A Data Center PUE Simulation Method and Platform Based on Dynamic Load Response

Abstract

A data center PUE simulation method and platform based on dynamic load response is disclosed. The method includes: dividing the computer room space into multiple load zones and determining local heat island zones based on temperature data within each load zone; converting the load of each load zone within the local heat island zone into regional cooling demand in chronological order to generate a temporal structure of regional cooling demand; deriving the operating intensity of air conditioning units in each load zone based on the temporal structure of regional cooling demand, and generating a regional cooling energy consumption sequence based on the operating intensity of the air conditioning units; generating a PUE curve based on the regional cooling energy consumption sequence and the time order, combined with the total energy consumption of the computer room at each moment, and the correspondence between the total energy consumption of the computer room and the regional cooling energy consumption; and correcting the PUE curve based on the regional cooling energy consumption deviations corresponding to different time periods in the PUE curve, effectively eliminating the overall underestimation caused by the averaging of regional cooling capacity.

Description

Technical Field

[0001] This application relates to the field of PUE simulation technology, and in particular to a data center PUE simulation method and platform based on dynamic load response. Background Technology

[0002] In large data centers, uneven rack distribution or inadequate hot and cold aisle isolation can lead to loads in certain areas far exceeding the average. Current solutions assume uniform cooling distribution across space and estimate unit start-up and shutdown status based on total load, neglecting to some extent the energy consumption differences in fans and chilled water branches caused by concentrated cooling in different areas. This is especially problematic in multi-story server rooms or modular containerized data centers, where localized heat island effects cause some air conditioners to operate at continuously high loads, while other areas remain undercooled. In such cases, PUE simulations struggle to accurately depict spatial energy consumption differences, resulting in an overall underestimate of the simulation value. Typically, precise correction requires establishing a three-dimensional spatial energy field coupling model, which is difficult to implement in existing engineering projects. Summary of the Invention

[0003] This application provides a data center PUE simulation method and platform based on dynamic load response, which solves the technical problem of simulation deviation caused by spatial mismatch of cooling demand in local heat island type computer room layout.

[0004] The present invention adopts the following technical solution.

[0005] The first aspect of this invention discloses a data center PUE simulation method based on dynamic load response, the method comprising:

[0006] The computer room space is divided into multiple load zones, and local heat island areas are determined based on the temperature data of each load zone;

[0007] The load of each load area in the local heat island region is converted into regional cooling demand in time sequence to generate a time sequence structure of regional cooling demand.

[0008] Based on the time sequence structure of the regional cooling demand, the operating intensity of the air conditioning units in each load area is derived, and a regional cooling energy consumption sequence is generated based on the operating intensity of the air conditioning units.

[0009] Based on the regional cooling energy consumption sequence and time order, and combined with the total energy consumption of the computer room at each time point, the correspondence between the total energy consumption of the computer room and the regional cooling energy consumption is calculated to generate the PUE curve.

[0010] The PUE curve is corrected based on the regional cooling energy consumption deviation corresponding to different time periods in the PUE curve to obtain the corrected PUE curve.

[0011] Furthermore, the process of dividing the computer room space into multiple load zones and determining local heat island areas based on temperature data within each load zone includes:

[0012] Obtain the distribution coordinates of server racks, air conditioners, and supply and return air paths within the server room, and divide the server room space into load areas based on the server rack distribution coordinates, air conditioner distribution coordinates, and supply and return air paths to obtain multiple load areas;

[0013] The cabinet power in each load area is summed at each time point, and a regional load time series is generated in chronological order based on the sum of the cabinet power in each load area.

[0014] Furthermore, the method of dividing the computer room space into multiple load zones and determining local heat island areas based on temperature data within each load zone also includes:

[0015] The temperature data of each load area at each time point is statistically analyzed, and the high temperature time window of the load area is determined based on the temperature data. The temperature data is the average value of the data collected by all temperature sensors in the corresponding load area.

[0016] Based on the high temperature time window and regional load time series of the load area, it is determined whether the corresponding load area meets the preset conditions, and the load area that meets the preset conditions is marked as the local heat island area.

[0017] The preset conditions are that the total power of the cabinet is higher than a first threshold for multiple consecutive times and the temperature data is higher than a second threshold for multiple consecutive times. The first threshold and the second threshold are the set upper limits for power and temperature, respectively.

[0018] Furthermore, the step of converting the load of each load area in the local heat island region into regional cooling demand in each time period according to time sequence to generate a time-series structure of regional cooling demand includes:

[0019] Obtain the air supply coverage area of ​​each air conditioner in the load area, and generate an air supply coverage matrix corresponding to each load area based on the air supply coverage area of ​​each air conditioner. At the same time, convert the air supply coverage matrix into the initial cooling capacity requirement of the area.

[0020] When the load area is located in the local heat island area, a heat compensation amount is added to the initial cooling demand of the area to obtain the area enhanced cooling demand by summing the initial cooling demand and the heat compensation amount.

[0021] Based on the time sequence and the air supply coverage area of ​​each air conditioner, the enhanced cooling demand of the region is integrated into a time sequence structure of the regional cooling demand.

[0022] Furthermore, the step of deriving the operating intensity of air conditioning units in each load area based on the temporal structure of the regional cooling demand, and generating a regional cooling energy consumption sequence based on the operating intensity of the air conditioning units, includes:

[0023] Based on the different height layers of the load area in the computer room space, the time sequence structure of the cooling demand of the area is divided into multiple height-layered cooling demand, and the effective fan operation intensity of the load area at each time is defined according to the height-layered cooling demand and the airflow delay coefficient.

[0024] The initial chilled water flow rate of the load area at each time moment is obtained, and the initial chilled water flow rate is adjusted in conjunction with the effective fan operating intensity to determine the regional chilled water flow rate.

[0025] The fan power component and the water pump power component are calculated based on the effective fan operating intensity and the regional chilled water flow rate, respectively. The invalid wind power component of the corresponding load area is defined based on the fan power component and the water pump power component to determine the regional cooling energy consumption.

[0026] The regional cooling energy consumption sequence is obtained by arranging the regional cooling energy consumption in chronological order, and the regional cooling energy consumption is the sum of the fan power component, water pump power component, and ineffective wind power component of the corresponding load area.

[0027] Furthermore, the step of calculating the correspondence between the total energy consumption of the data center and the regional cooling energy consumption based on the regional cooling energy consumption sequence and time order, combined with the total energy consumption of the data center at each time point, to generate a PUE curve, includes:

[0028] The regional cooling energy consumption of each load area is time-aligned, and the total cooling energy consumption of all load areas at a unified time is calculated based on the time-aligned regional cooling energy consumption. The energy consumption of IT equipment is then added to the total cooling energy consumption to obtain the total energy consumption of the computer room at each time.

[0029] Obtain the basic cooling energy consumption required for the IT equipment's energy consumption, and calculate the difference between the total cooling energy consumption and the basic cooling energy consumption to obtain the additional cooling energy consumption;

[0030] Based on the energy consumption of the IT equipment, the basic cooling energy consumption, and the additional cooling energy consumption, combined with the total energy consumption of the computer room, the PUE value at each time point is determined to generate a PUE curve of the PUE value over time.

[0031] Furthermore, the step of correcting the PUE curve based on the regional cooling energy consumption deviation corresponding to different time periods in the PUE curve to obtain the corrected PUE curve includes:

[0032] Calculate the spatial difference contribution value of each load area, and calculate the sum of the spatial difference contribution values ​​of all load areas based on the spatial difference contribution values ​​of each load area. Then, correct the additional cooling energy consumption by the sum of the spatial difference contribution values ​​of all load areas to obtain the corrected additional cooling component.

[0033] Based on the corrected additional cooling capacity component, combined with the IT equipment energy consumption and basic cooling capacity energy consumption, the PUE curve is corrected to output a corrected PUE curve.

[0034] A second aspect of this invention discloses a data center PUE simulation platform based on dynamic load response, used to implement the data center PUE simulation method based on dynamic load response as described in any one of the first aspects, the platform comprising:

[0035] The area division module is used to divide the computer room space into multiple load areas and determine local heat island areas based on the temperature data of each load area.

[0036] The cooling demand conversion module is used to convert the load of each load area in the local heat island area into regional cooling demand in each time period according to the time sequence, so as to generate a regional cooling demand time sequence structure.

[0037] The cooling energy consumption derivation module is used to derive the operating intensity of the air conditioning unit in each load area based on the time sequence structure of the regional cooling demand, and to generate a regional cooling energy consumption sequence based on the operating intensity of the air conditioning unit.

[0038] The PUE curve generation module is used to calculate the correspondence between the total energy consumption of the computer room and the regional cooling energy consumption based on the regional cooling energy consumption sequence and time sequence, combined with the total energy consumption of the computer room at each time point, so as to generate the PUE curve.

[0039] The PUE curve correction module is used to correct the PUE curve based on the regional cooling energy consumption deviation corresponding to different time periods in the PUE curve, so as to obtain the corrected PUE curve.

[0040] A third aspect of the present invention discloses a terminal, including a processor and a storage medium;

[0041] The storage medium is used to store instructions;

[0042] The processor is configured to operate according to the instructions to perform the steps of the method described in the first aspect.

[0043] A fourth aspect of the present invention discloses a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in the first aspect.

[0044] Compared with the prior art, this application has the following advantages:

[0045] (1) Based on the existing cabinet distribution, air conditioning distribution and supply and return air paths in the computer room, this invention divides the overall space of the computer room into multiple load areas. Each load area contains several cabinets and corresponding air conditioning vents, forming a regional load sequence. Local heat island areas are identified based on the sustained high temperature characteristics of temperature data within the area during the same time period. At the same time, by combining the regional air supply paths, air conditioning coverage and local heat island markers, the regional loads in each time period are transformed into a time sequence structure of regional cooling demand. By rearranging the regional cooling demand in chronological order according to the degree of heat concentration in the area, the cooling demand of high-load areas is significantly higher than that of ordinary areas, which facilitates accurate simulation of the spatial distribution of cooling demand in local heat island type computer room layouts.

[0046] (2) This invention derives the operating intensity of each air conditioning unit (including fan components and chilled water branches) based on the regional cooling demand. When the cooling demand of a certain region is high, the corresponding fan adjustment and chilled water branch flow rate will be increased, while the equipment in the region with lower demand will be in a low-load operating state. Finally, a regional cooling energy consumption sequence is generated, so that the cooling energy consumption of each region is directly bound to its corresponding cooling demand. In addition, combined with the global IT energy consumption at the corresponding time, the time correspondence between the global total energy consumption and the regional cooling energy consumption is calculated, and a dynamic PUE curve is generated. This makes PUE no longer dominated by a single cooling behavior, but by the accumulation of energy consumption in each region. Based on the correspondence between different time periods and regional cooling differences in the curve, the PUE is spatially corrected, so that the final PUE result not only reflects the temporal changes, but also reflects the spatial energy consumption differences. The corrected PUE curve can truly depict the energy consumption concentration phenomenon caused by high load in the heat island area, effectively eliminating the overall underestimation problem caused by the averaging of regional cooling. Attached Figure Description

[0047] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0048] Figure 1 This is a flowchart illustrating the data center PUE simulation method based on dynamic load response provided by the present invention.

[0049] Figure 2 This is a schematic diagram of the structure of the data center PUE simulation platform based on dynamic load response provided by the present invention.

[0050] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0051] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0052] like Figure 1 As shown, in one embodiment, a data center PUE simulation method based on dynamic load response includes the following steps:

[0053] Step S110: Divide the computer room space into multiple load areas and determine the local heat island area based on the temperature data of each load area.

[0054] In some embodiments, the data center PUE simulation method based on dynamic load response provided by the present invention includes the following steps in step S110:

[0055] Step S111: Obtain the rack distribution coordinates, air conditioner distribution coordinates, and supply and return air paths within the server room. Divide the server room space into load areas based on the rack distribution coordinates, air conditioner distribution coordinates, and supply and return air paths to obtain multiple load areas.

[0056] Step S112: At each time point, sum the cabinet power in each load area, and generate a regional load time series based on the total cabinet power in each load area in chronological order.

[0057] In some embodiments, the data center PUE simulation method based on dynamic load response provided by the present invention further includes the following steps in step S110:

[0058] Step S113: Calculate the temperature data of each load area at each time point, and determine the high temperature time window of the load area based on the temperature data. The temperature data is the average value of the data collected by all temperature sensors in the corresponding load area.

[0059] Step S114: Based on the high temperature time window of the load area and the load time series of the area, determine whether the corresponding load area meets the preset conditions, and mark the load area that meets the preset conditions as a local heat island area.

[0060] The preset conditions are that the total power of the cabinet is higher than a first threshold for multiple consecutive times and the temperature data is higher than a second threshold for multiple consecutive times. The first threshold and the second threshold are the set upper limits for power and temperature, respectively.

[0061] In a specific embodiment, the data center PUE simulation method based on dynamic load response provided by the present invention includes steps 1 to 5:

[0062] Step 1: Construct a regional load division and local heat island identification model.

[0063] Based on the existing rack distribution, air conditioning placement, and air supply and return organization in the data center, the overall space is divided into multiple load zones, each containing several racks and corresponding air conditioning vents. This step involves reorganizing the rack power monitoring records to form a zone-level load sequence, and identifying local heat island areas based on the sustained high-temperature characteristics of temperature sensors within the zone over the same time period. This includes the following sub-steps:

[0064] Sub-step 1.1: Establish the load region partitioning matrix.

[0065] Specifically, firstly, the coordinates of the server racks, air conditioners, and supply and return air paths within the data center are obtained. Based on these coordinates, the data center floor plan is divided into multiple load zones, forming a zone division matrix. Each element in the matrix corresponds to a zone number, with the zone number ranging from 1 to 20, which can be adjusted according to the size of the data center. The zone number for each coordinate point within the data center space is equal to the zone marker on the supply and return air path zoning map.

[0066] Sub-step 1.2 maps the cabinet power data to a regional load matrix.

[0067] Specifically, for each region number in the region division matrix, at each time point, the power values ​​of all cabinets belonging to that region are added together and arranged in chronological order to form a load region time series.

[0068] Sub-step 1.3: Extract regional temperature features and construct a high-temperature sequence.

[0069] Specifically, for each load area, the average value of all temperature sensor readings in that load area is calculated at each time point. This average value is the temperature characteristic of that load area. Finally, a high temperature time series of the area is generated in chronological order.

[0070] Sub-step 1.4: Generate heat island markers based on the continuous high temperature window.

[0071] Specifically, the high temperature time series and load area series of the region are jointly judged. When a region simultaneously meets the conditions of "the load remains high in a continuous time period (i.e. the total load exceeds the first threshold)" and "the temperature is continuously higher than the average value in the same time period (i.e. the temperature data exceeds the second threshold)", the load area is marked as a heat island area.

[0072] Step S120: Convert the load of each load area in the local heat island area into regional cooling demand in each time period according to the time sequence to generate a time sequence structure of regional cooling demand.

[0073] In some embodiments, the data center PUE simulation method based on dynamic load response provided by the present invention includes the following steps in step S120:

[0074] Step S121: Obtain the air supply coverage area of ​​each air conditioner in the load area, and generate the air supply coverage matrix corresponding to each load area based on the air supply coverage area of ​​each air conditioner. At the same time, convert the air supply coverage matrix into the initial cooling capacity requirement of the area.

[0075] Step S122: When the load area is located in the local heat island area, the heat compensation amount is added to the initial cooling demand of the area to obtain the area enhanced cooling demand by summing the initial cooling demand and the heat compensation amount.

[0076] Step S123: Based on the time sequence and the air supply coverage area of ​​each air conditioner, the regional enhanced cooling demand is integrated into a regional cooling demand time sequence structure.

[0077] In a specific embodiment, the data center PUE simulation method based on dynamic load response provided by the present invention includes step 2, which involves constructing a time-series structure of regional cooling demand based on regional load. The regional load sequence output in step 1 is input into the cooling demand calculation unit. Combined with the regional air supply path, air conditioning coverage area, and regional heat island markers, the regional load for each time period is transformed into a time-series structure of regional cooling demand. The core of this step is to rearrange the regional cooling demand in chronological order according to the degree of regional heat concentration, so that the cooling demand of high-load areas is significantly higher than that of ordinary areas. This includes the following sub-steps:

[0078] Sub-step 2.1: Construct the regional air supply coverage matrix.

[0079] Specifically, based on the number of each load area in the load area division matrix and the area that each air conditioner can cover in the area, the air supply coverage relationship of each air conditioner is mapped into an air supply coverage matrix. The elements in this matrix represent the air supply coverage area of ​​a certain number of air conditioners in a certain load area.

[0080] Sub-step 2.2: Construct the regional basic cooling demand matrix.

[0081] Specifically, based on the load area sequence matrix and the air supply coverage relationship of each air conditioner, the regional load is converted into the basic cooling demand of the corresponding area.

[0082] Sub-step 2.3: Generate an enhanced cooling demand matrix based on heat island markers.

[0083] Specifically, if the current area is a heat island area, then a heat compensation amount is added to the basic cooling demand in sub-step 2.2, so that the high-load area can obtain a higher cooling demand. The expression is:

[0084]

[0085] In the formula, The increased cooling demand for region q at time t; Let q be the basic cooling demand for region q at time t; The heat island enhancement factor for region q is 0.1-0.5.

[0086] Sub-step 2.4: Construct and output the time-series structure of regional cooling demand.

[0087] Specifically, considering the differences in the air supply paths of air conditioners in different regions, the enhanced cooling demand is converted into a regional cooling demand time sequence structure according to the coverage of the air conditioner. The regional cooling demand time sequence structure consists of the final cooling demand of each region at a certain moment. The final cooling demand is equal to the sum of the cooling provided by the air conditioner to the region and the enhanced cooling demand.

[0088] Step S130: Based on the time sequence structure of regional cooling demand, deduce the operating intensity of air conditioning units in each load area, and generate a regional cooling energy consumption sequence based on the operating intensity of air conditioning units.

[0089] In some embodiments, the data center PUE simulation method based on dynamic load response provided by the present invention includes the following steps in step S130:

[0090] Step S131: Based on the different height layers of the load area in the computer room space, the time sequence structure of the regional cooling demand is divided into multiple height-layered cooling demands, and the effective fan operation intensity of the load area at each time is defined according to the height-layered cooling demand and the airflow delay coefficient.

[0091] Step S132: Obtain the initial chilled water flow rate of the load area at each time, and adjust the initial chilled water flow rate in conjunction with the effective fan operating intensity to determine the regional chilled water flow rate.

[0092] Step S133: Calculate the fan power component and water pump power component based on the effective fan operating intensity and the regional chilled water flow rate, and define the invalid wind power component of the corresponding load area based on the fan power component and water pump power component to determine the regional cooling energy consumption.

[0093] Among them, the regional cooling energy consumption sequence is obtained by arranging the regional cooling energy consumption in chronological order. The regional cooling energy consumption is the sum of the fan power component, water pump power component, and ineffective wind power component of the corresponding load area.

[0094] In a specific embodiment, the data center PUE simulation method based on dynamic load response provided by the present invention includes step 3, which involves constructing a regional cooling equipment response model and generating a regional energy consumption sequence. The regional cooling demand time-series structure generated in step 2 is input into the cooling equipment response unit, and the operating intensity of each air conditioning unit (including fan components and chilled water branches) is derived based on the regional cooling demand. For example, when the cooling demand in a certain region is high, the corresponding fan adjustment and chilled water branch flow rate will be increased, while the equipment in regions with lower demand will operate under low load. This step ultimately generates a regional cooling energy consumption sequence, directly binding the cooling energy consumption of each region to its corresponding cooling demand, and includes the following sub-steps:

[0095] Sub-step 3.1: Construct a layered air supply response model and obtain the effective fan strength.

[0096] Specifically, the total cooling demand of a region is first broken down into stratified cooling demands based on the region's height. For example, the cooling demand of region q at time t at layer z is equal to the product of the proportion of cooling demand of region q at layer z and the total cooling demand. Then, using the airflow time delay coefficient, the effective response weights for each height stratification are constructed, expressed as:

[0097]

[0098] In the formula, Let be the effective response weights of region q at layer z. Let be the airflow time delay coefficient of region q at layer z. The larger, the better The smaller the value, the more difficult it is for the air conditioner to reach that area.

[0099] Finally, the effective fan operating intensity of the current load area at time t is defined. This effective fan operating intensity is the result of weighting the cooling demand of each height stratum according to the effective response weight.

[0100] Sub-step 3.2: Calculate the chilled water flow rate based on the effective fan strength and stratified cooling capacity requirements.

[0101] Specifically, since the chilled water flow rate must not only meet the total cooling capacity demand but also compensate for the ineffective airflow caused by difficulties in supplying air to higher floors, the basic chilled water flow rate component is defined as the product of the basic flow rate coefficient (valued at 0.6-1.2, used to convert cooling capacity demand into volumetric flow rate) and the sum of the stratified cooling capacity demands. Then, due to the linkage adjustment of the effective fan intensity on the chilled water flow rate, the corresponding linkage adjustment component (i.e., the chilled water linkage flow rate caused by the effective fan output) can be obtained, which is equal to the product of the linkage coefficient (valued at 0.1-0.3, used to characterize the increase in chilled water flow rate brought about by fan enhancement) and the effective fan operating intensity. Finally, by adding the basic chilled water flow rate component and the linkage adjustment component for the current load area, the final regional chilled water flow rate can be obtained.

[0102] Sub-step 3.3: Construct the operating power of the cooling equipment that takes into account the ineffective wind power.

[0103] Specifically, firstly, the power components of the fans and pumps are calculated. The fan power component equals the product of the effective fan operating intensity and the fan power coefficient (range 0.8-1.6), and the pump power component equals the product of the regional chilled water flow rate and the pump power coefficient (range 1.0-3.0). Next, to characterize the situation where some airflow circulates in the lower layers, causing ineffective work, an ineffective airflow component is introduced. This ineffective airflow component is determined based on the degree of deviation of stratified cooling demand from equilibrium. In determining the ineffective airflow component, the stratified cooling deviation degree for each region is first defined. This deviation is equal to the sum of the absolute values ​​of the differences between the average cooling demand of all strata at a certain height in that region at a certain moment. The stratified cooling deviation degree characterizes the degree of deviation between the cooling demand of each stratum and the average cooling demand. The larger the value, the more uneven the distribution of heat and cold, and the more likely the lower layers will be overcooled and the upper layers overheated. Next, the ineffective airflow component is defined as the product of the ineffective airflow coefficient (ranging from 0.05 to 0.5, used to convert deviation into power loss) and the stratified cooling capacity deviation. Finally, the sum of the fan power component, chilled water pump power component, and ineffective airflow component is calculated to obtain the operating power of the cooling equipment in the current area.

[0104] Sub-step 3.4 generates the regional cooling energy consumption sequence.

[0105] Specifically, based on a set time step (e.g., 5 min or 15 min), the operating power of the cooling equipment is converted into a time-sequenced sequence of regional cooling energy consumption.

[0106] Step S140: Based on the regional cooling energy consumption sequence and time order, and combined with the total energy consumption of the computer room at each time point, the correspondence between the total energy consumption of the computer room and the regional cooling energy consumption is used to generate the PUE curve.

[0107] In some embodiments, the data center PUE simulation method based on dynamic load response provided by the present invention includes the following steps in step S140:

[0108] Step S141: Perform time alignment processing on the regional cooling energy consumption of each load area, calculate the total cooling energy consumption of all load areas at a unified time based on the time-aligned regional cooling energy consumption, and add the IT equipment energy consumption to the total cooling energy consumption to obtain the total energy consumption of the computer room at each time.

[0109] Step S142: Obtain the basic cooling energy consumption required for the IT equipment's energy consumption, and calculate the difference between the total cooling energy consumption and the basic cooling energy consumption to obtain the additional cooling energy consumption.

[0110] Step S143: Based on the energy consumption of IT equipment, basic cooling energy consumption, and additional cooling energy consumption, combined with the total energy consumption of the computer room, determine the PUE value at each time point to generate a PUE curve of PUE value with respect to time.

[0111] In a specific embodiment, the data center PUE simulation method based on dynamic load response provided by the present invention includes step 4, which involves constructing a global energy consumption synthesis structure based on the regional energy consumption sequence and generating a dynamic PUE curve. The regional cooling energy consumption sequence output in step 3 is synthesized in the time direction, i.e., the energy consumption of all regions is added together at each time point, and combined with the global IT energy consumption at the corresponding time point, the time correspondence between the global total energy consumption and the regional cooling energy consumption is calculated. This generates a dynamic PUE curve, so that PUE is no longer dominated by a single cooling behavior, but is obtained by the accumulation of energy consumption from each region, including the following sub-steps:

[0112] Sub-step 4.1: Construct time-aligned regional cooling energy consumption and obtain the global cooling energy consumption sequence.

[0113] Specifically, firstly, the cooling energy consumption of each region is time-aligned to match the actual cooling capacity that the IT load can utilize at a certain moment. After time alignment, the cooling energy consumption of all regions at the same time is summed to obtain the global cooling energy consumption sequence.

[0114] Sub-step 4.2: Construct the time-aligned global total energy consumption sequence.

[0115] Specifically, by adding the IT equipment energy consumption to the time-aligned cooling energy consumption, the time-aligned global total energy consumption can be obtained, and then a global total energy consumption sequence can be generated in chronological order.

[0116] Sub-step 4.3: Construct the energy consumption correspondence structure between basic cooling capacity and additional cooling capacity.

[0117] Specifically, firstly, based on engineering experience, a basic cooling capacity ratio coefficient is given for the energy consumption of a unit of IT equipment, ranging from 0.3 to 0.8 (which can be determined according to the design conditions of the data center and the experience of normal seasonal operation). Basic cooling capacity energy consumption is defined as the product of IT equipment energy consumption and the basic cooling capacity ratio coefficient, representing the cooling energy consumed to ensure the normal operation of IT equipment under ideal uniform airflow and heat island-free conditions. Next, additional cooling capacity energy consumption is calculated, representing the additional cooling energy consumption caused by heat islands, uneven airflow, and cold short circuits, equal to the difference between the total global cooling energy consumption and the basic cooling capacity energy consumption. If the additional cooling capacity energy consumption is less than 0, it can be set to zero in actual operation, indicating that the current time is under a cooling energy-saving strategy. Finally, based on the above IT equipment energy consumption, basic cooling capacity energy consumption, and additional cooling capacity energy consumption, energy consumption curves for IT equipment energy consumption, basic cooling capacity energy consumption, and additional cooling capacity energy consumption with respect to time are constructed respectively. These three curves together constitute an energy consumption-correspondence structure matrix.

[0118] Sub-step 4.4 generates a dynamic PUE curve and binds it to the corresponding energy consumption structure.

[0119] Specifically, the dynamic PUE curve is a curve constructed with respect to time, representing the ratio of the total global energy consumption of IT devices after time alignment.

[0120] Step S150: Correct the PUE curve according to the regional cooling energy consumption deviation corresponding to different time periods in the PUE curve to obtain the corrected PUE curve.

[0121] In some embodiments, the data center PUE simulation method based on dynamic load response provided by the present invention includes the following steps in step S150:

[0122] Step S151: Calculate the spatial difference contribution value of each load area, and calculate the sum of the spatial difference contribution values ​​of all load areas based on the spatial difference contribution values ​​of each load area. Then, correct the additional cooling energy consumption by using the sum of the spatial difference contribution values ​​of all load areas to obtain the corrected additional cooling component.

[0123] Step S152: Based on the corrected additional cooling capacity component, combined with the IT equipment energy consumption and basic cooling capacity energy consumption, the PUE curve is corrected to output the corrected PUE curve.

[0124] In a specific embodiment, the data center PUE simulation method based on dynamic load response provided by the present invention includes step 5, which involves performing spatial difference correction based on the dynamic PUE curve and outputting the actual simulation results. The dynamic PUE curve obtained in step 4 is input into the correction unit. Based on the correspondence between different time periods and regional cooling differences in the curve, spatial difference correction is performed on the PUE, ensuring that the final PUE result reflects not only temporal changes but also spatial energy consumption differences. The PUE curve corrected in this step can realistically depict the energy consumption concentration phenomenon caused by high load in the heat island area, eliminating the overall underestimation caused by the averaging of regional cooling capacity. This includes the following sub-steps:

[0125] Sub-step 5.1: Calculate the spatial difference contribution coefficient and establish a spatial correction basis.

[0126] Specifically, in order to map spatial differences to the corresponding structure of additional cooling energy consumption, it is necessary to first construct a regional-level spatial difference contribution value, which is equal to the product of the cooling difference coefficient and the additional cooling energy consumption, to indicate which regions the additional cooling energy actually comes from.

[0127] Sub-step 5.2: Construct the spatiotemporal joint correction and apply it to the additional cooling component.

[0128] Specifically, first, the global spatial difference accumulation term is calculated, which equals the sum of the inter-regional difference contribution values ​​in the global environment at the same time. Then, the additional cooling energy consumption component is corrected using the sum of contribution values ​​to obtain the additional cooling correction amount, which is equal to the product of the correction coefficient (range 0.1-0.5, used to control the correction intensity) and the sum of contribution values. Finally, the additional cooling correction amount is summed with the original additional cooling energy consumption component to obtain the corrected additional cooling component. The corrected additional cooling energy consumption curve is then generated based on the corrected additional cooling component. Simultaneously, to ensure that the corrected additional cooling component does not conflict with the basic cooling component, the following constraint is introduced:

[0129] If the corrected additional cooling capacity component is less than 0, then the corrected additional cooling capacity component is equal to 0.

[0130] Sub-step 5.3 generates the final PUE curve based on the corrected additional cooling component.

[0131] Specifically, the corrected global energy consumption is defined as the sum of the IT equipment energy consumption, the basic cooling energy consumption, and the corrected additional cooling component. The final PUE value is equal to the ratio of the corrected global energy consumption (numerator) to the IT equipment energy consumption.

[0132] The data center PUE simulation platform based on dynamic load response provided by the present invention will be described below. The data center PUE simulation platform based on dynamic load response described below can be referred to in correspondence with the data center PUE simulation method based on dynamic load response described above.

[0133] like Figure 2 As shown in one embodiment, a data center PUE simulation platform based on dynamic load response includes a region division module, a cooling demand conversion module, a cooling energy consumption derivation module, a PUE curve generation module, and a PUE curve correction module.

[0134] The area division module is used to divide the computer room space into multiple load areas and determine local heat island areas based on the temperature data in each load area.

[0135] The cooling demand conversion module is used to convert the load of each load area in the local heat island area into regional cooling demand in each time period according to the time sequence, so as to generate the regional cooling demand time sequence structure.

[0136] The cooling energy consumption derivation module is used to derive the operating intensity of the air conditioning unit in each load area based on the time sequence structure of the regional cooling demand, and generate a regional cooling energy consumption sequence based on the operating intensity of the air conditioning unit.

[0137] The PUE curve generation module is used to generate PUE curves based on the regional cooling energy consumption sequence and time sequence, combined with the total energy consumption of the computer room at each time point, and the correspondence between the total energy consumption of the computer room and the regional cooling energy consumption.

[0138] The PUE curve correction module is used to correct the PUE curve based on the regional cooling energy consumption deviation corresponding to different time periods in the PUE curve, and obtain the corrected PUE curve.

[0139] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily essential to this application.

[0140] It should be further noted that although the steps in the flowchart are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowchart may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0141] It should be understood that the above-described device embodiments are merely illustrative, and the device of this application can also be implemented in other ways. For example, the division of units / modules in the above embodiments is only a logical functional division, and there may be other division methods in actual implementation. For example, multiple units, modules, or components may be combined, or integrated into another system, or some features may be ignored or not executed.

[0142] Furthermore, unless otherwise specified, the functional units / modules in the various embodiments of this application can be integrated into one unit / module, or each unit / module can exist physically separately, or two or more units / modules can be integrated together. The integrated units / modules described above can be implemented in hardware or as software program modules.

[0143] In the above embodiments, the descriptions of each embodiment have their own emphasis. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments. The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as the combination of these technical features does not contradict each other, it should be considered within the scope of this specification.

[0144] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the following claims.

[0145] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope.

Claims

1. A data center PUE simulation method based on dynamic load response, characterized in that, The method includes: The computer room space is divided into multiple load zones, and local heat island areas are determined based on the temperature data of each load zone; The load of each load area in the local heat island region is converted into regional cooling demand in time sequence to generate a time sequence structure of regional cooling demand. Based on the time sequence structure of the regional cooling demand, the operating intensity of the air conditioning units in each load area is derived, and a regional cooling energy consumption sequence is generated based on the operating intensity of the air conditioning units. Based on the regional cooling energy consumption sequence and time order, and combined with the total energy consumption of the computer room at each time point, the correspondence between the total energy consumption of the computer room and the regional cooling energy consumption is calculated to generate the PUE curve. The PUE curve is corrected based on the regional cooling energy consumption deviation corresponding to different time periods in the PUE curve to obtain the corrected PUE curve; The step of calculating the correspondence between the total energy consumption of the data center and the regional cooling energy consumption based on the regional cooling energy consumption sequence and time order, combined with the total energy consumption of the data center at each time point, to generate a PUE curve includes: The regional cooling energy consumption of each load area is time-aligned, and the total cooling energy consumption of all load areas at a unified time is calculated based on the time-aligned regional cooling energy consumption. The energy consumption of IT equipment is then added to the total cooling energy consumption to obtain the total energy consumption of the computer room at each time. Obtain the basic cooling energy consumption required for the IT equipment's energy consumption, and calculate the difference between the total cooling energy consumption and the basic cooling energy consumption to obtain the additional cooling energy consumption; Based on the energy consumption of the IT equipment, the basic cooling energy consumption, and the additional cooling energy consumption, combined with the total energy consumption of the computer room, the PUE value at each time point is determined to generate a PUE curve of the PUE value over time. The step of correcting the PUE curve based on the regional cooling energy consumption deviation corresponding to different time periods in the PUE curve to obtain the corrected PUE curve includes: Calculate the spatial difference contribution value of each load area, and calculate the sum of the spatial difference contribution values ​​of all load areas based on the spatial difference contribution values ​​of each load area. Then, correct the additional cooling energy consumption by the sum of the spatial difference contribution values ​​of all load areas to obtain the corrected additional cooling component. Based on the corrected additional cooling capacity component, combined with the IT equipment energy consumption and basic cooling capacity energy consumption, the PUE curve is corrected to output a corrected PUE curve.

2. The data center PUE simulation method based on dynamic load response according to claim 1, characterized in that, The process of dividing the computer room space into multiple load zones and determining local heat island areas based on temperature data within each load zone includes: Obtain the distribution coordinates of server racks, air conditioners, and supply and return air paths within the server room, and divide the server room space into load areas based on the server rack distribution coordinates, air conditioner distribution coordinates, and supply and return air paths to obtain multiple load areas; The cabinet power in each load area is summed at each time point, and a regional load time series is generated in chronological order based on the sum of the cabinet power in each load area.

3. The data center PUE simulation method based on dynamic load response according to claim 2, characterized in that, The method of dividing the computer room space into multiple load zones and determining local heat island areas based on temperature data within each load zone also includes: The temperature data of each load area at each time point is statistically analyzed, and the high temperature time window of the load area is determined based on the temperature data. The temperature data is the average value of the data collected by all temperature sensors in the corresponding load area. Based on the high temperature time window and regional load time series of the load area, it is determined whether the corresponding load area meets the preset conditions, and the load area that meets the preset conditions is marked as the local heat island area. The preset conditions are that the total power of the cabinet is higher than a first threshold for multiple consecutive times and the temperature data is higher than a second threshold for multiple consecutive times. The first threshold and the second threshold are the set upper limits for power and temperature, respectively.

4. The data center PUE simulation method based on dynamic load response according to claim 1, characterized in that, The step of converting the load of each load area in the local heat island region into regional cooling demand in each time period according to time sequence to generate a time-series structure of regional cooling demand includes: Obtain the air supply coverage area of ​​each air conditioner in the load area, and generate an air supply coverage matrix corresponding to each load area based on the air supply coverage area of ​​each air conditioner. At the same time, convert the air supply coverage matrix into the initial cooling capacity requirement of the area. When the load area is located in the local heat island area, a heat compensation amount is added to the initial cooling demand of the area to obtain the area enhanced cooling demand by summing the initial cooling demand and the heat compensation amount. Based on the time sequence and the air supply coverage area of ​​each air conditioner, the enhanced cooling demand of the region is integrated into a time sequence structure of the regional cooling demand.

5. The data center PUE simulation method based on dynamic load response according to claim 1, characterized in that, The step of deriving the operating intensity of air conditioning units in each load area based on the time-series structure of regional cooling demand, and generating a regional cooling energy consumption sequence based on the operating intensity of the air conditioning units, includes: Based on the different height layers of the load area in the computer room space, the time sequence structure of the cooling demand of the area is divided into multiple height-layered cooling demand, and the effective fan operation intensity of the load area at each time is defined according to the height-layered cooling demand and the airflow delay coefficient. The initial chilled water flow rate of the load area at each time moment is obtained, and the initial chilled water flow rate is adjusted in conjunction with the effective fan operating intensity to determine the regional chilled water flow rate. The fan power component and the water pump power component are calculated based on the effective fan operating intensity and the regional chilled water flow rate, respectively. The invalid wind power component of the corresponding load area is defined based on the fan power component and the water pump power component to determine the regional cooling energy consumption. The regional cooling energy consumption sequence is obtained by arranging the regional cooling energy consumption in chronological order, and the regional cooling energy consumption is the sum of the fan power component, water pump power component, and ineffective wind power component of the corresponding load area.

6. A data center PUE simulation platform based on dynamic load response, characterized in that, The platform for implementing the data center PUE simulation method based on dynamic load response as described in any one of claims 1 to 5 includes: The area division module is used to divide the computer room space into multiple load areas and determine local heat island areas based on the temperature data of each load area. The cooling demand conversion module is used to convert the load of each load area in the local heat island area into regional cooling demand in each time period according to the time sequence, so as to generate a regional cooling demand time sequence structure. The cooling energy consumption derivation module is used to derive the operating intensity of the air conditioning unit in each load area based on the time sequence structure of the regional cooling demand, and to generate a regional cooling energy consumption sequence based on the operating intensity of the air conditioning unit. The PUE curve generation module is used to calculate the correspondence between the total energy consumption of the computer room and the regional cooling energy consumption based on the regional cooling energy consumption sequence and time sequence, combined with the total energy consumption of the computer room at each time point, so as to generate the PUE curve. The PUE curve correction module is used to correct the PUE curve based on the regional cooling energy consumption deviation corresponding to different time periods in the PUE curve, so as to obtain the corrected PUE curve.

7. A terminal, comprising a processor and a storage medium; characterized in that: The storage medium is used to store instructions; The processor is configured to operate according to the instructions to perform the steps of the method according to any one of claims 1-5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the steps of the method according to any one of claims 1-5.

Images