use of commercial power, photovoltaic, and energy storage devices, the goal is to meet the power demand of the machine room while reducing electricity costs. Correspondingly, the target function is used to reflect the calculation formula for minimizing the electricity cost.

[0048] In order to facilitate the understanding of the solving process of the initial power allocation strategy, the following takes the power supply module including a commercial power supply module, a photovoltaic power generation module, and an energy storage device module as an example for illustration. In some embodiments of the present application, the calculation formula of the target function can be:

[0049] [Formula 1]

[0050] Where C is the electricity cost, Pgrid,t P is the power supplied by the grid at time t, Price grid,t P is the price of the grid at time t, Price battery,discharge,t P is the discharging power of the energy storage device at time t, Price battery,discharge T is the total number of optimization periods (all optimization periods make up a preset period of time).

[0051] The constraint condition is a rule that must be met in the optimization problem. In some embodiments of the present application, the power balance constraint is to ensure that the total power supplied by the grid, the photovoltaic and the energy storage device is equal to the total load demand of the machine room, and its mathematical expression is:

[0052] [Formula 2]

[0053] P is the charging power of the energy storage device at time t, P PV,t P is the power supplied by the grid at time t, Price battery,charge,t P is the charging power of the energy storage device at time t.

[0054] It should be noted that the energy storage device module supplies power to the target machine room using the energy storage battery. On this basis, the constraint condition corresponding to the optimization target can be determined in the following manner: obtaining the state of charge interval corresponding to the energy storage battery, and determining the state of charge constraint according to the state of charge interval; obtaining the charging power interval corresponding to the energy storage battery, and determining the charging power constraint according to the charging power interval; obtaining the discharging power interval corresponding to the energy storage battery, and determining the discharging power constraint according to the discharging power interval; determining the constraint condition corresponding to the optimization target according to the state of charge constraint, the charging power constraint and the discharging power constraint.

[0055] Specifically, the setting of the energy storage device state constraint ensures that the energy storage battery group operates within a safe range, avoids excessive charging and discharging of the battery, and prolongs the service life of the battery. Its mathematical expression is:

[0056] SOC min ≤ SOC t ≤ SOC max [Formula 3]

[0057] P battery,charge,min ≤ P battery,charge,t ≤ P battery,charge,max [Formula 4]

[0058] P battery,discharge,min ≤ P battery,discharge,t ≤ P battery,discharge,max [Formula 5]

[0059] SOC min is the minimum state of charge of the energy storage device, SOC t is the state of charge of the energy storage device at time t, SOCmax Pmax is the maximum state of charge of the energy storage device battery,charge,min Pmin is the minimum power limit of the energy storage device when charging battery,charge,max Pmax is the maximum power limit of the energy storage device when charging battery,discharge,min Pmin is the minimum power limit of the energy storage device when discharging battery,discharge,max Pmax is the maximum power limit of the energy storage device when discharging.

[0060] After determining the objective function and constraints, advanced algorithms such as dynamic programming, genetic algorithm, particle swarm optimization, etc. can be used to solve the above optimization problem, for example, taking dynamic programming as an example, the basic idea is to convert the multi-stage decision problem into a series of single-stage optimal decision problems, and obtain the global optimal solution by step-by-step solving. In specific implementation, state variables (such as energy storage device state, time, etc.), decision variables (such as grid power supply power, energy storage device charging and discharging power, etc.), instantaneous cost (such as electricity cost, etc.) and state transition equation need to be defined, and the optimal power allocation scheme (i.e. initial power allocation strategy) is obtained by iteratively solving the dynamic programming equation.

[0061] In some embodiments of the present application, the initial power allocation strategy can be determined by the following method: dividing the preset time period into multiple sub-time periods, and discretizing the state of charge of the energy storage battery within the range of the state of charge constraint according to a preset step size to obtain multiple state variables, wherein each state variable corresponds to a sub-time period; determining the target electricity cost corresponding to each state variable, wherein the target electricity cost includes the electricity price cost and the discharging cost of the energy storage battery; determining the state transition equation according to the charging power constraint and the discharging power constraint, wherein the state transition equation is used to determine the state of charge of the second sub-time period in the first sub-time period, wherein the second sub-time period is the next sub-time period of the first sub-time period; determining the initial power allocation strategy according to the state transition equation, the target electricity cost, the state variable and the objective function.

[0062] Specifically, the dynamic programming algorithm can include the following steps:

[0063] (1) State space division:

[0064] SOC discretization: divided into 101 states (0-100%) according to 1% step size, each state variable corresponds to the state of charge of the energy storage device in a sub-time period. This discretization makes the dynamic programming algorithm able to find the optimal solution in a limited state space, while retaining the continuous change characteristics of the state of charge.

[0065] Time granularity: 15 minutes per period (96 periods T per day), i.e., a preset time period (e.g., 24 hours) is divided into 96 sub-periods according to a time granularity of 15 minutes, which enables the system to finely adjust the power output of the power supply module and improve the flexibility and accuracy of scheduling.

[0066] (2) Constraint processing:

[0067] Power balance: (corresponding to formula 2 above).

[0068] SOC constraint: 20%≤SOC≤90% (corresponding to formula 3 above).

[0069] Charge and discharge power limit: 0≤P ≤100kW, 0≤P ≤80kW (corresponding to formula 4 and formula 5 above, respectively).

[0070] Optimization objective function: (corresponding to formula 1 above).

[0071] (3) Initialization:

[0072] According to the total number of set optimization periods T, and the grid price P ricegrid,t , photovoltaic power prediction value P PV,t , and machine room load prediction value P load,t , etc.; initialize the state of the energy storage device, including the initial state of charge SOC0, the minimum state of charge SOC min , the maximum state of charge SOC max , and the minimum charging power P battery,charge,min , the maximum charging power P battery,charge,max , the minimum discharging power P battery,discharge,min , and the maximum discharging power P battery,discharge,max , etc.

[0073] (4) State definition and transition:

[0074] Define the state variable as the state of charge SOC t of the energy storage device and the time t, and the state transition equation represents the change of the state of charge of the energy storage device from time t to time t+1, i.e.:

[0075] [formula 6]

[0076] where η charge and η discharge are the charging efficiency and discharging efficiency of the energy storage device, respectively, and Δt is the time step.

[0077] (5) Decision variables and instant cost (i.e. target electricity cost) determination:

[0078] The decision variables are the grid power P grid,t , the energy storage device charging power P battery,charge,t and the discharging power P battery,discharge,t ; the instant cost is the electricity cost at time t, i.e.:

[0079] [Formula 7]

[0080] wherein Cost t is the instant cost.

[0081] (6) Recursive solution:

[0082] The recursive solution is calculated from time T in reverse, and the minimum instant cost and the corresponding optimal decision variables of each state SOC t at time t are calculated, wherein the recursive formula is:

[0083] [Formula 8]

[0084] wherein J t (SOC t ) represents the minimum instant cost when the state of charge of the energy storage device is SOC t at time t.

[0085] (7) Strategy execution:

[0086] On the forward time axis, the optimal decision variable sequence obtained by the reverse recursive solution is used to execute the power allocation strategy. In some embodiments of the present application, the intelligent control module can issue the prepared initial power allocation strategy to each module, and each module adjusts its power output according to the received instructions.

[0087] It should be noted that in addition to using the dynamic programming algorithm, a genetic algorithm, a particle swarm optimization algorithm, etc. can also be used to determine the initial power allocation strategy, which is not limited herein.

[0088] In some embodiments of the present application, a multi-algorithm collaborative optimization framework can also be constructed, and the global optimization layer and the real-time adjustment layer work collaboratively to achieve a balance between long-term planning and instant response. For example, the global optimization layer is responsible for searching for the global optimal solution based on the predicted data, and the real-time adjustment layer is responsible for rapid correction based on real-time data. Specifically:

[0089] (1) Global optimization layer:

[0090] The global optimization layer can use dynamic programming or genetic algorithms to search for a 24-hour-ahead global optimal solution, generate a baseline power allocation scheme for each sub-time period based on predicted grid electricity prices, data center loads, and photovoltaic power generation, so that the system can pre-plan an optimal power allocation strategy based on predicted data in scenarios where electricity prices fluctuate significantly and load curves are relatively stable, providing a baseline and starting point for subsequent real-time adjustments.

[0091] The task of the global optimization layer is to consider various constraints (such as power balance, SOC range limits) and optimization objectives (such as minimizing electricity costs, maximizing energy utilization efficiency) over a longer prediction period, and generate a globally optimal power allocation strategy, which helps the system make more reasonable energy scheduling decisions in the long term.

[0092] (2) Real-time adjustment layer:

[0093] The real-time adjustment layer can use particle swarm optimization or model predictive control algorithms to perform minute-level dynamic adjustments based on real-time monitoring of grid electricity prices, data center loads, and photovoltaic power generation data, to quickly respond to dramatic fluctuations in photovoltaic output and sudden changes in data center loads, generate corrected power allocation instructions, and ensure the real-time and adaptability of the strategy.

[0094] The real-time adjustment layer focuses on immediate cost control through real-time adjustment of power allocation for grid power, photovoltaic power, and energy storage devices, allowing the system to quickly respond to scenarios with dramatic photovoltaic fluctuations and sudden load changes, reducing energy waste and controlling electricity cost increases.

[0095] Before determining the initial power allocation strategy for the target data center in the preset time period based on the state data, the following steps can be performed: determining the prediction error corresponding to the electricity price prediction value, the load prediction value, and the power prediction value; comparing the prediction error with the first preset threshold and the second preset threshold to obtain a second comparison result, wherein the second preset threshold is greater than the first preset threshold; in the case where the second comparison result indicates that the prediction error is greater than the first preset threshold and less than or equal to the second preset threshold, determining to continue generating the initial power allocation strategy for the preset time period; in the case where the second comparison result indicates that the prediction error is greater than the second preset threshold, adjusting the real-time power allocation strategy corresponding to the current time.

[0096] In some embodiments of the present application, the intelligent control module includes a prediction unit for predicting the trends of grid electricity price, data center load and photovoltaic power generation power in a future period of time, and dynamically adjusting the power distribution strategy according to the prediction results, real-time data and algorithmic formula to adapt to the fluctuations of grid electricity price and data center load, for example, increasing the charging power of the energy storage device during the low electricity price period and increasing the discharging power of the energy storage device during the high electricity price period to reduce the electricity cost.

[0097] It should be noted that the prediction objects can include but are not limited to grid electricity price, data center load, and photovoltaic power generation power, and specifically as shown in Table 1 below:

[0098]

[0099] Table 1: Prediction model matrix table.

[0100] The prediction error refers to the difference between the predicted value and the actual value, which is used to measure the accuracy of the prediction, and can be calculated based on the comparison of historical predicted values and actual values, for example, statistical indicators such as mean square error (MSE) and mean absolute error (MAE) can be used to measure.

[0101] Comparing the prediction error with the first preset threshold and the second preset threshold respectively, the subsequent strategy making or adjustment can be determined according to the comparison result, for example, the first preset threshold is set to 5%, and the second preset threshold is set to 15%, when the prediction error is less than or equal to 5%, the system considers that the prediction is relatively accurate, and continues to execute the strategy generation of the global optimization layer; when the prediction error is greater than 5% and less than or equal to 15%, the system starts the real-time adjustment layer and prepares to adjust the initial strategy; when the prediction error is greater than 15%, the system directly adjusts the real-time power distribution strategy at the current time to cope with significant uncertainty.

[0102] In some embodiments of the present application, a three-stage adjustment strategy can also be based on the prediction error, specifically:

[0103] (1) Pre-adjustment stage (e.g. 4 hours in advance): generate a baseline scheme based on the global optimization result, the triggering condition of this stage can be, for example: prediction error > 10%.

[0104] (2) Real-time adjustment stage (real-time update): use PSO algorithm to correct the current period scheme, the triggering condition of this stage can be, for example: real-time data and prediction deviation > 15%.

[0105] (3) Emergency adjustment stage (second-level response): start security constrained optimal power flow (SCOPF), the triggering condition of this stage can be, for example: SOC < 20% or P load >110% rated value.

[0106] The three-stage strategy adjustment mechanism is a multi-level and multi-time scale dynamic response strategy for the data center energy management system to cope with uncertainties. The pre-adjustment stage focuses on the future and optimizes the strategy based on predicted data, which is suitable for long-term trend prediction. The real-time adjustment stage focuses on the present and adjusts the strategy to adapt to instantaneous market and environmental fluctuations through rapid response to real-time changes. The emergency adjustment stage is an emergency response measure when the system encounters extreme conditions, ensuring the stability of data center energy supply and the safety of the system, while minimizing economic losses as much as possible.

[0107] In step S206, the preset time period is divided into multiple electricity price periods according to the actual electricity price value.

[0108] In the above step S206, the electricity price period can be divided into valley, flat and peak electricity price periods according to different levels of actual electricity price value, each period corresponding to different electricity price levels and energy scheduling strategies.

[0109] In some embodiments of the present application, the data acquisition unit in the intelligent control module monitors the electricity price of the power grid in real time, and divides the preset time period into valley, flat and peak electricity price periods according to the actual fluctuations of the electricity price. For example, when the electricity price is lower than a certain percentage (such as 60%) of the average value, the period is defined as a valley electricity price period; when the electricity price fluctuates around the average value, it is a flat electricity price period; when the electricity price is higher than a certain percentage (such as 140%) of the average value, it is considered as a peak electricity price period.

[0110] Traditional electricity price period division may be too static and cannot adapt well to dynamic changes in electricity prices. To this end, the division standard of the electricity price period can also be dynamically adjusted based on the comparison and analysis of real-time electricity price data and historical electricity price data. For example, the system analyzes the electricity price fluctuation law based on historical data and sets a baseline electricity price; then, according to the comparison of real-time electricity price and baseline electricity price, the definition interval of valley, flat and peak is dynamically adjusted to capture the electricity price fluctuation characteristics more accurately.

[0111] In step S208, a control strategy corresponding to the electricity price period is determined in each electricity price period, and the initial power distribution strategy is adjusted according to the control strategy to obtain a target power distribution strategy, wherein the control strategy is used to reflect the power supply priority of the multiple power supply modules in the electricity price period.

[0112] In the above step S208, the control strategy is used to guide how to allocate power among the multiple power supply modules (mains, photovoltaic and energy storage) in a specific electricity price period to achieve cost minimization or target optimization. The control strategy reflects the power supply priority, such as charging energy storage in valley electricity price period and using photovoltaic and energy storage in peak electricity price period.

[0113] The target power allocation strategy includes a final power allocation strategy generated by the system after considering real-time electricity prices, prediction errors and control strategies, which guides the actual power output of the power supply module in each electricity price period, ensuring that the computer room load is met while the cost and emissions are optimized.

[0114] In some embodiments of the present application, the power supply priority can be dynamically adjusted based on real-time monitoring and prediction errors. For example, when the prediction error is less than a first preset threshold, the preset control strategy is maintained; when the prediction error is greater than the first preset threshold but less than a second preset threshold, the control strategy is fine-tuned according to real-time data, such as increasing the charging rate of energy storage during the valley electricity price period; when the prediction error is greater than the second preset threshold, emergency adjustment is immediately started to re-allocate power supply priority to ensure system stability and computer room load.

[0115] In each electricity price period, a control strategy corresponding to the electricity price period is determined. Specifically, in the case of the electricity price period being the first electricity price period, the first power supply module using grid power is determined as the highest power supply priority; in the case of the electricity price period being the second electricity price period, the load information of the target computer room and the power generation of the second power supply module using photovoltaic power are obtained, and the power supply priorities of the first power supply module and the second power supply module are determined according to the load information and the power generation, wherein the minimum electricity price of the second electricity price period is greater than the maximum electricity price of the first electricity price period; in the case of the electricity price period being the third electricity price period, the second power supply module is determined as the highest power supply priority, wherein the minimum electricity price of the third electricity price period is greater than the maximum electricity price of the second electricity price period.

[0116] It should be noted that the plurality of power supply modules can also include a third power supply module using energy storage battery power; in the case of the electricity price period being the first electricity price period, the third power supply module is charged; in the case of the electricity price period being the third electricity price period, the power supply priority of the third power supply module is higher than that of the first power supply module.

[0117] The first electricity price period refers to the period when the grid electricity price is the lowest, such as night or off-peak period, at this time the system sets the power supply priority of the mains (first power supply module) to the highest to fully utilize the low-cost power resources and reduce costs; the electricity price level of the second electricity price period is higher than that of the first period, but is still relatively low, at this time the system needs to dynamically adjust the power supply proportion of the mains and photovoltaic according to the real-time computer room load information and photovoltaic power generation (second power supply module) to find the best balance point between cost and green energy utilization; the third electricity price period is the period when the electricity price is the highest, such as peak electricity period, at this time the system sets the priority of photovoltaic power supply to the highest to use photovoltaic and energy storage devices (third power supply module, such as energy storage battery) as much as possible to avoid the use of high-priced mains and achieve cost minimization.

[0118] In some embodiments of the present application, the control strategy includes but is not limited to:

[0119] (1) Control strategy of the first electricity price period:

[0120] When the system detects that the current period belongs to the first electricity price period, the priority of the mains power supply is raised to the highest, and the mains power is preferentially used to supply power to the machine room, and the energy storage device is considered to be charged for subsequent peak electricity price period, specifically, the control method unit adjusts the power output of the mains power supply module, and sends an instruction to the charge and discharge control unit to control the energy storage device to charge.

[0121] By preferentially using the mains power in the first electricity price period, not only the electricity cost can be reduced, but also the energy storage device can be reserved for energy scheduling in subsequent periods.

[0122] (2) Control strategy of the second electricity price period:

[0123] In the second electricity price period, the system needs to dynamically adjust the power supply proportion of the mains and the photovoltaic according to the real-time machine room load and photovoltaic power generation, for example, the control method unit can set a threshold value, when the photovoltaic power generation is higher than a certain proportion of the machine room load, the photovoltaic power supply is preferentially used; when the photovoltaic power generation is insufficient, the mains is enabled to supplement, and the charge and discharge strategy of the energy storage is considered to maintain the continuity of power supply and the economy of cost.

[0124] In the period with relatively high electricity price, by flexibly adjusting the power supply proportion of the mains and the photovoltaic, the system can reduce the use of the mains while ensuring the demand of the machine room load, thereby reducing the electricity cost.

[0125] (3) Control strategy of the third electricity price period:

[0126] In the third electricity price period, since the electricity price reaches the highest, the system sets the priority of the photovoltaic power supply to the highest. If the photovoltaic power generation is insufficient to meet the machine room load, the system will release the energy from the energy storage device to avoid using the high-cost mains as much as possible. The control method unit closely monitors the photovoltaic output and the machine room load, and as soon as it is found that the photovoltaic output is insufficient, the energy storage discharge strategy is started immediately to ensure that the machine room load is met, while the electricity cost is controlled.

[0127] In the peak period of electricity price, by preferentially using the photovoltaic and energy storage energy, the system can greatly reduce the dependence on high-priced mains, thereby minimizing the electricity cost.

[0128] In some embodiments of the present application, the following steps can also be performed: in the case of powering the target machine room with the second power supply module, comparing the power generation of the second power supply module with the demand load of the target machine room to obtain a first comparison result; in the case where the first comparison result indicates that the power generation is greater than the demand load, storing the excess photovoltaic power corresponding to the power generation to the third power supply module; in the case where the first comparison result indicates that the power generation is less than or equal to the demand load, powering the target machine room by discharging the third power supply module.

[0129] Specifically, the excess photovoltaic power refers to the part of the photovoltaic power that exceeds the demand load of the machine room. This part of the power can be stored for subsequent use. By using the energy storage device (third power supply module), the excess photovoltaic power can be fully utilized, and the overall utilization rate of energy can be improved. For example, when the photovoltaic power is greater than the load of the machine room, the system stores the excess photovoltaic power in the energy storage device. When the photovoltaic power is insufficient or the load of the machine room increases, the energy storage device can release the stored power to power the machine room, thereby avoiding waste of energy.

[0130] It should be noted that in addition to adjusting the initial power distribution strategy according to the electricity price period, other adjustment strategies can also be used, including but not limited to:

[0131] (1) When the predicted photovoltaic fluctuation coefficient > 20kW, automatically switch to PSO algorithm (Particle Swarm Optimization).

[0132] The predicted photovoltaic fluctuation coefficient refers to the fluctuation range of the system's prediction of the photovoltaic power within a period of time. When the predicted photovoltaic power fluctuation exceeds 20kW, the optimization algorithm unit in the intelligent control module will automatically switch from the global optimization layer to the real-time adjustment layer, and use the particle swarm optimization (PSO) algorithm to dynamically adjust the power distribution strategy in the current period. This is because the PSO algorithm can quickly respond to high-frequency fluctuations and is suitable for real-time adjustment of power distribution to respond to sudden changes in photovoltaic power.

[0133] (2) When the electricity price prediction error > 15%, start rolling optimization (Model Predictive Control, abbreviated as MPC).

[0134] The electricity price prediction error refers to the deviation between the system's prediction of the electricity price and the actual electricity price. For example, when the prediction error exceeds 15%, the optimization algorithm unit will start the model predictive control (MPC) algorithm for rolling optimization. The MPC algorithm can continuously optimize the power distribution strategy within a future period of time based on real-time data and prediction information at the current time, in order to respond to the cost risk caused by inaccurate electricity price prediction.

[0135] In the case of large electricity price prediction errors, the MPC algorithm can continuously adjust the power allocation strategy for future periods through the rolling optimization mechanism, ensuring that even if the electricity price fluctuates beyond expectations, the system can respond in time to adjust the power supply priority and energy storage strategy to minimize electricity costs.

[0136] (3) When SOC is lower than 25%, preferentially execute the charging strategy.

[0137] SOC (State of Charge) refers to the state of charge of the energy storage device, i.e., the ratio of the remaining capacity to the maximum capacity of the battery. When the SOC is monitored to be lower than 25%, the control method unit in the system will preferentially execute the charging strategy, using the mains or photovoltaic power to charge the energy storage device, to avoid the low state of charge of the energy storage device affecting the power supply capacity in the subsequent peak period.

[0138] In the case of low SOC, the preferential charging strategy can ensure that the energy storage device stores as much energy as possible during the off-peak electricity price period to provide reserves for the subsequent peak electricity price period, avoiding additional use of mains and cost increases due to insufficient energy storage.

[0139] (4) When the mains price is 1.5 times higher than the energy storage discharge cost, suggest stopping mains power supply.

[0140] When the real-time monitored mains price is 1.5 times higher than the energy storage discharge cost, the system will send a suggestion to the operator through the user interface module in the intelligent control module to stop using mains power supply and instead use photovoltaic power and energy storage discharge to meet the data center load.

[0141] In the case of abnormally high mains price, the stop mains power supply strategy can avoid high electricity costs and reduce dependence on high-cost grid power.

[0142] Through the above steps S202 to S208, in a data-driven manner, by integrating the multi-source state data of the target data center and combining the actual fluctuations in electricity prices to divide the time periods, in different electricity price periods, the power allocation strategy is dynamically adjusted according to the real-time power supply priority, achieving the purpose of intelligent optimization of energy allocation, thereby realizing the technical effects of minimizing electricity costs and improving energy utilization efficiency, and further solving the technical problems of the related art, i.e., the dependence of the data center energy management system on pre-set rules or simple optimization models, which cannot cope with electricity price fluctuations, leading to energy waste or cost increases.

[0143] Figure 3 The overall flowchart of a method for determining a power allocation strategy according to an embodiment of the present application is shown in FIG. Figure 3 As shown in FIG.

[0144] Step S302: Prediction error analysis.

[0145] The data acquisition unit collects real-time data such as grid electricity price, data center load, photovoltaic power generation, and energy storage device status. The prediction unit predicts the future trend of electricity price, load, and photovoltaic power based on historical data and machine learning algorithms. The prediction error analysis compares real-time data with predicted data to calculate the prediction error, such as the deviation between predicted photovoltaic power and actual photovoltaic power, and the deviation between predicted electricity price and actual electricity price.

[0146] Step S304a: Global optimization when prediction error > 10% and ≤ 15%.

[0147] When the prediction error exceeds 10% but does not reach 15%, the system considers that global optimization is needed. At this time, the algorithm of the global optimization layer is used, such as dynamic programming (DP) or genetic algorithm (GA), to generate a benchmark power distribution scheme for the future period based on prediction data. The benchmark scheme details the power output of the grid, photovoltaic, and energy storage device in each period to minimize electricity cost.

[0148] Step S304b: Real-time adjustment when prediction error ≥ 15%.

[0149] Once the condition of prediction error ≥ 15% is triggered, the system will switch from the global optimization layer to the real-time adjustment layer, using faster response algorithms such as particle swarm optimization (PSO) or model predictive control (MPC) to make real-time adjustments to the power distribution strategy. These algorithms can quickly update the control strategy based on the latest real-time data to cope with significant deviations between predicted and actual values.

[0150] Step S306: Generate benchmark scheme.

[0151] The optimization algorithm unit generates a detailed benchmark power distribution scheme through iterative calculation, combining optimization objective functions and constraints, including specific power values for grid power supply, photovoltaic power generation, and energy storage device charging and discharging in each period. The scheme generation needs to ensure that the data center load demand is met while minimizing electricity cost and carbon emissions.

[0152] Step S308: Strategy execution.

[0153] The control execution unit sends control instructions to the grid power supply module, photovoltaic power generation module, and energy storage device module based on the generated power distribution strategy, including the benchmark scheme and real-time adjusted scheme, to execute the power distribution strategy in real time to meet the data center load demand and optimize energy use.

[0154] Step S310: Real-time monitoring.

[0155] The data acquisition unit continuously monitors real-time data such as grid electricity price, computer room load, photovoltaic power generation, and energy storage device status, providing the latest operation information for subsequent strategy adjustment and control.

[0156] Step S312: Constraint check.

[0157] The control method unit checks whether the real-time adjusted strategy meets the preset constraint conditions, such as whether the state of charge (SOC) of the energy storage device is within a safe range, and whether the power output of photovoltaic and energy storage is within the allowed range.

[0158] Step S314a: Emergency adjustment when exceeding the limit.

[0159] In the constraint check, if the strategy execution result exceeds an important constraint, such as the SOC being lower than the set lower limit (e.g., 20%) or higher than the upper limit (e.g., 80%), the system will immediately start the emergency adjustment strategy. For example, when the SOC is too low, the system prioritizes using the mains to charge the energy storage device to avoid system risks caused by excessive discharge of the energy storage device.

[0160] Step S314b: Data feedback when not exceeding the limit.

[0161] If the strategy does not exceed any important constraint, the system will feed back the detailed information of the current strategy execution to the optimization algorithm unit, including the actual power allocation result, SOC change, electricity cost, etc.

[0162] Step S316: Algorithm parameter update.

[0163] The strategy adjustment unit in the optimization algorithm unit updates the algorithm parameters, such as the objective function weight of dynamic programming, the selection probability of genetic algorithm, and the learning factor of particle swarm optimization, based on the collected strategy execution feedback data, to improve the adaptability and control accuracy of the algorithm.

[0164] Figure 4 is another structural schematic diagram of a computer room energy management system according to an embodiment of the present application, as shown in Figure 4 The system includes:

[0165] The power supply module 404 is connected with the control module 402. The control module is connected with the power supply module, is used for acquiring state data of the target machine room in a preset time period. The target machine room includes a plurality of power supply modules. The state data includes at least one of the following: the price prediction value of the power grid for supplying power to the target machine room, the load prediction value of the target machine room, and the power prediction value corresponding to each of the plurality of power supply modules. The initial power distribution strategy of the target machine room in the preset time period is determined according to the state data. The initial power distribution strategy is used to reflect the output power distribution of the plurality of power supply modules. The preset time period is divided into a plurality of price periods according to the actual price value. In each price period, the control strategy corresponding to the price period is determined, and the initial power distribution strategy is adjusted according to the control strategy to obtain the target power distribution strategy. The control strategy is used to reflect the power supply priority of the plurality of power supply modules in the price period. The power supply module is connected with the control module, and is used to adjust the output power according to the target power distribution strategy.

[0166] In some embodiments of the present application, the power supply module includes a commercial power supply module 404c, a photovoltaic power generation module 404a, and an energy storage device module 404b. The commercial power supply module is connected with the control module, and is used to supply power to the target machine room by using grid power. The photovoltaic power generation module is connected with the control module, and is used to convert solar energy into electrical energy and supply power to the target machine room by using the electrical energy. The energy storage device module is connected with the photovoltaic power generation module, the commercial power supply module, and the control module, and is used to store electrical energy generated by the photovoltaic power generation module and / or charged by the commercial power supply module, and supply power to the target machine room by discharging.

[0167] In some embodiments of the present application, a terminal device module 406 is further included. The terminal device module is connected with the control module, and is used to display state information of the target machine room and receive optimization target setting information of a target object. The state information includes energy use and electricity cost of the target machine room. The optimization target setting information is used to determine the initial power distribution strategy.

[0168] In some embodiments of the present application, the control module further includes:

[0169] (1) A data acquisition unit 402a is used to acquire data such as grid price, machine room load, photovoltaic power generation power, and energy storage device state in real time.

[0170] (2) An optimization algorithm unit 402b is used to calculate an optimal power distribution scheme of commercial power, photovoltaic power, and energy storage device according to the acquired data and a preset optimization target. Dynamic programming, genetic algorithm, and particle swarm optimization are integrated.

[0171] (3) Control execution unit 402c, according to the calculation result of the optimization algorithm unit, real-time adjustment of power supply module, photovoltaic power generation module and energy storage device module power output, to meet the load demand and reduce the cost of electricity.

[0172] (4) Control method unit 402d, implementation of specific control logic and strategy, including valley value electricity price period strategy, flat value electricity price period strategy and peak value electricity price period strategy, to adjust the power distribution of each module according to different electricity price period and load condition.

[0173] It should be noted that, Figure 4 The data room energy management system shown is used to execute Figure 2 The determination method of power distribution strategy shown, so Figure 2 The related explanation and description in the determination method of power distribution strategy in Figure 4 The data room energy management system shown, hereinafter will not be repeated.

[0174] Figure 5 It is a structure diagram of a power distribution strategy determination device according to an embodiment of the application, as shown in Figure 5 The device comprises:

[0175] The acquisition module 502 is used to acquire the state data of the target data room in the preset time period, wherein the target data room comprises a plurality of power supply modules, and the state data comprises at least one of the following: the electricity price prediction value of the power grid for supplying power to the target data room, the load prediction value of the target data room, and the power prediction value corresponding to each of the plurality of power supply modules;

[0176] The distribution module 504 is used to determine the initial power distribution strategy of the target data room in the preset time period according to the state data, wherein the initial power distribution strategy is used to reflect the output power distribution of the plurality of power supply modules;

[0177] The division module 506 is used to divide the preset time period into a plurality of electricity price periods according to the electricity price actual value;

[0178] The determination module 508 is used to determine the control strategy corresponding to each electricity price period in each electricity price period, and adjust the initial power distribution strategy according to the control strategy to obtain the target power distribution strategy, wherein the control strategy is used to reflect the power supply priority of the plurality of power supply modules in the electricity price period.

[0179] It should be noted that, Figure 5 The power distribution strategy determination device shown is used to execute Figure 2 The determination method of power distribution strategy shown, so Figure 2 The related explanation and description in the determination method of power distribution strategy in Figure 5The determination apparatus of the power distribution strategy is shown, and details are not repeated here.

[0180] The electronic device includes a memory and a processor, wherein the memory is configured to store program instructions; the processor is connected with the memory and is configured to execute steps of the method for determining the power distribution strategy in various embodiments of the present application.

[0181] The non-volatile storage medium includes a stored computer program, wherein a device in which the non-volatile storage medium is located executes steps of the method for determining the power distribution strategy in various embodiments of the present application by running the computer program.

[0182] The computer program product includes computer instructions, which, when executed by a processor, implement steps of the method for determining the power distribution strategy in various embodiments of the present application.

[0183] The computer program, when executed by a processor, implements steps of the method for determining the power distribution strategy in various embodiments of the present application.

[0184] The above sequence numbers of the embodiments of the present application are only for description, and do not represent advantages or disadvantages of the embodiments.

[0185] In the above embodiments of the present application, the description of each embodiment has its own focus, and the parts not described in detail in a certain embodiment can be referred to the related description of other embodiments.

[0186] In the several embodiments provided in the present application, it should be understood that the disclosed technology can be implemented in other ways. Of course, the device embodiment described above is only schematic. For example, the division of the units can be a logical function division, and there can be another division manner in actual implementation, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interface, unit or module, and can be electrical or other forms.

[0187] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, that is, they can be located in one place, or can be distributed on multiple units. According to actual needs, part or all of the units can be selected to achieve the purpose of the embodiment scheme.

[0188] In addition, each function unit in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software function unit.

[0189] When the integrated unit is realized in the form of a software function unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present application. The foregoing storage medium includes: a U disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a mobile hard disk, a magnetic disk or an optical disk, and various media that can store program codes.

[0190] The above is only the preferred embodiment of the present application, and it should be pointed out that, for those skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, and these improvements and refinements should be regarded as the protection scope of the present application.

Claims

1. A method for determining a power allocation strategy, characterized in that, The method comprises: acquiring state data of a target machine room in a preset time period, wherein the target machine room comprises a plurality of power supply modules, and the state data comprises at least one of the following: an electricity price prediction value of a power grid supplying power to the target machine room, a load prediction value of the target machine room, and a power prediction value corresponding to each of the plurality of power supply modules; determining an initial power distribution strategy of the target machine room in the preset time period according to the state data, wherein the initial power distribution strategy is used to reflect the output power distribution of the plurality of power supply modules; dividing the preset time period into a plurality of electricity price periods according to actual electricity prices; in each electricity price period, determining a control strategy corresponding to the electricity price period, and adjusting the initial power distribution strategy according to the control strategy to obtain a target power distribution strategy, wherein the control strategy is used to reflect the power supply priority of the plurality of power supply modules in the electricity price period.

2. The method of claim 1, wherein, determining an initial power distribution strategy of the target machine room in the preset time period according to the state data comprises: taking minimizing the electricity cost of the target machine room as an optimization target, and determining a target function corresponding to the optimization target; determining a constraint condition corresponding to the optimization target, wherein the constraint condition comprises a power balance constraint that the output power of the plurality of power supply modules is equal to the input power of the target machine room; determining a power distribution strategy that satisfies the constraint condition and the target function as the initial power distribution strategy.

3. The method of claim 2, wherein, The plurality of power supply modules comprises an energy storage device module, wherein the energy storage device module uses an energy storage battery to supply power to the target machine room; determining a constraint condition corresponding to the optimization target comprises: acquiring a state of charge interval corresponding to the energy storage battery, and determining a state of charge constraint according to the state of charge interval; acquiring a charging power interval corresponding to the energy storage battery, and determining a charging power constraint according to the charging power interval; acquiring a discharging power interval corresponding to the energy storage battery, and determining a discharging power constraint according to the discharging power interval; determining a constraint condition corresponding to the optimization target according to the state of charge constraint, the charging power constraint, and the discharging power constraint.

4. The method of claim 3, wherein, determining a power distribution strategy that satisfies the constraint condition and the target function as the initial power distribution strategy comprises: dividing the preset time period into a plurality of sub-time periods, and discretizing the state of charge of the energy storage battery within the range of the state of charge constraint according to a preset step size to obtain a plurality of state variables, wherein each state variable corresponds to a sub-time period; determining a target electricity cost corresponding to each state variable, wherein the target electricity cost comprises an electricity price cost and a discharging cost of the energy storage battery; determining a state transition equation according to the charging power constraint and the discharging power constraint, wherein the state transition equation is used to determine the state of charge of a second sub-time period in a first sub-time period, and the second sub-time period is the next sub-time period of the first sub-time period; Determine the initial power allocation strategy according to the state transition equation, the target electricity cost, the state variable, and the target function.

5. The method of claim 1, wherein, In each of the electricity price time periods, a control strategy corresponding to the electricity price time period is determined, including: In the case where the electricity price time period is a first electricity price time period, a first power supply module powered by a power grid is determined as the highest power supply priority; In the case where the electricity price time period is a second electricity price time period, load information of the target machine room and power generation power of a second power supply module powered by photovoltaic power are obtained, and the power supply priorities of the first power supply module and the second power supply module are determined according to the load information and the power generation power, wherein the minimum electricity price of the second electricity price time period is greater than the maximum electricity price of the first electricity price time period; In the case where the electricity price time period is a third electricity price time period, the second power supply module is determined as the highest power supply priority, wherein the minimum electricity price of the third electricity price time period is greater than the maximum electricity price of the second electricity price time period.

6. The method of claim 5, wherein, The plurality of power supply modules includes a third power supply module powered by an energy storage battery; the method further includes: In the case where the electricity price time period is the first electricity price time period, the third power supply module is charged; In the case where the electricity price time period is the third electricity price time period, the power supply priority of the third power supply module is greater than the power supply priority of the first power supply module.

7. The method of claim 6, wherein, The method further includes: In the case where the target machine room is powered by the second power supply module, the power generation power of the second power supply module is compared with the demand load of the target machine room to obtain a first comparison result; In the case where the first comparison result indicates that the power generation power is greater than the demand load, the remaining photovoltaic power corresponding to the power generation power is stored to the third power supply module; In the case where the first comparison result indicates that the power generation power is less than or equal to the demand load, the target machine room is powered by discharging of the third power supply module.

8. The method of claim 1, wherein, Before determining the initial power allocation strategy of the target machine room in the preset time period according to the state data, the method further includes: Determine a prediction error corresponding to the electricity price prediction value, the load prediction value, and the power prediction value; Compare the prediction error with a first preset threshold and a second preset threshold respectively to obtain a second comparison result, wherein the second preset threshold is greater than the first preset threshold; In the case where the second comparison result indicates that the prediction error is greater than the first preset threshold and less than or equal to the second preset threshold, it is determined to continue to generate the initial power allocation strategy in the preset time period; In the case where the second comparison result indicates that the prediction error is greater than the second preset threshold, adjust a real-time power allocation strategy corresponding to the current time.

9. A machine room energy management system, characterized by, Include: a power supply module and a control module, wherein The control module is connected with the power supply module, and is used to acquire state data of a target machine room in a preset time period, wherein the target machine room comprises a plurality of power supply modules, and the state data comprises at least one of the following: a price prediction value of a power grid supplying power to the target machine room, a load prediction value of the target machine room, and a power prediction value corresponding to each of the plurality of power supply modules; an initial power distribution strategy of the target machine room in the preset time period is determined according to the state data, wherein the initial power distribution strategy is used to reflect output power distribution of the plurality of power supply modules; the preset time period is divided into a plurality of price periods according to actual price values; in each of the price periods, a control strategy corresponding to the price period is determined, and the initial power distribution strategy is adjusted according to the control strategy to obtain a target power distribution strategy, wherein the control strategy is used to reflect power supply priorities of the plurality of power supply modules in the price period; The power supply module is connected with the control module, and is used to adjust output power according to the target power distribution strategy.

10. The system of claim 9, wherein, The power supply module comprises a commercial power supply module, a photovoltaic power generation module and an energy storage device module, wherein The commercial power supply module is connected with the control module, and is used to supply power to the target machine room by using grid power; The photovoltaic power generation module is connected with the control module, and is used to convert solar energy into electric energy, and supply power to the target machine room by using the electric energy; The energy storage device module is connected with the photovoltaic power generation module, the commercial power supply module and the control module respectively, and is used to store electric energy generated by the photovoltaic power generation module and / or charged by the commercial power supply module, and supply power to the target machine room by discharging.

11. The system of claim 9, wherein, Further comprising a terminal device module, wherein The terminal device module is connected with the control module, and is used to display state information of the target machine room, and receive optimization target setting information of a target object, wherein the state information comprises energy use and electricity cost of the target machine room, and the optimization target setting information is used to determine the initial power distribution strategy.

12. A device for determining a power allocation strategy, characterized in that, Comprise: An acquisition module is used to acquire state data of a target machine room in a preset time period, wherein the target machine room comprises a plurality of power supply modules, and the state data comprises at least one of the following: a price prediction value of a power grid supplying power to the target machine room, a load prediction value of the target machine room, and a power prediction value corresponding to each of the plurality of power supply modules; A distribution module is used to determine an initial power distribution strategy of the target machine room in the preset time period according to the state data, wherein the initial power distribution strategy is used to reflect output power distribution of the plurality of power supply modules; A division module is used to divide the preset time period into a plurality of price periods according to actual price values; The determining module is configured to determine a control strategy corresponding to each of the electricity price time periods, and adjust the initial power distribution strategy according to the control strategy to obtain a target power distribution strategy, wherein the control strategy is used to reflect the power supply priority of the plurality of power supply modules in the electricity price time period.

13. An electronic device, comprising: The method comprises the following steps: The memory is configured to store program instructions; and the processor is connected with the memory and configured to execute the method for determining the power distribution strategy according to any one of claims 1 to 8.

14. A non-volatile storage medium, comprising: The non-volatile storage medium comprises a stored computer program, wherein the device where the non-volatile storage medium is located executes the method for determining the power distribution strategy according to any one of claims 1 to 8 by running the computer program.

15. A computer program product comprising computer instructions, characterized in that, The computer instructions are executed by the processor to implement the method for determining the power distribution strategy according to any one of claims 1 to 8.

Images