Data center computing power energy storage cooling deep coupling source network load storage collaborative control method

Abstract

The application discloses a data center computing power energy storage cooling deep coupling source network load storage collaborative control method and relates to the technical field of data center energy efficiency, comprising the following steps: obtaining the multi-dimensional correlation characteristics of coupling analysis parameters by using a parameter coupling analysis method; constructing a collaborative control method by using a single parameter correlation method; and performing real-time control on various parameters in the data center by using the collaborative control method of all sub-parameters of the coupling analysis parameters; the application is used for solving the problem that in the existing source network load storage collaborative control method, only one kind or part of parameters is regulated in the parameter collaborative regulation, the system parameters are mutually separated and independently operated without linkage matching, the linkage compensation is not performed after the system parameters are adjusted, and the overall operation energy efficiency and economic benefits are reduced.

Description

Technical Field

[0001] This invention relates to the field of data center energy efficiency technology, specifically to a source-grid-load-storage collaborative control method that deeply couples data center computing power, energy storage, and cooling. Background Technology

[0002] Source-grid-load-storage coordinated control is a new power management mechanism that uses intelligent scheduling and technical means to achieve dynamic coordination among power sources, grids, loads, and energy storage in a power system. Source-grid-load-storage coordinated control, which deeply couples computing power, energy storage, and cooling in data centers, is an intelligent control mechanism that treats the computing tasks, energy storage, and heat dissipation systems of data centers as adjustable loads and achieves multi-dimensional coordination with the power sources and grids of the power system. It aims to improve energy efficiency, reduce carbon emissions, and enhance grid flexibility.

[0003] Existing methods for coordinated control of power generation, grid, load, and storage typically aim at intelligent scheduling of power sources and improving efficiency and flexibility during power supply. These methods integrate information such as load parameters, energy storage parameters, and distance of the power source, and construct logical filtering models to achieve rational energy allocation, avoid energy waste, and improve energy utilization. While this approach can optimize power source control through comprehensive analysis, adjusting only one or a few parameters in parameter coordination leads to fragmented and independent operation of system parameters without coordinated matching. This results in the inability of the power generation, grid, load, and storage scheduling to accurately adapt to fluctuations in the data center's electrical and thermal loads. Furthermore, the lack of coordinated compensation after parameter adjustments leads to a decrease in overall operational energy efficiency and economic benefits, as illustrated in patent publication CN118074226A. The application discloses a source-grid-load-storage coordinated control method for power distribution networks. This scheme comprehensively analyzes information such as load parameters, energy storage parameters, and distance of each power source, and uses a pre-trained logic screening model to achieve intelligent scheduling and matching of power sources, thereby improving power supply efficiency and flexibility. Other improvements to source-grid-load-storage coordinated control methods are usually improvements in scheduling scheme optimization. In terms of parameter coordinated regulation, they still cannot solve the problem of regulating only one or some parameters, resulting in the parameters of each system being isolated and operating independently without linkage and matching. This leads to the inability of source-grid-load-storage scheduling to accurately adapt to the fluctuations in the electrical and thermal loads of data centers. After the parameters of each system are adjusted, there is no linkage compensation, which in turn causes the overall operating energy efficiency to decline and the economic benefits to decrease. In view of this, it is necessary to improve the existing source-grid-load-storage coordinated control method. Summary of the Invention

[0004] This invention aims to at least partially solve one of the technical problems in the prior art. By proposing a source-grid-load-storage collaborative control method that deeply couples computing power, energy storage, and cooling in data centers, this invention addresses the issue that existing source-grid-load-storage collaborative control methods only regulate one or some parameters in terms of parameter coordination, resulting in the parameters of each system being isolated and operating independently without linkage and matching. This leads to the inability of source-grid-load-storage scheduling to accurately adapt to the fluctuations in the electrical and thermal loads of data centers, and the lack of linkage compensation after the adjustment of each system parameter, ultimately causing a decrease in overall operating energy efficiency and economic benefits.

[0005] To achieve the above objectives, this application provides a source-grid-load-storage coordinated control method for deep coupling of data center computing power, energy storage, and cooling, comprising the following steps:

[0006] Obtain historical data of coupling analysis parameters within the data center, and based on the historical data of each coupling analysis parameter, use the parameter coupling analysis method to obtain the multidimensional correlation features of each coupling analysis parameter;

[0007] Based on the multidimensional correlation characteristics of all coupling analysis parameters, a collaborative control method for the sub-parameters corresponding to each coupling analysis parameter is constructed using a single-parameter correlation method.

[0008] When controlling the data center, a collaborative control method based on all sub-parameters of all coupled analysis parameters is used to control various parameters within the data center in real time.

[0009] Furthermore, historical data of coupling analysis parameters within the data center are obtained, and based on the historical data of each coupling analysis parameter, the multidimensional correlation features of each coupling analysis parameter are obtained using the parameter coupling analysis method, including:

[0010] All parameters related to computing power, energy storage, cooling, energy, power grid, and load within the data center are denoted as coupling analysis parameters. Based on the historical operating data in the data, the historical data corresponding to each coupling analysis parameter is obtained.

[0011] For any coupling analysis parameter, the parameters that can be quantified in the historical data of the coupling analysis parameter are recorded as sub-parameters of the coupling analysis parameter; obtain all sub-parameters corresponding to each coupling analysis parameter.

[0012] Furthermore, parameter coupling analysis methods include:

[0013] For any coupling analysis parameter α, based on the sub-parameters of all coupling analysis parameters other than α, the one-way multidimensional analysis method is sequentially performed on each sub-parameter of the coupling analysis parameter α, and the one-way multidimensional features corresponding to each sub-parameter are obtained based on the results of the one-way multidimensional analysis method.

[0014] The unidirectional multidimensional features of all sub-parameters are denoted as the multidimensional correlation features of the coupling analysis parameter α.

[0015] Furthermore, unidirectional multidimensional analysis includes:

[0016] For any sub-parameter γ of the coupling analysis parameter α and any coupling analysis parameter β other than the coupling analysis parameter α, the sub-parameters of the coupling analysis parameter β are sequentially denoted as correlation analysis parameters GF1 to GF2. t ;

[0017] For any correlation analysis parameter GF r The sub-parameter γ is correlated with the correlation analysis parameter GF. r The corresponding units are denoted as Z1 and Z2, respectively, where r is a positive integer less than or equal to t and greater than or equal to 1; a Cartesian coordinate system is established with the unit of the horizontal coordinate being Z1 and the unit of the vertical coordinate being Z2, denoted as the correlation analysis coordinate system.

[0018] Furthermore, unidirectional multidimensional analysis also includes:

[0019] Record the sub-parameter γ and the correlation analysis parameter GF simultaneously from the historical data of the coupling analysis parameter α and the coupling analysis parameter β. r The time interval is denoted as the synchronic analysis interval, and multiple synchronic analysis intervals may exist.

[0020] For any synchronicity analysis interval, based on the parameter γ and the correlation analysis parameter GF at the same moment in the synchronicity analysis interval... r The recorded values ​​are plotted in the correlation analysis coordinate system to represent the parameters γ and GF within the synchronic analysis interval. r The relationship curve is denoted as the synchronic analysis curve;

[0021] Obtain the synchronicity analysis curves for all synchronicity analysis intervals, and denot the curve obtained by fitting all synchronicity analysis curves in the same correlation analysis coordinate system as the sub-parameter γ and the correlation analysis parameter GF. r The unidirectional characteristic curve.

[0022] Furthermore, unidirectional multidimensional analysis also includes:

[0023] Obtain the one-way characteristic curves of sub-parameter γ and all correlation analysis parameters, and denote the one-way characteristic curves of sub-parameter γ and all correlation analysis parameters as the one-way one-dimensional feature of sub-parameter γ and coupling analysis parameter β;

[0024] Obtain the one-way one-dimensional features of sub-parameter γ with all coupling analysis parameters except coupling analysis parameter α, and denote the one-way one-dimensional features of sub-parameter γ with all coupling analysis parameters except coupling analysis parameter α as the one-way multi-dimensional features of sub-parameter γ;

[0025] Obtain the unidirectional multidimensional features of all sub-parameters of the coupling analysis parameter α.

[0026] Furthermore, single-parameter correlation methods include:

[0027] For any sub-parameter γ of any coupling analysis parameter α, the set of units of the ordinates of all unidirectional feature curves in the unidirectional multidimensional features of sub-parameter γ is denoted as the associated feature set of sub-parameter γ.

[0028] The horizontal spans of all unidirectional feature curves in the unidirectional multidimensional features of sub-parameter γ are denoted as M1 to Mn, where n is the number of unidirectional feature curves in the unidirectional multidimensional features of sub-parameter γ; the average value of M1 to Mn is denoted as the control analysis span of sub-parameter γ.

[0029] Furthermore, single-parameter correlation methods also include:

[0030] Establish a Cartesian coordinate system with the horizontal axis in Z1 and the vertical axis in Ω, and denote it as the collaborative analysis coordinate system. Ω can be set to any unit in the associated feature set. Place all unidirectional feature curves of the unidirectional multidimensional features of the sub-parameter γ in the collaborative analysis coordinate system, and adjust the horizontal span of all unidirectional feature curves to control the analysis span based on curve fitting.

[0031] For any x-coordinate interval [X1, X2] in the co-analysis coordinate system, when all unidirectional characteristic curves in [X1, X2] are monotonically increasing curves, [X1, X2] is denoted as the positive growth interval of the sub-parameter γ; when all unidirectional characteristic curves in [X1, X2] are monotonically decreasing curves, [X1, X2] is denoted as the negative decreasing interval of the sub-parameter γ.

[0032] Obtain all unidirectional growth intervals and reverse decrease intervals of the sub-parameter γ within the co-analysis coordinate system.

[0033] Furthermore, the cooperative control method for sub-parameter γ includes:

[0034] When performing real-time control of the data center, if the value corresponding to sub-parameter γ is less than the standard threshold of sub-parameter γ, the value corresponding to sub-parameter γ is recorded as a real-time growth parameter; when the real-time growth parameter is within the positive growth range of sub-parameter γ, the parameters corresponding to all units in the associated feature set of sub-parameter γ are recorded as scalable parameters, and based on the data center's control method for all scalable parameters, the value of each scalable parameter is increased sequentially, where the standard threshold is the value of sub-parameter γ in the data center under standard operating conditions;

[0035] When the real-time growth parameter is in the reverse decreasing range of the sub-parameter γ, all parameters corresponding to all units in the associated feature set of the sub-parameter γ are recorded as reducible parameters, and the value of each reducible parameter is reduced in turn based on the data center's control method for all reducible parameters.

[0036] When the real-time growth parameter is outside the positive growth range and the negative decrease range, the straight line X = real-time growth parameter in the collaborative analysis coordinate system of sub-parameter γ is recorded as the adjustment baseline. For any unidirectional characteristic curve in the collaborative analysis coordinate system, the intersection of the unidirectional characteristic curve and the adjustment baseline is recorded as the adjustment intersection point. When the slope of the adjustment intersection point in the unidirectional characteristic curve is greater than 0, the value of the corresponding ordinate parameter is increased based on the data center's control method for the ordinate parameter of the unidirectional characteristic curve. When the slope of the adjustment intersection point in the unidirectional characteristic curve is less than 0, the value of the corresponding ordinate parameter of the unidirectional characteristic curve is decreased based on the data center's control method for the ordinate parameter of the unidirectional characteristic curve. When the slope of the adjustment intersection point in the unidirectional characteristic curve is equal to 0, the ordinate parameter of the unidirectional characteristic curve is not controlled.

[0037] Furthermore, the cooperative control method for sub-parameter γ also includes:

[0038] When performing real-time control of the data center, if the value corresponding to sub-parameter γ is greater than the standard threshold of sub-parameter γ, the value corresponding to sub-parameter γ is recorded as the real-time reduction parameter; when the real-time reduction parameter is within the positive growth range of sub-parameter γ, the parameters corresponding to all units in the associated feature set of sub-parameter γ are recorded as reducible parameters, and based on the data center's control method for all reducible parameters, the value of each reducible parameter is reduced in turn.

[0039] When the real-time reduction parameter is in the reverse reduction range of the sub-parameter γ, all parameters corresponding to all units in the associated feature set of the sub-parameter γ are recorded as scalable parameters, and the value of each scalable parameter is increased in turn based on the data center's control method for all scalable parameters.

[0040] When the real-time reduction parameter is outside the positive growth range and the negative reduction range, the straight line X = real-time reduction parameter in the collaborative analysis coordinate system of sub-parameter γ is recorded as the adjustment baseline. For any unidirectional characteristic curve in the collaborative analysis coordinate system, the intersection of the unidirectional characteristic curve and the adjustment baseline is recorded as the adjustment intersection point. When the slope of the adjustment intersection point in the unidirectional characteristic curve is greater than 0, the value corresponding to the ordinate parameter is reduced based on the data center's control method for the ordinate parameter of the unidirectional characteristic curve. When the slope of the adjustment intersection point in the unidirectional characteristic curve is less than 0, the value corresponding to the ordinate parameter of the unidirectional characteristic curve is increased based on the data center's control method for the ordinate parameter of the unidirectional characteristic curve. When the slope of the adjustment intersection point in the unidirectional characteristic curve is equal to 0, the ordinate parameter of the unidirectional characteristic curve is not controlled.

[0041] The beneficial effects of this invention are as follows: This application first obtains historical data of coupling analysis parameters within the data center, and then uses the parameter coupling analysis method to obtain the multidimensional correlation characteristics of each coupling analysis parameter based on the historical data of each coupling analysis parameter. The advantage of this is that by obtaining the historical data of coupling analysis parameters and using the parameter coupling analysis method to obtain the multidimensional correlation characteristics of coupling analysis parameters, the relationship between various system parameters can be effectively analyzed, thereby obtaining the parameter change relationship between a single system parameter and other system parameters. This is so that in subsequent analysis, when a single system parameter is adjusted, the direction in which other system parameters need to be adjusted can be obtained based on the multidimensional correlation characteristics, thereby realizing the linkage and matching of the operation of various system parameters and ensuring that the source-grid-load-storage scheduling is accurately adapted to the fluctuation of the data center's electric and heat load.

[0042] This application also constructs a collaborative control method for the sub-parameters corresponding to each coupling analysis parameter based on the multi-dimensional correlation characteristics of all coupling analysis parameters using a single-parameter correlation method. Finally, when controlling the data center, the various parameters within the data center are controlled in real time based on the collaborative control method of all sub-parameters of all coupling analysis parameters. The advantage of this is that by using the single-parameter correlation method to obtain the collaborative control method of the sub-parameters of each coupling analysis parameter, when the sub-parameters of the coupling analysis parameter are adjusted, the parameters in other system parameters that are related to the sub-parameters can be directly controlled based on the collaborative control method of the sub-parameters. This ensures that the system parameters can be compensated in a coordinated manner after adjustment, avoiding the problem of overall operational energy efficiency degradation caused by adjusting only a single parameter. Attached Figure Description

[0043] Figure 1 This is a flowchart of the steps of the method of the present invention;

[0044] Figure 2 This is a flowchart illustrating the collaborative control method for the sub-parameter γ of the present invention.

[0045] Figure 3 This is a schematic diagram of the electronic device of the present invention. Detailed Implementation

[0046] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0047] Example 1, please refer to Figure 1 As shown, this application provides a source-grid-load-storage coordinated control method for deep coupling of data center computing power, energy storage, and cooling, including the following steps:

[0048] Step S1: Obtain historical data of coupling analysis parameters within the data center, and based on the historical data of each coupling analysis parameter, use the parameter coupling analysis method to obtain the multidimensional correlation features of each coupling analysis parameter;

[0049] Step S1 includes: Step S101, recording all computing power parameters, energy storage parameters, cooling parameters, energy parameters, power grid parameters, and load parameters in the data center as coupling analysis parameters; and obtaining historical data corresponding to each coupling analysis parameter based on historical operating data in the data.

[0050] Step S102: For any coupling analysis parameter, record the parameters that can be quantized in the historical data of the coupling analysis parameter as sub-parameters of the coupling analysis parameter; obtain all sub-parameters corresponding to each coupling analysis parameter;

[0051] In the specific implementation process, the sub-parameters corresponding to computing power parameters, energy storage parameters, cooling parameters, energy parameters, power grid parameters, and load parameters can be specifically set according to the quantifiable parameters in the data center in the actual analysis. In the data analysis of this embodiment, for example, for computing power parameters, the quantifiable computing power parameters in the data center include: single rack computing power consumption, total cluster computing power active power, instantaneous computing power surge power, computing power concurrent scheduling, computing power load rate, computing power peak-valley load difference, computing power temperature rise rate, computing power steady-state operating temperature, number of online computing power clusters, and maximum adjustable load reduction ratio of computing power. In summary, all the computing power parameters mentioned above can be recorded as sub-parameters corresponding to computing power parameters.

[0052] Step S103, the parameter coupling analysis method includes: Step S1031, for any coupling analysis parameter α, based on the sub-parameters of all coupling analysis parameters other than coupling analysis parameter α, sequentially perform a one-way multidimensional analysis method on each sub-parameter of coupling analysis parameter α, and obtain the one-way multidimensional features corresponding to each sub-parameter based on the results of the one-way multidimensional analysis method;

[0053] Step S1032: Record the unidirectional multidimensional features of all sub-parameters as the multidimensional correlation features of the coupling analysis parameter α.

[0054] The one-way multidimensional analysis method includes: Step V1, for any sub-parameter γ of the coupling analysis parameter α and any coupling analysis parameter β other than the coupling analysis parameter α, the sub-parameters of the coupling analysis parameter β are sequentially denoted as correlation analysis parameters GF1 to GF2. t ;

[0055] Step V2, for any association analysis parameter GF r The sub-parameter γ is correlated with the correlation analysis parameter GF. r The corresponding units are denoted as Z1 and Z2, respectively, where r is a positive integer less than or equal to t and greater than or equal to 1; a Cartesian coordinate system is established with the unit of the horizontal axis being Z1 and the unit of the vertical axis being Z2, denoted as the correlation analysis coordinate system;

[0056] In the data analysis of this embodiment, for example, the coupling analysis parameter α and the coupling analysis parameter β are computing power parameters and energy storage parameters, respectively, and the sub-parameter γ is "single rack computing power consumption"; when the correlation analysis parameter is "real-time charging and discharging power" in the energy storage parameters, a correlation analysis coordinate system with the horizontal axis in kW and the vertical axis in kW can be constructed to obtain the unidirectional characteristic curves of "single rack computing power consumption" and "real-time charging and discharging power";

[0057] In the specific implementation process, if there is no quantitative relationship between the correlation analysis parameter and the sub-parameter γ, that is, the change of the sub-parameter γ is not affected by the correlation analysis parameter, the unidirectional characteristic curve of the sub-parameter γ and the correlation analysis parameter should not be obtained, so as to avoid performing erroneous control in the collaborative control method of the sub-parameter γ.

[0058] The one-way multidimensional analysis method also includes: step V3, which involves simultaneously recording the sub-parameter γ and the correlation analysis parameter GF in the historical data of the coupling analysis parameter α and the coupling analysis parameter β. r The time interval is denoted as the synchronic analysis interval, and multiple synchronic analysis intervals may exist.

[0059] Step V4: For any synchronicity analysis interval, based on the parameter γ and the correlation analysis parameter GF at the same time in the synchronicity analysis interval... rThe recorded values ​​are plotted in the correlation analysis coordinate system to represent the parameters γ and GF within the synchronic analysis interval. r The relationship curve is denoted as the synchronic analysis curve;

[0060] Step V5: Obtain the synchronicity analysis curves for all synchronicity analysis intervals, and denot the curve obtained by fitting all synchronicity analysis curves in the same correlation analysis coordinate system as the sub-parameter γ and the correlation analysis parameter GF. r The unidirectional characteristic curve.

[0061] The one-way multidimensional analysis method also includes: step V6, obtaining the one-way characteristic curve of sub-parameter γ and all correlation analysis parameters, and recording the one-way characteristic curve of sub-parameter γ and all correlation analysis parameters as the one-way one-dimensional feature of sub-parameter γ and coupling analysis parameter β;

[0062] In the data analysis of this embodiment, for example, after obtaining the unidirectional characteristic curves of "single rack computing power consumption" and "real-time charging and discharging power" in the above analysis, by obtaining the unidirectional characteristic curves corresponding to the quantified correlation sub-parameters of "single rack computing power consumption" and other sub-parameters of energy storage parameters, all the obtained unidirectional characteristic curves can be recorded as the unidirectional unit characteristics of "single rack computing power consumption" and energy storage parameters.

[0063] Step V7: Obtain the one-way one-dimensional features of sub-parameter γ and all coupling analysis parameters except coupling analysis parameter α, and record the one-way one-dimensional features of sub-parameter γ and all coupling analysis parameters except coupling analysis parameter α as the one-way multi-dimensional features of sub-parameter γ;

[0064] Step V8: Obtain the unidirectional multidimensional features of all sub-parameters of the coupling analysis parameter α.

[0065] Step S2: Based on the multidimensional correlation characteristics of all coupling analysis parameters, a collaborative control method for the sub-parameters corresponding to each coupling analysis parameter is constructed using a single-parameter correlation method.

[0066] The single-parameter association method includes: step S201, for any sub-parameter γ of any coupling analysis parameter α, the set of units of the ordinates of all unidirectional feature curves in the unidirectional multidimensional features of sub-parameter γ is denoted as the association feature set of sub-parameter γ.

[0067] Step S202: The horizontal spans of all unidirectional feature curves in the unidirectional multidimensional features of sub-parameter γ are denoted as M1 to Mn, where n is the number of unidirectional feature curves in the unidirectional multidimensional features of sub-parameter γ; the average value of M1 to Mn is denoted as the control analysis span of sub-parameter γ.

[0068] In the data analysis of this embodiment, for example, in a single data analysis, when the sub-parameter γ is "single rack computing power consumption", by obtaining the horizontal span of all unidirectional feature curves in the unidirectional multidimensional feature of "single rack computing power consumption", the obtained M1 to Mn are 9, 33, 14, 35, 10, 39, 8, 11, 31, 19, 13, 13, 18, 16, 35, 22, 35, 34, 37, 38, 36, 26, 10, 24, 30, 22, 26, 12, 9, and 21, that is, the value of n is 30; by calculation, the control analysis span of "single rack computing power consumption" is 22.87; by obtaining the control analysis of "single rack computing power consumption" The span ensures that when all unidirectional characteristic curves of "single rack computing power consumption" are placed in the same collaborative analysis coordinate system, they can be uniformly adjusted based on the control analysis span. This helps to avoid correlation distortion caused by inconsistent lateral spans in the collaborative control method of "single rack computing power consumption," and improves the accuracy of parameter linkage control and source-grid-load-storage collaborative control. If all unidirectional characteristic curves are not uniformly adjusted based on the control analysis span, there will be omissions in parameter control in the collaborative control method of "single rack computing power consumption," and other parameters cannot be fully linked and compensated after "single rack computing power consumption" is adjusted, resulting in an overall imbalance in parameter control.

[0069] The single-parameter association method also includes: step S203, establishing a Cartesian coordinate system with the unit of the horizontal axis being Z1 and the unit of the vertical axis being Ω, and denoting it as the collaborative analysis coordinate system, wherein Ω can be set as any unit in the association feature set; placing all unidirectional feature curves of the unidirectional multidimensional features of the sub-parameter γ in the collaborative analysis coordinate system, and adjusting the horizontal span of all unidirectional feature curves to control the analysis span based on curve fitting;

[0070] Step S204: For any abscissa interval [X1, X2] in the collaborative analysis coordinate system, when all unidirectional characteristic curves in [X1, X2] are monotonically increasing curves, [X1, X2] is recorded as the positive growth interval of sub-parameter γ; when all unidirectional characteristic curves in [X1, X2] are monotonically decreasing curves, [X1, X2] is recorded as the negative decreasing interval of sub-parameter γ.

[0071] Step S205: Obtain all unidirectional growth intervals and reverse decrease intervals of sub-parameter γ within the collaborative analysis coordinate system.

[0072] Please see Figure 2As shown, the collaborative control method for sub-parameter γ includes: Step Q1, when the data center is controlled in real time, if the value corresponding to sub-parameter γ is less than the standard threshold of sub-parameter γ, the value corresponding to sub-parameter γ is recorded as a real-time growth parameter; when the real-time growth parameter is within the positive growth range of sub-parameter γ, all parameters corresponding to all units in the associated feature set of sub-parameter γ are recorded as scalable parameters, and based on the data center's control method for all scalable parameters, the value of each scalable parameter is increased sequentially, wherein the standard threshold is the value of sub-parameter γ in the data center under standard operating conditions;

[0073] Step Q2: When the real-time growth parameter is in the reverse decreasing range of the sub-parameter γ, all parameters corresponding to all units in the associated feature set of the sub-parameter γ are recorded as reducible parameters, and the value of each reducible parameter is reduced in turn based on the data center's control method for all reducible parameters.

[0074] Step Q3: When the real-time growth parameter is outside the positive growth range and the negative decrease range, the straight line X = real-time growth parameter in the collaborative analysis coordinate system of sub-parameter γ is recorded as the adjustment baseline. For any unidirectional characteristic curve in the collaborative analysis coordinate system, the intersection of the unidirectional characteristic curve and the adjustment baseline is recorded as the adjustment intersection point. When the slope of the adjustment intersection point in the unidirectional characteristic curve is greater than 0, the value corresponding to the ordinate parameter is increased based on the data center's control method for the ordinate parameter of the unidirectional characteristic curve. When the slope of the adjustment intersection point in the unidirectional characteristic curve is less than 0, the value corresponding to the ordinate parameter of the unidirectional characteristic curve is decreased based on the data center's control method for the ordinate parameter of the unidirectional characteristic curve. When the slope of the adjustment intersection point in the unidirectional characteristic curve is equal to 0, the ordinate parameter of the unidirectional characteristic curve is not controlled.

[0075] In the data analysis of this embodiment, for example, the sub-parameter γ being analyzed is "single rack computing power consumption". When the data center is controlled in real time, the value corresponding to "single rack computing power consumption" is 60kW, and the standard threshold of "single rack computing power consumption" is 50kW. This indicates that the value corresponding to "single rack computing power consumption" should be reduced at this time. Therefore, 60kW should be recorded as the parameter to be reduced in real time.

[0076] The collaborative control method for sub-parameter γ also includes: step Q4, when the data center is controlled in real time, if the value corresponding to sub-parameter γ is greater than the standard threshold of sub-parameter γ, the value corresponding to sub-parameter γ is recorded as the real-time reduction parameter; when the real-time reduction parameter is within the positive growth range of sub-parameter γ, the parameters corresponding to all units in the associated feature set of sub-parameter γ are recorded as reducible parameters, and based on the control method of the data center for all reducible parameters, the value of each reducible parameter is reduced in turn;

[0077] Step Q5: When the real-time reduction parameter is in the reverse reduction range of the sub-parameter γ, all parameters corresponding to all units in the associated feature set of the sub-parameter γ are recorded as scalable parameters, and the value of each scalable parameter is increased in turn based on the data center's control method for all scalable parameters.

[0078] Step Q6: When the real-time reduction parameter is outside the positive growth range and the negative reduction range, the straight line X = real-time reduction parameter in the collaborative analysis coordinate system of sub-parameter γ is recorded as the adjustment baseline. For any unidirectional characteristic curve in the collaborative analysis coordinate system, the intersection of the unidirectional characteristic curve and the adjustment baseline is recorded as the adjustment intersection point. When the slope of the adjustment intersection point in the unidirectional characteristic curve is greater than 0, the value corresponding to the ordinate parameter is reduced based on the data center's control method for the ordinate parameter of the unidirectional characteristic curve. When the slope of the adjustment intersection point in the unidirectional characteristic curve is less than 0, the value corresponding to the ordinate parameter of the unidirectional characteristic curve is increased based on the data center's control method for the ordinate parameter of the unidirectional characteristic curve. When the slope of the adjustment intersection point in the unidirectional characteristic curve is equal to 0, the ordinate parameter of the unidirectional characteristic curve is not controlled.

[0079] In the data analysis of this embodiment, for example, the positive growth range and the negative reduction range corresponding to "single rack computing power consumption" in the above analysis are [10kW, 20kW] and [65kW, 70kW], respectively. Through analysis, it can be found that when the real-time reduction parameter is 60kW, the real-time reduction parameter is outside the positive growth range and the negative reduction range. Therefore, the straight line X=60kW in the collaborative analysis coordinate system of "single rack computing power consumption" should be obtained as the adjustment baseline. Then, for the intersection point of the unidirectional characteristic curve corresponding to "real-time charging and discharging power" with the vertical coordinate parameter in the collaborative analysis coordinate system and the adjustment point of X=60kW, if the slope of the unidirectional characteristic curve at the adjustment intersection point is greater than 0, it means that if "single rack computing power consumption" is to be reduced, the value corresponding to "real-time charging and discharging power" should be reduced simultaneously. Therefore, the value corresponding to "real-time charging and discharging power" should be reduced based on the data center's control method for "real-time charging and discharging power".

[0080] In the specific implementation process, when controlling the ordinate parameters of the data center that can be reduced, increased, or unidirectional characteristic curves, the parameters can be increased or decreased according to the actual working conditions of the data center and the preset quantization parameter thresholds.

[0081] Step S3: When controlling the data center, the various parameters within the data center are controlled in real time based on the collaborative control method of all sub-parameters of all coupled analysis parameters.

[0082] Example 2, please refer to Figure 3 As shown, Figure 3A schematic diagram of an electronic device is provided, which may include a processor, a communication interface, a memory, and a communication bus. The processor, communication interface, and memory communicate with each other via the communication bus. The memory stores computer-readable instructions, which the processor can call. When the processor executes a computer-readable instruction, it runs steps such as those in a source-grid-load-storage coordinated control method for deep coupling of computing power, energy storage, and cooling in a data center, to achieve the following functions: First, historical data of coupling analysis parameters within the data center is acquired. Based on the historical data of each coupling analysis parameter, a parameter coupling analysis method is used to obtain the multidimensional correlation characteristics of each coupling analysis parameter. Then, based on the multidimensional correlation characteristics of all coupling analysis parameters, a single-parameter correlation method is used to construct a coordinated control method for the sub-parameters corresponding to each coupling analysis parameter. Finally, when controlling the data center, the parameters within the data center are controlled in real time based on the coordinated control method for all sub-parameters of all coupling analysis parameters.

[0083] Furthermore, when the logical instructions in the aforementioned memory can be implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0084] Example 3: This application also provides a computer program product, which includes a computer program stored on a computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the source-grid-load-storage collaborative control method for deep coupling of data center computing power, energy storage, and cooling provided by the above methods. This method includes: firstly, acquiring historical data of coupling analysis parameters within the data center, and based on the historical data of each coupling analysis parameter, using a parameter coupling analysis method to obtain the multidimensional correlation characteristics of each coupling analysis parameter; then, based on the multidimensional correlation characteristics of all coupling analysis parameters, using a single-parameter correlation method to construct a collaborative control method for the sub-parameters corresponding to each coupling analysis parameter; finally, when controlling the data center, performing real-time control of various parameters within the data center based on the collaborative control method of all sub-parameters of all coupling analysis parameters.

[0085] Example 4: This application also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it runs the steps in the above-mentioned source-grid-load-storage collaborative control method for deep coupling of data center computing power, energy storage, and cooling, to achieve the following functions: First, it acquires historical data of coupling analysis parameters within the data center, and based on the historical data of each coupling analysis parameter, it uses the parameter coupling analysis method to obtain the multidimensional correlation characteristics of each coupling analysis parameter; then, based on the multidimensional correlation characteristics of all coupling analysis parameters, it uses the single-parameter correlation method to construct a collaborative control method for the sub-parameters corresponding to each coupling analysis parameter; finally, when controlling the data center, it uses the collaborative control method for all sub-parameters of all coupling analysis parameters to perform real-time control of various parameters within the data center.

[0086] Based on the above description of the embodiments, the embodiments of the present invention can be provided as methods, systems, or computer program products. Based on this understanding, the above technical solutions, in essence or in terms of their contribution to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or certain parts of the embodiments.

[0087] In the embodiments provided in this application, it should be understood that the disclosed system or method can be implemented in other ways. The embodiments described above are merely illustrative. For example, the division of modules or units is only a logical functional division, and there may be other division methods in actual implementation. Furthermore, multiple modules or units may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the coupling or direct coupling or communication connection shown or discussed may be through some communication interfaces. The indirect coupling or communication connection between systems, modules, and units may be electrical, mechanical, or other forms.

[0088] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A source-grid-load-storage coordinated control method for deep coupling of computing power, energy storage, and cooling in data centers, characterized in that: Includes the following steps: Obtain historical data of coupling analysis parameters within the data center, and based on the historical data of each coupling analysis parameter, use the parameter coupling analysis method to obtain the multidimensional correlation features of each coupling analysis parameter; Based on the multidimensional correlation characteristics of all coupling analysis parameters, a collaborative control method for the sub-parameters corresponding to each coupling analysis parameter is constructed using a single-parameter correlation method. When controlling the data center, a collaborative control method based on all sub-parameters of all coupled analysis parameters is used to control various parameters within the data center in real time.

2. The data center computing power, energy storage, and cooling deep coupling source-grid-load-storage coordinated control method according to claim 1, characterized in that, Historical data of coupling analysis parameters within the data center is obtained, and based on the historical data of each coupling analysis parameter, multidimensional correlation features of each coupling analysis parameter are obtained using parametric coupling analysis, including: All parameters related to computing power, energy storage, cooling, energy, power grid, and load within the data center are denoted as coupling analysis parameters. Based on the historical operating data in the data, the historical data corresponding to each coupling analysis parameter is obtained. For any coupling analysis parameter, the parameters that can be quantified in the historical data of the coupling analysis parameter are recorded as sub-parameters of the coupling analysis parameter; obtain all sub-parameters corresponding to each coupling analysis parameter.

3. The source-grid-load-storage coordinated control method for deep coupling of data center computing power, energy storage, and cooling, as described in claim 2, is characterized in that... Parametric coupling analysis methods include: For any coupling analysis parameter α, based on the sub-parameters of all coupling analysis parameters other than α, the one-way multidimensional analysis method is sequentially performed on each sub-parameter of the coupling analysis parameter α, and the one-way multidimensional features corresponding to each sub-parameter are obtained based on the results of the one-way multidimensional analysis method. The unidirectional multidimensional features of all sub-parameters are denoted as the multidimensional correlation features of the coupling analysis parameter α.

4. The data center computing power, energy storage, and cooling deep coupling source-grid-load-storage coordinated control method according to claim 3, characterized in that, One-way multidimensional analysis includes: For any sub-parameter γ of the coupling analysis parameter α and any coupling analysis parameter β other than the coupling analysis parameter α, the sub-parameters of the coupling analysis parameter β are sequentially denoted as correlation analysis parameters GF1 to GF2. t ; For any correlation analysis parameter GF r The sub-parameter γ is correlated with the correlation analysis parameter GF. r The corresponding units are denoted as Z1 and Z2, respectively, where r is a positive integer less than or equal to t and greater than or equal to 1; a Cartesian coordinate system is established with the unit of the horizontal coordinate being Z1 and the unit of the vertical coordinate being Z2, denoted as the correlation analysis coordinate system.

5. The data center computing power, energy storage, and cooling deep coupling source-grid-load-storage coordinated control method according to claim 4, characterized in that, One-way multidimensional analysis also includes: Record the sub-parameter γ and the correlation analysis parameter GF simultaneously from the historical data of the coupling analysis parameter α and the coupling analysis parameter β. r The time interval is denoted as the synchronic analysis interval, and multiple synchronic analysis intervals may exist. For any synchronicity analysis interval, based on the parameter γ and the correlation analysis parameter GF at the same moment in the synchronicity analysis interval... r The recorded values ​​are plotted in the correlation analysis coordinate system to represent the parameters γ and GF within the synchronic analysis interval. r The relationship curve is denoted as the synchronic analysis curve; Obtain the synchronicity analysis curves for all synchronicity analysis intervals, and denot the curve obtained by fitting all synchronicity analysis curves in the same correlation analysis coordinate system as the sub-parameter γ and the correlation analysis parameter GF. r The unidirectional characteristic curve.

6. The source-grid-load-storage coordinated control method for deep coupling of data center computing power, energy storage, and cooling, as described in claim 5, is characterized in that... One-way multidimensional analysis also includes: Obtain the one-way characteristic curves of sub-parameter γ and all correlation analysis parameters, and denote the one-way characteristic curves of sub-parameter γ and all correlation analysis parameters as the one-way one-dimensional feature of sub-parameter γ and coupling analysis parameter β; Obtain the one-way one-dimensional features of sub-parameter γ with all coupling analysis parameters except coupling analysis parameter α, and denote the one-way one-dimensional features of sub-parameter γ with all coupling analysis parameters except coupling analysis parameter α as the one-way multi-dimensional features of sub-parameter γ; Obtain the unidirectional multidimensional features of all sub-parameters of the coupling analysis parameter α.

7. The data center computing power, energy storage, and cooling deep coupling source-grid-load-storage coordinated control method according to claim 6, characterized in that, Single-parameter correlation methods include: For any sub-parameter γ of any coupling analysis parameter α, the set of units of the ordinates of all unidirectional feature curves in the unidirectional multidimensional features of sub-parameter γ is denoted as the associated feature set of sub-parameter γ. The horizontal spans of all unidirectional feature curves in the unidirectional multidimensional features of sub-parameter γ are denoted as M1 to Mn, where n is the number of unidirectional feature curves in the unidirectional multidimensional features of sub-parameter γ; the average value of M1 to Mn is denoted as the control analysis span of sub-parameter γ.

8. The data center computing power, energy storage, and cooling deep coupling source-grid-load-storage coordinated control method according to claim 7, characterized in that, Single-parameter correlation methods also include: Establish a Cartesian coordinate system with the horizontal axis in Z1 and the vertical axis in Ω, and denote it as the collaborative analysis coordinate system. Ω can be set to any unit in the associated feature set. Place all unidirectional feature curves of the unidirectional multidimensional features of the sub-parameter γ in the collaborative analysis coordinate system, and adjust the horizontal span of all unidirectional feature curves to control the analysis span based on curve fitting. For any x-coordinate interval [X1, X2] in the co-analysis coordinate system, when all unidirectional characteristic curves in [X1, X2] are monotonically increasing curves, [X1, X2] is denoted as the positive growth interval of the sub-parameter γ; when all unidirectional characteristic curves in [X1, X2] are monotonically decreasing curves, [X1, X2] is denoted as the negative decreasing interval of the sub-parameter γ. Obtain all unidirectional growth intervals and reverse decrease intervals of the sub-parameter γ within the co-analysis coordinate system.

9. The source-grid-load-storage coordinated control method for deep coupling of data center computing power, energy storage, and cooling as described in claim 8, is characterized in that... The cooperative control methods for sub-parameter γ include: When performing real-time control of the data center, if the value corresponding to sub-parameter γ is less than the standard threshold of sub-parameter γ, the value corresponding to sub-parameter γ is recorded as a real-time growth parameter; when the real-time growth parameter is within the positive growth range of sub-parameter γ, the parameters corresponding to all units in the associated feature set of sub-parameter γ are recorded as scalable parameters, and based on the data center's control method for all scalable parameters, the value of each scalable parameter is increased sequentially, where the standard threshold is the value of sub-parameter γ in the data center under standard operating conditions; When the real-time growth parameter is in the reverse decreasing range of the sub-parameter γ, all parameters corresponding to all units in the associated feature set of the sub-parameter γ are recorded as reducible parameters, and the value of each reducible parameter is reduced in turn based on the data center's control method for all reducible parameters. When the real-time growth parameter is outside the positive growth range and the negative decrease range, the straight line X = real-time growth parameter in the collaborative analysis coordinate system of sub-parameter γ is recorded as the adjustment baseline. For any unidirectional characteristic curve in the collaborative analysis coordinate system, the intersection of the unidirectional characteristic curve and the adjustment baseline is recorded as the adjustment intersection point. When the slope of the adjustment intersection point in the unidirectional characteristic curve is greater than 0, the value of the corresponding ordinate parameter is increased based on the data center's control method for the ordinate parameter of the unidirectional characteristic curve. When the slope of the adjustment intersection point in the unidirectional characteristic curve is less than 0, the value of the corresponding ordinate parameter of the unidirectional characteristic curve is decreased based on the data center's control method for the ordinate parameter of the unidirectional characteristic curve. When the slope of the adjustment intersection point in the unidirectional characteristic curve is equal to 0, the ordinate parameter of the unidirectional characteristic curve is not controlled.

10. The data center computing power, energy storage, and cooling deep coupling source-grid-load-storage coordinated control method according to claim 9, characterized in that, The cooperative control methods for sub-parameter γ also include: When performing real-time control of the data center, if the value corresponding to sub-parameter γ is greater than the standard threshold of sub-parameter γ, the value corresponding to sub-parameter γ is recorded as the real-time reduction parameter; when the real-time reduction parameter is within the positive growth range of sub-parameter γ, the parameters corresponding to all units in the associated feature set of sub-parameter γ are recorded as reducible parameters, and based on the data center's control method for all reducible parameters, the value of each reducible parameter is reduced in turn. When the real-time reduction parameter is in the reverse reduction range of the sub-parameter γ, all parameters corresponding to all units in the associated feature set of the sub-parameter γ are recorded as scalable parameters, and the value of each scalable parameter is increased in turn based on the data center's control method for all scalable parameters. When the real-time reduction parameter is outside the positive growth range and the negative reduction range, the straight line X = real-time reduction parameter in the collaborative analysis coordinate system of sub-parameter γ is recorded as the adjustment baseline. For any unidirectional characteristic curve in the collaborative analysis coordinate system, the intersection of the unidirectional characteristic curve and the adjustment baseline is recorded as the adjustment intersection point. When the slope of the adjustment intersection point in the unidirectional characteristic curve is greater than 0, the value corresponding to the ordinate parameter is reduced based on the data center's control method for the ordinate parameter of the unidirectional characteristic curve. When the slope of the adjustment intersection point in the unidirectional characteristic curve is less than 0, the value corresponding to the ordinate parameter of the unidirectional characteristic curve is increased based on the data center's control method for the ordinate parameter of the unidirectional characteristic curve. When the slope of the adjustment intersection point in the unidirectional characteristic curve is equal to 0, the ordinate parameter of the unidirectional characteristic curve is not controlled.

Images