Method and device for establishing power consumption characteristic model library, electronic equipment and storage medium

Abstract

Embodiments of the present application provide a power consumption characteristic model library establishment method and device, electronic equipment and a storage medium. The method comprises: obtaining real-time running data of each type of device in a target machine room in different time periods; performing characteristic modeling on the real-time running data based on a time-varying parameter random volatility rate vector autoregressive model to generate a time-varying influence coefficient graph and an impulse response graph corresponding to each type of device; determining the variable relationship between related variables in the running process of each type of device according to the time-varying influence coefficient graph and the impulse response graph; and constructing a power consumption characteristic model library based on an application scenario corresponding to each type of device in the target machine room according to the variable relationship. The power consumption characteristic model library based on an application scenario established by the embodiments of the present application can effectively improve power utilization while ensuring the safety of communication machine room power operation.

Description

Technical Field

[0001] This application relates to the field of communication technology, and in particular to a method, apparatus, electronic device, and storage medium for establishing a power consumption characteristic model library. Background Technology

[0002] Currently, new-generation information technologies such as cloud computing, big data, mobile internet, and the Internet of Things are booming, leading to a surge in demand for data storage, exchange, and computing applications. Data centers, as a representative of new information infrastructure, have become crucial carriers and supports for the development of smart cities and the digital economy. Energy conservation and carbon reduction are critical tasks throughout the entire lifecycle of data centers. Currently, data centers and communication buildings generally suffer from low power utilization rates during operation, resulting not only in wasted investment but also becoming a major obstacle to reducing PUE (Power Usage Effectiveness).

[0003] During the operation of data centers and communication buildings, the operational risks are high due to the setting of operation and maintenance guarantee assessment indicators and the restrictions of operation and maintenance specifications, especially without adequate technical support. Therefore, the power supply for the managed equipment is subject to strict requirements, and the equipment is generally installed and powered according to the "rated power". However, there is no unified calculation method for "rated power". Currently, the commonly used technical methods are: (1) the maximum output power of the power module; (2) the power consumption of various boards of the equipment provided by the manufacturer; (3) the comprehensive average empirical value calculated through years of practice and combined with previous power monitoring data. All of the above calculation methods have resulted in low power utilization to varying degrees.

[0004] Currently, mainstream manufacturers' equipment has functions such as "power caps" and "power walls" to control the maximum power consumption of individual devices. All devices are configured with this setting to keep the overall rack power consumption within a safe range. The operation and maintenance system combines power consumption and load conditions to dynamically adjust service orchestration to improve rack utilization efficiency. "Power caps" physically reduce the CPU (Central Processing Processor) frequency, thus reducing service processing capacity. Since "power caps" are often triggered during peak service periods, this can lead to service congestion. Furthermore, there are certain information barriers between the IT and network design and operation levels and the design and operation levels of communication equipment buildings and data center rooms, failing to fundamentally solve the problem of low power utilization in communication equipment buildings and data centers. Summary of the Invention

[0005] The technical problem to be solved by the embodiments of this application is to provide a method, apparatus, electronic device and storage medium for establishing a power consumption characteristic model library, so as to effectively improve power utilization while ensuring the safe operation of power supply in communication equipment rooms.

[0006] In a first aspect, embodiments of this application provide a method for establishing a power consumption characteristic model library, including:

[0007] Acquire real-time operational data of various types of equipment in the target data center at different times;

[0008] Based on the time-varying parameter random volatility vector autoregression model, the real-time operating data is modeled to generate time-varying influence coefficient diagrams and impulse response diagrams for each type of equipment.

[0009] Based on the time-varying influence coefficient diagram and the impulse response diagram, determine the variable relationships between relevant variables of each type of equipment during operation;

[0010] Based on the variable relationships, a power consumption characteristic model library based on application scenarios is constructed for each type of device in the target computer room.

[0011] Optionally, acquiring real-time operational data of various types of equipment within the target computer room at different time periods includes:

[0012] Obtain the CPU utilization rate of various types of equipment in the target computer room at different time periods;

[0013] Based on the device identifier corresponding to each type of device and the CPU utilization, obtain the operating power of each type of device in the corresponding time period;

[0014] The CPU utilization and the operating power are used as real-time operating data of various types of equipment in the target computer room at different time periods, and the real-time operating data is stored in a time-series database.

[0015] Optionally, the step of modeling the characteristics of the real-time operating data based on the time-varying parameter stochastic volatility vector autoregressive model to generate time-varying influence coefficient diagrams and impulse response diagrams corresponding to various types of equipment includes:

[0016] Obtain real-time operating data of various types of devices from the time-series database;

[0017] Perform a stability test on the real-time operating data, and obtain the device operating data in the real-time operating data whose stability is lower than a set value;

[0018] The device operating data is differentially processed to obtain differential operating data;

[0019] The differential operating data and the power consumption of the benchmark equipment corresponding to each type of equipment are input into the time-varying parameter random volatility vector autoregressive model to obtain the time-varying influence coefficient diagram and the impulse response diagram corresponding to each type of equipment.

[0020] Optionally, the step of constructing a power consumption characteristic model library based on application scenarios for each type of equipment in the target data center according to the variable relationships includes:

[0021] Based on the aforementioned variable relationships, power consumption characteristics data of various types of equipment in the target computer room are obtained throughout the day.

[0022] Based on the power consumption characteristic data, a power consumption characteristic model based on the application scenario is constructed for each type of device in the target computer room.

[0023] The power consumption characteristic model library is generated based on the power consumption characteristic models corresponding to each type of device;

[0024] The application scenarios include at least one of: stable scenarios and non-stable scenarios;

[0025] The stable scenario is used to indicate a scenario where the power consumption is stable throughout the day, while the non-stable scenario is used to indicate a scenario where the power consumption has a set peak / valley value throughout the day.

[0026] Optionally, after constructing the power consumption characteristic model library based on application scenarios for each type of equipment in the target data center according to the variable relationships, the method further includes:

[0027] When the target device is installed in the target computer room, the target device type of the target device is obtained;

[0028] Based on the target device type, the target power consumption characteristic model corresponding to the target device is selected from the power consumption characteristic model library;

[0029] Based on the target power consumption characteristic model, determine the simulated operating power consumption of the target device in the next day;

[0030] Based on the simulated operating power consumption and the system power consumption of the equipment in the target computer room other than the target equipment, determine whether the power-on conditions of the target equipment are met;

[0031] If the power-on conditions of the target device are met, the target device is powered on.

[0032] Optionally, determining whether the power-on conditions for the target device are met based on the simulated operating power consumption and the system power consumption of devices other than the target device in the target computer room includes:

[0033] Compare the sum of the simulated operating power consumption and the system power consumption with the power consumption threshold.

[0034] Based on the comparison results, determine whether the power-on conditions of the target device are met.

[0035] Optionally, after determining whether the power-on conditions of the target device are met based on the simulated operating power consumption and the system power consumption of devices other than the target device in the target computer room, the method further includes:

[0036] If the power-on conditions of the target equipment are not met, determine whether the power environment of the target area where the target computer room is located has the conditions for expansion.

[0037] If it is determined that the power environment of the target area is capable of expansion, the target computer room is expanded and the target equipment is powered on.

[0038] If it is determined that the power supply environment of the target area does not meet the requirements for capacity expansion, the racking area of ​​the target device shall be changed.

[0039] Secondly, embodiments of this application provide a power consumption characteristic model library establishment apparatus, comprising:

[0040] The runtime data acquisition module is used to acquire real-time runtime data of various types of equipment in the target computer room at different time periods;

[0041] The equipment coefficient diagram generation module is used to perform characteristic modeling on the real-time operating data based on the time-varying parameter random volatility vector autoregression model, and generate time-varying influence coefficient diagrams and impulse response diagrams corresponding to various types of equipment.

[0042] The variable relationship determination module is used to determine the variable relationships between relevant variables of various types of equipment during operation based on the time-varying influence coefficient diagram and the impulse response diagram;

[0043] The power consumption characteristic model library construction module is used to construct a power consumption characteristic model library based on application scenarios for each type of device in the target computer room according to the variable relationship.

[0044] Optionally, the runtime data acquisition module includes:

[0045] CPU utilization acquisition unit is used to acquire the CPU utilization of various types of equipment in the target computer room at different time periods.

[0046] The operating power acquisition unit is used to acquire the operating power of each type of device in the corresponding time period based on the device identifier corresponding to each type of device and the CPU utilization rate.

[0047] The real-time operation data acquisition unit is used to use the CPU utilization and the operating power as real-time operation data of various types of equipment in the target computer room at different time periods, and to store the real-time operation data in a time-series database.

[0048] Optionally, the equipment coefficient diagram generation module includes:

[0049] The running data acquisition unit is used to acquire real-time running data of various types of devices in the time series database;

[0050] The equipment operation data acquisition unit is used to perform a stability test on the real-time operation data and acquire equipment operation data in the real-time operation data whose stability is lower than a set value.

[0051] The differential operation data acquisition unit is used to perform differential processing on the device operation data to obtain differential operation data;

[0052] The equipment coefficient diagram acquisition unit is used to input the differential operating data and the power consumption of the benchmark equipment corresponding to each type of equipment into the time-varying parameter random volatility vector autoregressive model to obtain the time-varying influence coefficient diagram and the impulse response diagram corresponding to each type of equipment.

[0053] Optionally, the power consumption characteristic model library construction module includes:

[0054] The power consumption characteristic data acquisition unit is used to acquire power consumption characteristic data of various types of equipment in the target computer room throughout the day based on the variable relationship.

[0055] The power consumption characteristic model construction unit is used to construct, based on the power consumption characteristic data, a power consumption characteristic model corresponding to each type of device in the target computer room and the application scenario.

[0056] The power consumption characteristic model library generation unit is used to generate the power consumption characteristic model library according to the power consumption characteristic models corresponding to each type of device.

[0057] The application scenarios include at least one of: stable scenarios and non-stable scenarios;

[0058] The stable scenario is used to indicate a scenario where the power consumption is stable throughout the day, while the non-stable scenario is used to indicate a scenario where the power consumption has a set peak / valley value throughout the day.

[0059] Optionally, the device further includes:

[0060] The target device type acquisition module is used to acquire the target device type of the target device when the target device is installed in the target computer room;

[0061] The target power consumption model filtering module is used to filter the target power consumption characteristic model corresponding to the target device from the power consumption characteristic model library according to the target device type;

[0062] The simulated operating power consumption determination module is used to determine the simulated operating power consumption of the target device in the next day based on the target power consumption characteristic model.

[0063] The power-on condition determination module is used to determine whether the power-on conditions of the target device are met based on the simulated operating power consumption and the system power consumption of the devices in the target computer room other than the target device.

[0064] The target device power-on module is used to power on the target device when the power-on conditions of the target device are met.

[0065] Optionally, the power-on condition determination module includes:

[0066] A power consumption comparison unit is used to compare the sum of the simulated operating power consumption and the system power consumption with a power consumption threshold.

[0067] The power-on condition determination unit is used to determine whether the power-on conditions of the target device are met based on the comparison results.

[0068] Optionally, the device further includes:

[0069] The expansion condition determination module is used to determine whether the power environment of the target area where the target computer room is located meets the expansion conditions if the power supply conditions of the target device are not met.

[0070] The power-on processing module is used to perform capacity expansion processing on the target computer room and power-on processing on the target equipment when it is determined that the power environment of the target area has the capacity expansion conditions.

[0071] The racking area change module is used to change the racking area of ​​the target device when it is determined that the power environment of the target area does not meet the expansion conditions.

[0072] Thirdly, embodiments of this application provide an electronic device, including:

[0073] A processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, implements the power consumption characteristic model library establishment method described above.

[0074] Fourthly, embodiments of this application provide a computer-readable storage medium that, when the instructions in the storage medium are executed by a processor of an electronic device, enables the electronic device to execute the power consumption characteristic model library establishment method described in any of the preceding claims.

[0075] Compared with the prior art, the embodiments of this application have the following advantages:

[0076] In this embodiment, real-time operating data of various types of equipment in the target equipment room at different time periods is acquired. Based on a time-varying parameter stochastic volatility vector autoregressive model, the real-time operating data is modeled to generate time-varying influence coefficient diagrams and impulse response diagrams for each type of equipment. Based on these diagrams, the variable relationships between relevant variables during the operation of each type of equipment are determined. Based on these relationships, a power consumption characteristic model library based on application scenarios for each type of equipment in the target equipment room is constructed. This embodiment, by combining real-time operating data of various types of equipment in the equipment room at different time periods to construct a power consumption characteristic model library based on application scenarios for each type of equipment in the target equipment room, allows for the assignment of actual power consumption values ​​to new equipment during power-on evaluation after modeling is completed. This also incorporates time-dimensional factors for power-on evaluation, ensuring the safe operation of power consumption in the communication equipment room while ultimately improving power utilization.

[0077] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0078] Figure 1 A flowchart illustrating the steps of a method for establishing a power consumption characteristic model library as provided in this application embodiment;

[0079] Figure 2 A flowchart illustrating the steps of a method for acquiring runtime data provided in this application embodiment;

[0080] Figure 3 A flowchart illustrating the steps of a method for obtaining an equipment coefficient diagram, as provided in this application embodiment;

[0081] Figure 4 A flowchart illustrating the steps of a method for generating a power consumption characteristic model library, as provided in an embodiment of this application;

[0082] Figure 5 A flowchart illustrating the steps of a method for powering on a target device, as provided in an embodiment of this application;

[0083] Figure 6 A flowchart illustrating the steps of a method for determining power-on conditions provided in this application embodiment;

[0084] Figure 7 A flowchart illustrating the steps of a method for changing the shelf area provided in this application embodiment;

[0085] Figure 8 A schematic diagram of a power consumption characteristic model library establishment device provided in this application embodiment;

[0086] Figure 9 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0087] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0088] The terminology used in the embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. The singular forms “a,” “the,” and “the” used in the embodiments of this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise.

[0089] In existing technical solutions, the following two methods are typically used:

[0090] 1. In a data center's cloud resource pool, the platform can intelligently control the overhead of calling specific servers and storage devices when scheduling and allocating resources, which indirectly controls the power consumption level of the devices. However, the platform is not linked to the overall power resource situation of the data center where the cloud resource pool is located. The cloud resource pool designers and platform system maintenance personnel are unaware of the overall power situation and potential of the data center space. Conversely, the data center's environmental monitoring system does not monitor the operating mechanism of the cloud resource pool system's IT (Internet Technology) devices. The data center's power maintenance personnel are also unaware that the cloud resource pool's IT system's intelligent scheduling technology can achieve intelligent scheduling of resource expenditure and control of power consumption. They usually adopt conservative equipment racking and power control strategies, resulting in low rack space utilization and power utilization for a long time. This leads to increased data center construction costs, wasted power energy, and difficulty in reducing the data center's PUE.

[0091] 2. In a telecommunications building, the power-consuming equipment of various specialties such as switching, transmission, data, and wireless is more complex than that of a data center. In the initial planning stage, when each specialty submits its power requirements to the telecommunications building's control room, it usually submits the requirements based on the power consumption values ​​provided by the manufacturer. (For example, for a certain 5G (5th Generation Mobile Communication Technology) BBU (Building Baseband Unit) equipment, the manufacturer reported a power consumption of 20kW. The control room configured the infrastructure according to this power consumption standard. As a result, the actual power consumption after the equipment went online deviated significantly from the estimated value, resulting in low actual power utilization.) Furthermore, there is a lack of linkage between the network management of the communication equipment and the network management of the environmental protection system in the control room after the equipment went online. Similarly, there is a problem of no information exchange between systems during the operation phase. When managing power supply, the on-site power maintenance personnel also verify the equipment power based on the estimated value provided by the manufacturer. This conservative power supply strategy also results in a consistently low level of power utilization.

[0092] In order to solve the above-mentioned technical problems, this application embodiment models the relationship between device CPU utilization and actual operating power consumption, and establishes an interaction mechanism between various systems. This enables the power utilization rate to be effectively improved while ensuring the safe operation of power supply in communication equipment rooms, and also contributes to saving investment and reducing energy consumption for enterprises.

[0093] The technical solutions of the embodiments of this application will be described in detail below with reference to specific examples.

[0094] Reference Figure 1 The flowchart illustrates the steps of a method for establishing a power consumption characteristic model library according to an embodiment of this application. Figure 1 As shown, the method for establishing this power consumption characteristic model library may include the following steps:

[0095] Step 101: Obtain real-time operating data of various types of equipment in the target computer room at different time periods.

[0096] The embodiments of this application can be applied to scenarios where a power consumption characteristic model library is constructed by combining the operating data of various types of equipment in the computer room over multiple time periods, so as to evaluate the power-on performance of new equipment in conjunction with the model library.

[0097] The target equipment room is the communication equipment room that needs to be modeled. The equipment placed in the target equipment room may include: signal transmitters, transmission equipment (such as SDH (Synchronous Digital Hierarchy), PDH (Plesiochronous Digital Hierarchy) equipment, optical distribution frames (ODF), digital distribution frames (DDF), BBU equipment, etc.

[0098] Real-time operational data refers to the operational data of various types of equipment in the target data center at different times. Real-time operational data can include data such as CPU utilization and power consumption of various types of equipment at different times.

[0099] When building a power consumption characteristic model library for the target data center, real-time operating data of various types of equipment within the data center can be obtained at different times. The implementation process for obtaining this real-time operating data can be combined with... Figure 2 The following is a detailed description.

[0100] Reference Figure 2 The flowchart illustrates the steps of a method for acquiring runtime data according to an embodiment of this application. Figure 2 As shown, the method for obtaining runtime data may include steps 201, 202, and 203.

[0101] Step 201: Obtain the CPU utilization rate of each type of equipment in the target computer room at different time periods.

[0102] In this embodiment, CPU utilization refers to the CPU resources occupied by the running program, representing the program running status of the device at a certain point in time.

[0103] It can obtain the CPU utilization of various types of devices in the target data center at different times.

[0104] After obtaining the CPU utilization of various types of devices in the target data center at different times, proceed to step 202.

[0105] Step 202: Based on the device identifier corresponding to each type of device and the CPU utilization rate, obtain the operating power of each type of device in the corresponding time period.

[0106] Equipment identification can be used to indicate information such as the manufacturer and model of various types of equipment.

[0107] After obtaining the CPU utilization rates of various types of equipment in the target data center at different times, the operating power of each type of equipment during the corresponding time period can be obtained based on its corresponding equipment identifier and CPU utilization rate. In specific implementation, the manufacturer and model of each type of equipment can be determined based on its equipment identifier. Then, by combining the power score and CPU utilization rate corresponding to that manufacturer and model, the real-time operating power of each type of equipment can be obtained.

[0108] After obtaining the operating power of each type of device in the corresponding time period based on the device identifier and CPU utilization, step 202 is executed.

[0109] Step 203: Use the CPU utilization and the operating power as real-time operating data of various types of equipment in the target computer room at different time periods, and store the real-time operating data in a time-series database.

[0110] After obtaining the operating power of each type of device in the corresponding time period based on the device identifier and CPU utilization, the CPU utilization and operating power can be used to track the real-time operating data of each type of device in the target data room in different time periods. This real-time operating data is stored in a time-series database, which can record the operating data of each communication device in the target data room in each time period, so as to facilitate the subsequent establishment of a power consumption characteristic model library for the target data room.

[0111] The system's data acquisition includes: using an integrated cloud management platform operation system, developing relevant interfaces and application functions for the cloud management platform, collecting real-time operating power consumption of equipment in network, transmission, and IT systems, collecting real-time power consumption of racks, central office rooms, and other unit sets in the environmental system, and using the SNMP protocol for interaction between network, transmission, IT, and power and environmental systems, with a data acquisition granularity of 5 minutes / time (if the equipment and system support a better Telemetry protocol, the acquisition granularity frequency can be increased to 5 seconds / time). It achieves full interaction between information from various systems in the data center, including network, transmission, IT, and power, and stores the collected real-time data such as CPU utilization and actual power consumption of various devices at different times, as well as the real-time data such as actual power consumption of communication equipment room racks and central office rooms at different times, into a time-series database.

[0112] After obtaining the real-time operating data of various types of equipment in the target data center at different times, proceed to step 102.

[0113] Step 102: Based on the time-varying parameter stochastic volatility vector autoregressive model, perform characteristic modeling on the real-time operating data to generate time-varying influence coefficient diagrams and impulse response diagrams corresponding to each type of equipment.

[0114] Compared to vector autoregression (VAR), the time-varying parameter-stochastic volatility-vector autoregression (TVP-VAR) model does not assume homoscedasticity, making it more realistic. Furthermore, the assumption of stochastic volatility in the time-varying parameter better captures the relationships and characteristics of variables across different time contexts (i.e., variable effects), significantly improving estimation performance.

[0115] After obtaining the real-time operating data of various types of equipment in the target computer room at different time periods, characteristic modeling of the real-time operating data of various types of equipment in the target computer room can be performed based on TVP-VAR, thereby generating time-varying influence coefficient diagrams and impulse response diagrams corresponding to each type of equipment.

[0116] The time-varying influence coefficient diagram reflects the peak and trough states of the communication equipment during all-weather operation. The impulse response diagram reflects the CPU utilization and actual operating power of the communication equipment at different times.

[0117] The generation process of time-varying influence coefficient diagrams and impulse response diagrams can be combined with... Figure 3 The following is a detailed description.

[0118] Reference Figure 3 The flowchart illustrates the steps of a method for obtaining a device coefficient diagram according to an embodiment of this application. Figure 3 As shown, the method for obtaining the equipment coefficient diagram may include steps 301, 302, 303, and 304.

[0119] Step 301: Obtain real-time operating data of each type of device in the time series database.

[0120] In this embodiment, when obtaining the time-varying influence coefficient diagram and the impulse response diagram corresponding to each type of device, real-time operating data of each type of device can be obtained from the time series database.

[0121] After obtaining the real-time operating data of various types of devices in the time series database, proceed to step 302.

[0122] Step 302: Perform a stationarity test on the real-time operating data and obtain the device operating data whose stationarity is lower than the set value in the real-time operating data.

[0123] The set value refers to the stability threshold that is preset by business personnel to obtain data with poor stability in the real-time operation data of communication equipment. The specific value of the set value can be determined according to business needs, and this embodiment does not impose any restrictions on it.

[0124] After obtaining real-time operating data for various types of devices from the time-series database, a stationarity test can be performed on the real-time operating data to identify devices whose stationarity is below a set value. In practice, the ADF unit root test can be used to verify the stationarity of the data.

[0125] After obtaining equipment operation data with stability lower than the set value from the real-time operation data by performing a stability test on the real-time operation data, step 303 is executed.

[0126] Step 303: Perform differential processing on the equipment operation data to obtain differential operation data.

[0127] After obtaining the equipment operation data for each type of equipment, differential processing can be performed on the equipment operation data to obtain differential operation data for each type of equipment. This process eliminates some fluctuations in the equipment operation data, making the data more stable.

[0128] After differential processing of the equipment operation data of various types of equipment to obtain differential operation data, step 304 is executed.

[0129] Step 304: Input the differential operating data and the power consumption of the reference equipment corresponding to each type of equipment into the time-varying parameter random volatility vector autoregressive model to obtain the time-varying influence coefficient diagram and the impulse response diagram corresponding to each type of equipment.

[0130] The power consumption of the reference device differs from that of the actual device. The power consumption of the reference device is the power consumption of various types of devices during operation, as specified by the manufacturer, taking into account the characteristics of the device structure and other factors.

[0131] After differentially processing the equipment operation data of various types of equipment to obtain differential operation data, the differential operation data and the corresponding benchmark equipment power consumption of each type of equipment can be input into a time-varying parameter stochastic volatility vector autoregressive model to obtain the corresponding time-varying influence coefficient diagram and impulse response diagram for each type of equipment. Specifically, the time-varying parameter stochastic volatility vector autoregressive model can be used to model the characteristics based on the differential operation data, thereby obtaining the corresponding time-varying influence coefficient diagram and impulse response diagram for each type of equipment. Specifically, the time-varying parameter stochastic volatility vector autoregressive model can be used to analyze the nominal power consumption, actual operating power consumption, and CPU utilization data of the equipment to establish the impulse response law of each type of equipment, thereby determining the correlation between the actual operating power consumption and the CPU utilization of each type of equipment. That is, the nominal power consumption (i.e., benchmark equipment power consumption), actual operating power consumption, and CPU utilization of each type of equipment are used as model variables and substituted into the standard formula of the vector autoregressive model for modeling.

[0132] The specific implementation process is as follows:

[0133] 1. Define a standard form of vector autoregression model:

[0134] Ay t =F1y t-1 +...+F s y t-s +μ t ,t=s+1,...,n

[0135] Let y t Let A be a k×1 dimensional observation variable, and F be a k×k dimensional coefficient matrix. s For a k×k dimensional lag operator, μ t Let μ be a k×1 dimensional random perturbation term. t ~N(0,∑∑), where ∑ is a diagonal matrix.

[0136]

[0137] Assuming A is a lower triangular matrix, the relationship of a specified structural impact is identified recursively.

[0138]

[0139] Multiply both sides of the standard form of the vector autoregressive model by A. -1 It can be rewritten as:

[0140] y t =B1y t-1 +...+B s y t-s +A -1 ∑ε t

[0141] Among them, B i =A -1 F i ,i=1...s, defined ( (representing the Kronecker product), ε t ~N(0,I k ).

[0142] y t =X t β+A -1 ∑ε t In the expression, all parameters are time-varying, allowing the derivation of a time-varying vector autoregressive (TVP-VAR) model among different device model variables. Therefore, the TVP-VAR model can be expressed in the following form:

[0143] y t =X t β+A -1 ∑ t εt ,t=s+1,...,n

[0144] Where, β t for 2D time-varying coefficient vector, disturbance term ε t It is a k×1 structural shock. The coefficient vector β is... t Matrix A t The sum of the covariance matrix ∑ t Both are time-varying, and the time-varying matrix A t This indicates that the impact of the shock to the i-th variable on the j-th variable changes over time.

[0145] After generating time-varying influence coefficient diagrams and impulse response diagrams for various types of equipment by performing characteristic modeling of real-time operating data based on the time-varying parameter stochastic volatility vector autoregressive model, step 103 is executed.

[0146] Step 103: Based on the time-varying influence coefficient diagram and the impulse response diagram, determine the variable relationships between relevant variables of each type of equipment during operation.

[0147] In this embodiment, the relevant variables may refer to variables such as CPU utilization and actual power consumption.

[0148] After generating time-varying influence coefficient diagrams and impulse response diagrams for various types of equipment by modeling the characteristics of real-time operating data using a time-varying parameter stochastic volatility vector autoregressive model, the variable relationships between relevant variables of each type of equipment during operation can be determined based on these diagrams. Specifically, by analyzing the time-varying influence coefficient diagrams and impulse response diagrams, the relationships between CPU utilization and actual power consumption of each type of equipment during peak and trough periods throughout the day can be obtained.

[0149] After determining the variable relationships between relevant variables of various types of equipment during operation based on the time-varying influence coefficient diagram and impulse response diagram, proceed to step 104.

[0150] Step 104: Based on the variable relationships, construct a power consumption characteristic model library corresponding to each type of equipment in the target computer room based on the application scenario.

[0151] After determining the variable relationships between relevant variables of various types of equipment during operation based on the time-varying influence coefficient diagram and impulse response diagram, a power consumption characteristic model library based on application scenarios can be constructed according to these variable relationships. Specifically, a power consumption characteristic model corresponding to each type of equipment can be constructed based on the variable relationships of each type of equipment. Then, by combining the power consumption characteristic models of multiple types of equipment, a power consumption characteristic model library based on application scenarios can be obtained for each type of equipment in the target data room. Specifically, the data collected by the system can be modeled, and the Monte Carlo Markov Chain (MCMC) algorithm can be used to estimate... The time-varying parameter values ​​in the equations are analyzed using impulse response functions to examine the influence between different time-varying vectors in the TVP-VAR model, and a power consumption characteristic model library based on application scenarios is established.

[0152] The process of building a power consumption characteristic model library can be combined with... Figure 4 The following is a detailed description.

[0153] Reference Figure 4 The flowchart illustrates the steps of a method for generating a power consumption characteristic model library according to an embodiment of this application. Figure 4 As shown, the method for generating the power consumption characteristic model library may include steps 401, 402 and 403.

[0154] Step 401: Based on the variable relationship, obtain the power consumption characteristics data of each type of equipment in the target computer room throughout the day.

[0155] In this embodiment, after obtaining the variable relationships corresponding to each type of device, the power consumption characteristics data of each type of device in the target computer room throughout the day can be obtained based on the variable relationships corresponding to each type of device, namely CPU utilization, operating power and other data.

[0156] After obtaining the power consumption characteristics data of various types of equipment in the target data center throughout the day based on variable relationships, step 402 is executed.

[0157] Step 402: Based on the power consumption characteristic data, construct a power consumption characteristic model based on the application scenario for each type of device in the target computer room.

[0158] In this embodiment, the application scenarios may include at least one of the following: stable scenarios (i.e., object scenarios) and non-stable scenarios (i.e., human scenarios). Stable scenarios can be used to indicate scenarios where power consumption is stable throughout the day, while non-stable scenarios are used to indicate scenarios where power consumption has a set peak / valley value throughout the day. A scenario where power consumption is relatively stable throughout the day, with virtually no obvious peaks or valleys, can be defined as an "object scenario," in which devices typically operate stably 24 / 7. A scenario where power consumption has very obvious peaks and valleys throughout the day can be defined as a "human scenario," in which device power consumption exhibits a tidal phenomenon, mainly related to human activities such as rest and travel.

[0159] After obtaining the power consumption characteristic data of each type of device in the target data center throughout the day based on the variable relationship, a power consumption characteristic model based on the application scenario for each type of device in the target data center can be constructed based on the power consumption characteristic data of each type of device throughout the day.

[0160] After constructing the power consumption characteristic model based on the application scenario for each type of device in the target data room based on the power consumption characteristic data, step 403 is executed.

[0161] Step 403: Generate the power consumption characteristic model library based on the power consumption characteristic models corresponding to each type of device.

[0162] After constructing application scenario-based power consumption characteristic models for each type of device in the target data center based on power consumption characteristic data, a power consumption characteristic model library for the target data center can be generated based on the power consumption characteristic models for each type of device.

[0163] This application uses a time-varying parameter stochastic volatility vector autoregressive model to model the actual operating power consumption of equipment. This model does not assume homoscedasticity, making it more realistic. The assumption of stochastic volatility in the time-varying parameters better captures the relationships and characteristics (time-varying effects) of various variables across different time dimensions. Establishing a time-series database of parameters such as CPU utilization and actual power consumption for various equipment is a universally applicable modeling and calculation method, characterized by speed, accuracy, and adaptability to local conditions. After modeling, using the actual power consumption of the equipment superimposed with the time dimension as the power-on standard allows for a stable and smooth gradual approximation of the optimal power utilization method. This overcomes the disadvantages of traditional methods that rely entirely on empirical values, effectively improving power utilization while ensuring electrical safety.

[0164] In practical applications, during the initial stage of model library establishment, application scenarios can be simply defined as two categories of power consumption characteristic model libraries: "object scenarios" and "human scenarios." Later, with the continuous operation of the data center and the continuous expansion of the collected data and models, the model library can be further subdivided based on industries, administrative regions, etc. Because when evaluating the power-on of new equipment, the system model not only assigns a value to the actual power consumption of the equipment but also incorporates the time dimension factor. This will comprehensively and further explore the potential of infrastructure space and power resources, ultimately achieving the goal of improving power utilization.

[0165] The implementation process of power-on assessment can be combined with Figure 5 The following is a detailed description.

[0166] Reference Figure 5 This document illustrates a flowchart of the steps involved in a method for powering on a target device according to an embodiment of this application. Figure 5 As shown, the power-on method for the target device may include steps 501, 502, 503, 504, and 505.

[0167] Step 501: When the target device is installed in the target computer room, obtain the target device type of the target device.

[0168] In this embodiment, the target device refers to a newly installed device in the target computer room. That is, the target device is the communication device that was recently placed in the target computer room.

[0169] When installing a target device in the target data center, the target device type can be obtained.

[0170] After obtaining the target device type of the target device, proceed to step 502.

[0171] Step 502: Based on the target device type, select the target power consumption characteristic model corresponding to the target device from the power consumption characteristic model library.

[0172] The target power consumption characteristic model refers to the power consumption characteristic model that matches the target device type.

[0173] After obtaining the target device type, the corresponding target power consumption characteristic model can be selected from the power consumption characteristic model library based on the target device type. In this example, the power consumption characteristic model library stores power consumption characteristic models corresponding to different device types. After obtaining the target device type, the power consumption characteristic model corresponding to that target device type can be obtained from the power consumption characteristic model library. For example, the power consumption characteristic model library stores power consumption characteristic model 1 for device type A, power consumption characteristic model 2 for device type B, and power consumption characteristic model 3 for device type C. When the target device type is device type C, power consumption characteristic model 3 is considered the target power consumption characteristic model. When the target device type is device type A, power consumption characteristic model 1 is considered the target power consumption characteristic model, and so on.

[0174] It is understood that the above examples are merely examples listed to better understand the technical solutions of the embodiments of this application, and are not intended to be the only limitation on the embodiments.

[0175] After selecting the target power consumption characteristic model corresponding to the target device from the power consumption characteristic model library based on the target device type, proceed to step 503.

[0176] Step 503: Determine the simulated operating power consumption of the target device in the next day based on the target power consumption characteristic model.

[0177] After selecting the target power consumption characteristic model corresponding to the target device from the power consumption characteristic model library based on the target device type, the simulated operating power consumption of the target device in the next day can be determined based on the target power consumption characteristic model. Specifically, the reference device power consumption of the target device can be used as the input of the target power consumption characteristic model to predict the simulated operating power consumption of the target device in the next day.

[0178] After determining the simulated operating power consumption of the target device in the next day based on the target power consumption characteristic model, proceed to step 504.

[0179] Step 504: Based on the simulated operating power consumption and the system power consumption of the devices in the target computer room other than the target device, determine whether the power-on conditions of the target device are met.

[0180] Power-on conditions refer to the pre-set conditions used to determine whether newly installed communication equipment can be powered on.

[0181] After determining the simulated operating power consumption of the target device over the next day based on the target power consumption characteristic model, it can be determined whether the power-on conditions for the target device are met based on the simulated operating power consumption and the system power consumption of other devices in the target equipment room besides the target device. In this example, the power-on condition can be a power consumption threshold condition. By comparing the sum of the simulated operating power consumption and the system power consumption with the power consumption threshold, it can be determined whether the power-on conditions for the newly installed communication equipment are met. This implementation process can be combined with... Figure 6 Provide a detailed description.

[0182] Reference Figure 6 The flowchart illustrates the steps of a power-on condition determination method provided in an embodiment of this application, as shown below. Figure 6 As shown, the method for determining the power-on conditions may include steps 601 and 602.

[0183] Step 601: Compare the sum of the simulated operating power consumption and the system power consumption with the power consumption threshold.

[0184] In this embodiment, after estimating the simulated operating power consumption of the target device over the next day, the sum of the simulated operating power consumption and the system power consumption can be calculated. Specifically, when calculating the sum of power consumption, the simulated operating power consumption and the system power consumption for each time period within a day can be summed to obtain the sum of power consumption for multiple time periods within a day.

[0185] After obtaining the sum of the simulated operating power consumption and the system power consumption, the relationship between this sum and the power consumption threshold can be compared.

[0186] Step 602: Based on the comparison results, determine whether the power-on conditions of the target device are met.

[0187] After comparing the sum of the simulated operating power consumption and the system power consumption with the power consumption threshold, it can be determined whether the power-on conditions for the target device are met. Specifically, if the comparison result indicates that the sum of the simulated operating power consumption and the system power consumption is greater than or equal to the power consumption threshold, then the power-on conditions for the target device are not met. If the comparison result indicates that the sum of the simulated operating power consumption and the system power consumption is less than the power consumption threshold, then the power-on conditions for the target device are met.

[0188] This application embodiment sets a power consumption threshold to determine whether the power-on conditions for newly installed communication equipment are met, thereby ensuring the safe operation of the data center.

[0189] After determining that the power-on conditions of the target device are met, proceed to step 505.

[0190] Step 505: If the power-on conditions of the target device are met, power on the target device.

[0191] After determining that the power-on conditions of the target device are met, the target device can be powered on.

[0192] In evaluating the power-on of new equipment, the system modeling in this application not only considers the peak value of actual power consumption, but also adds time dimension factors. After comprehensive integration, it further explores the potential of infrastructure space and power resources, and ultimately achieves the goal of improving power utilization.

[0193] In this embodiment, if the power-on conditions of the target device are not met, the power supply environment of the target equipment room can be expanded, or the racking area of ​​the target device can be changed to accommodate the target device. This implementation process can be combined with... Figure 7 The following is a detailed description.

[0194] Reference Figure 7 The flowchart illustrates the steps of a method for changing the shelf area provided in an embodiment of this application. Figure 7 As shown, the method for changing the shelf area may include steps 701, 702 and 703.

[0195] Step 701: If the power-on conditions of the target device are not met, determine whether the power environment of the target area where the target computer room is located has the conditions for expansion.

[0196] In this embodiment, if the power-on conditions for the target device are not met, it can be determined whether the power environment of the target area where the target equipment room is located is suitable for expansion. Specifically, it can be determined whether the load of the existing communication equipment in the target area has reached a preset threshold. If so, it is determined that the target area is not suitable for expansion; otherwise, it is determined that the target area is suitable for expansion.

[0197] If it is determined that the power environment of the target area where the target data center is located is suitable for capacity expansion, proceed to step 702. If it is determined that the power environment of the target area where the target data center is located is not suitable for capacity expansion, proceed to step 703.

[0198] Step 702: If it is determined that the power environment of the target area is capable of expansion, the target computer room is expanded and the target equipment is powered on.

[0199] Once it is determined that the power supply environment of the target area is suitable for expansion, the target data center can be expanded. After expansion, the target equipment can be powered on to ensure the safe operation of the target data center.

[0200] Step 703: If it is determined that the power environment of the target area does not meet the expansion requirements, change the racking area of ​​the target device.

[0201] If the power environment of the target area does not meet the expansion requirements, the racking area of ​​the target equipment can be changed. The above process can be performed on the racking area to determine whether the power-on conditions are met. If they are met, the target equipment can be racked; otherwise, the expansion conditions can be determined.

[0202] This application embodiment powers on the target device by changing the racking area of ​​the target device or expanding the capacity of the target data center. This can ensure the safe operation of the target data center while powering on the newly added target device to meet the user's communication needs.

[0203] The power consumption characteristic model library establishment method provided in this application embodiment obtains real-time operating data of various types of equipment in the target equipment room at different time periods, performs characteristic modeling on the real-time operating data based on a time-varying parameter random volatility vector autoregression model, generates time-varying influence coefficient diagrams and impulse response diagrams corresponding to each type of equipment, determines the variable relationships between relevant variables of each type of equipment during operation based on the time-varying influence coefficient diagrams and impulse response diagrams, and constructs a power consumption characteristic model library based on application scenarios for each type of equipment in the target equipment room based on the variable relationships. This application embodiment constructs a power consumption characteristic model library based on application scenarios for each type of equipment in the target equipment room by combining real-time operating data of various types of equipment in the equipment room at different time periods. When powering on new equipment after modeling is completed, the actual power consumption of the equipment can be assigned a value, and the time dimension factor can be added for powering on evaluation. This can ensure the safe operation of power consumption in the communication equipment room while ultimately improving power utilization.

[0204] Reference Figure 8 The diagram shows a schematic representation of a power consumption characteristic model library establishment device provided in an embodiment of this application. Figure 8 As shown, the power consumption characteristic model library establishment device 800 may include the following modules:

[0205] The runtime data acquisition module 810 is used to acquire real-time runtime data of various types of equipment in the target computer room at different time periods.

[0206] The equipment coefficient diagram generation module 820 is used to perform characteristic modeling on the real-time operating data based on the time-varying parameter random volatility vector autoregression model, and generate time-varying influence coefficient diagrams and impulse response diagrams corresponding to various types of equipment.

[0207] The variable relationship determination module 830 is used to determine the variable relationships between relevant variables of various types of equipment during operation based on the time-varying influence coefficient diagram and the impulse response diagram;

[0208] The power consumption characteristic model library construction module 840 is used to construct a power consumption characteristic model library based on application scenarios for each type of device in the target computer room according to the variable relationship.

[0209] Optionally, the runtime data acquisition module 810 includes:

[0210] CPU utilization acquisition unit is used to acquire the CPU utilization of various types of equipment in the target computer room at different time periods.

[0211] The operating power acquisition unit is used to acquire the operating power of each type of device in the corresponding time period based on the device identifier corresponding to each type of device and the CPU utilization rate.

[0212] The real-time operation data acquisition unit is used to use the CPU utilization and the operating power as real-time operation data of various types of equipment in the target computer room at different time periods, and to store the real-time operation data in a time-series database.

[0213] Optionally, the equipment coefficient diagram generation module 820 includes:

[0214] The running data acquisition unit is used to acquire real-time running data of various types of devices in the time series database;

[0215] The equipment operation data acquisition unit is used to perform a stability test on the real-time operation data and acquire equipment operation data in the real-time operation data whose stability is lower than a set value.

[0216] The differential operation data acquisition unit is used to perform differential processing on the device operation data to obtain differential operation data;

[0217] The equipment coefficient diagram acquisition unit is used to input the differential operating data and the power consumption of the benchmark equipment corresponding to each type of equipment into the time-varying parameter random volatility vector autoregressive model to obtain the time-varying influence coefficient diagram and the impulse response diagram corresponding to each type of equipment.

[0218] Optionally, the power consumption characteristic model library construction module 840 includes:

[0219] The power consumption characteristic data acquisition unit is used to acquire power consumption characteristic data of various types of equipment in the target computer room throughout the day based on the variable relationship.

[0220] The power consumption characteristic model construction unit is used to construct, based on the power consumption characteristic data, a power consumption characteristic model corresponding to each type of device in the target computer room and the application scenario.

[0221] The power consumption characteristic model library generation unit is used to generate the power consumption characteristic model library according to the power consumption characteristic models corresponding to each type of device.

[0222] The application scenarios include at least one of: stable scenarios and non-stable scenarios;

[0223] The stable scenario is used to indicate a scenario where the power consumption is stable throughout the day, while the non-stable scenario is used to indicate a scenario where the power consumption has a set peak / valley value throughout the day.

[0224] Optionally, the device further includes:

[0225] The target device type acquisition module is used to acquire the target device type of the target device when the target device is installed in the target computer room;

[0226] The target power consumption model filtering module is used to filter the target power consumption characteristic model corresponding to the target device from the power consumption characteristic model library according to the target device type;

[0227] The simulated operating power consumption determination module is used to determine the simulated operating power consumption of the target device in the next day based on the target power consumption characteristic model.

[0228] The power-on condition determination module is used to determine whether the power-on conditions of the target device are met based on the simulated operating power consumption and the system power consumption of the devices in the target computer room other than the target device.

[0229] The target device power-on module is used to power on the target device when the power-on conditions of the target device are met.

[0230] Optionally, the power-on condition determination module includes:

[0231] A power consumption comparison unit is used to compare the sum of the simulated operating power consumption and the system power consumption with a power consumption threshold.

[0232] The power-on condition determination unit is used to determine whether the power-on conditions of the target device are met based on the comparison results.

[0233] Optionally, the device further includes:

[0234] The expansion condition determination module is used to determine whether the power environment of the target area where the target computer room is located meets the expansion conditions if the power supply conditions of the target device are not met.

[0235] The power-on processing module is used to perform capacity expansion processing on the target computer room and power-on processing on the target equipment when it is determined that the power environment of the target area has the capacity expansion conditions.

[0236] The racking area change module is used to change the racking area of ​​the target device when it is determined that the power environment of the target area does not meet the expansion conditions.

[0237] The power consumption characteristic model library establishment device provided in this application embodiment acquires real-time operating data of various types of equipment in the target equipment room at different time periods. Based on a time-varying parameter random volatility vector autoregression model, it models the characteristics of the real-time operating data, generating time-varying influence coefficient diagrams and impulse response diagrams corresponding to each type of equipment. Based on these diagrams, it determines the variable relationships between relevant variables of each type of equipment during operation. Based on these relationships, it constructs a power consumption characteristic model library for each type of equipment in the target equipment room based on application scenarios. This application embodiment, by combining real-time operating data of various types of equipment in the equipment room at different time periods to construct a power consumption characteristic model library for each type of equipment in the target equipment room based on application scenarios, allows for the assignment of actual power consumption values ​​to new equipment during power-on evaluation after modeling is completed. It also incorporates time dimension factors for power-on evaluation, ultimately improving power utilization while ensuring the safe operation of power consumption in the communication equipment room.

[0238] This application also provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the above-described method for establishing a power consumption characteristic model library.

[0239] Figure 9 A schematic diagram of the structure of an electronic device 900 according to an embodiment of the present invention is shown. Figure 9 As shown, the electronic device 900 includes a central processing unit (CPU) 901, which can perform various appropriate actions and processes according to computer program instructions stored in read-only memory (ROM) 902 or loaded from storage unit 908 into random access memory (RAM) 903. The RAM 903 can also store various programs and data required for the operation of the electronic device 900. The CPU 901, ROM 902, and RAM 903 are interconnected via bus 904. An input / output (I / O) interface 905 is also connected to bus 904.

[0240] Multiple components in electronic device 900 are connected to I / O interface 905, including: input unit 906, such as keyboard, mouse, microphone, etc.; output unit 907, such as various types of monitors, speakers, etc.; storage unit 908, such as disk, optical disk, etc.; and communication unit 909, such as network card, modem, wireless transceiver, etc. Communication unit 909 allows electronic device 900 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0241] The various processes and handling described above can be executed by processing unit 901. For example, the methods of any of the above embodiments can be implemented as computer software programs, which are tangibly contained in a computer-readable medium, such as storage unit 908. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 900 via ROM 902 and / or communication unit 909. When the computer program is loaded into RAM 903 and executed by CPU 901, one or more actions of the methods described above can be performed.

[0242] Additionally, embodiments of this application also provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned method for establishing a power consumption characteristic model library.

[0243] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0244] Those skilled in the art will understand that embodiments of this application can be provided as methods, apparatus, or computer program products. Therefore, embodiments of this application can take the form of entirely hardware embodiments, entirely software embodiments, or embodiments combining software and hardware aspects. Furthermore, embodiments of this application can take the form of computer program products implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0245] This application describes embodiments with reference to flowchart illustrations and / or block diagrams of methods, terminals (systems), and computer program products according to embodiments of this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing terminal to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0246] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing terminal to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0247] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal, causing a series of operational steps to be executed on the computer or other programmable terminal to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable terminal for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0248] Although preferred embodiments of the present application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the embodiments of the present application.

[0249] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal that includes said element.

[0250] The above provides a detailed description of a method for establishing a power consumption characteristic model library, a device for establishing a power consumption characteristic model library, an electronic device, and a computer-readable storage medium provided in this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for establishing a power consumption characteristic model library, characterized in that, include: Acquire real-time operating data of various types of equipment in the target computer room at different time periods, and store the real-time operating data in a time-series database; Based on the time-varying parameter stochastic volatility vector autoregressive model, the real-time operating data is modeled to generate time-varying influence coefficient diagrams and impulse response diagrams for each type of equipment. The time-varying influence coefficient diagrams represent the peak and trough states of the equipment during all-weather operation, and the impulse response diagrams represent the CPU utilization and actual operating power of the equipment in different time periods. Based on the time-varying influence coefficient diagram and the impulse response diagram, determine the variable relationships between relevant variables of each type of equipment during operation; Based on the variable relationships, a power consumption characteristic model library based on application scenarios is constructed for each type of equipment in the target computer room. When the target device is installed in the target computer room, the target device type of the target device is obtained; Based on the target device type, the target power consumption characteristic model corresponding to the target device is selected from the power consumption characteristic model library; Based on the target power consumption characteristic model, determine the simulated operating power consumption of the target device in the next day; Based on the simulated operating power consumption and the system power consumption of the equipment in the target computer room other than the target equipment, determine whether the power-on conditions of the target equipment are met; If the power-on conditions of the target device are met, the target device is powered on. The method of modeling the characteristics of the real-time operating data based on the time-varying parameter stochastic volatility vector autoregressive model to generate time-varying influence coefficient diagrams and impulse response diagrams corresponding to various types of equipment includes: Obtain real-time operating data of various types of devices from the time-series database; Perform a stability test on the real-time operating data, and obtain the device operating data in the real-time operating data whose stability is lower than a set value; The device operating data is differentially processed to obtain differential operating data; The differential operating data and the power consumption of the benchmark equipment corresponding to each type of equipment are input into the time-varying parameter random volatility vector autoregressive model to obtain the time-varying influence coefficient diagram and the impulse response diagram corresponding to each type of equipment. The step of constructing a power consumption characteristic model library based on application scenarios for each type of equipment in the target data center according to the variable relationships includes: Based on the aforementioned variable relationships, power consumption characteristics data of various types of equipment in the target computer room are obtained throughout the day. Based on the power consumption characteristic data, the Monte Carlo Markov chain algorithm is used to estimate the time-varying parameter values ​​in the time-varying parameter random volatility vector autoregressive model. The impulse response function is used to analyze the influence between different time-varying vectors in the time-varying parameter random volatility vector autoregressive model, and the power consumption characteristic model based on the application scenario corresponding to each type of equipment in the target data center is constructed. The power consumption characteristic model library is generated based on the power consumption characteristic models corresponding to each type of device.

2. The method according to claim 1, characterized in that, The acquisition of real-time operational data of various types of equipment in the target computer room at different time periods includes: Obtain the CPU utilization rate of various types of equipment in the target computer room at different time periods; Based on the device identifier corresponding to each type of device and the CPU utilization, obtain the operating power of each type of device in the corresponding time period; The CPU utilization and the operating power are used as real-time operating data of various types of equipment in the target computer room at different time periods, and the real-time operating data is stored in a time-series database.

3. The method according to claim 1, characterized in that, The application scenarios include at least one of: stable scenarios and non-stationary scenarios; The stable scenario is used to indicate a scenario where the power consumption is stable throughout the day, while the non-stable scenario is used to indicate a scenario where the power consumption has a set peak / valley value throughout the day.

4. The method according to claim 1, characterized in that, The step of determining whether the power-on conditions for the target device are met based on the simulated operating power consumption and the system power consumption of devices other than the target device in the target computer room includes: Compare the sum of the simulated operating power consumption and the system power consumption with the power consumption threshold. Based on the comparison results, determine whether the power-on conditions of the target device are met.

5. The method according to claim 1, characterized in that, After determining whether the power-on conditions for the target device are met based on the simulated operating power consumption and the system power consumption of devices other than the target device in the target computer room, the method further includes: If the power-on conditions of the target equipment are not met, determine whether the power environment of the target area where the target computer room is located has the conditions for expansion. If it is determined that the power environment of the target area is capable of expansion, the target computer room is expanded and the target equipment is powered on. If it is determined that the power supply environment of the target area does not meet the requirements for capacity expansion, the racking area of ​​the target device shall be changed.

6. A device for establishing a power consumption characteristic model library, characterized in that, include: The operation data acquisition module is used to acquire real-time operation data of various types of equipment in the target computer room at different time periods, and the real-time operation data is stored in a time-series database. The equipment coefficient diagram generation module is used to perform characteristic modeling on the real-time operating data based on the time-varying parameter random volatility vector autoregression model, and generate time-varying influence coefficient diagrams and impulse response diagrams corresponding to various types of equipment; the time-varying influence coefficient diagrams represent the peak and trough states of the equipment in all-weather operation, and the impulse response diagrams represent the CPU utilization and actual operating power of the equipment in different time periods. The variable relationship determination module is used to determine the variable relationships between relevant variables of various types of equipment during operation based on the time-varying influence coefficient diagram and the impulse response diagram; The power consumption characteristic model library construction module is used to construct a power consumption characteristic model library based on application scenarios for each type of device in the target computer room according to the variable relationship. The target device type acquisition module is used to acquire the target device type of the target device when the target device is installed in the target computer room; The target power consumption model filtering module is used to filter the target power consumption characteristic model corresponding to the target device from the power consumption characteristic model library according to the target device type; The simulated operating power consumption determination module is used to determine the simulated operating power consumption of the target device in the next day based on the target power consumption characteristic model. The power-on condition determination module is used to determine whether the power-on conditions of the target device are met based on the simulated operating power consumption and the system power consumption of the devices in the target computer room other than the target device. A target device power-on module is used to power on the target device when the power-on conditions of the target device are met. The equipment coefficient diagram generation module includes: The running data acquisition unit is used to acquire real-time running data of various types of devices in the time series database; The equipment operation data acquisition unit is used to perform a stability test on the real-time operation data and acquire equipment operation data in the real-time operation data whose stability is lower than a set value. The differential operation data acquisition unit is used to perform differential processing on the device operation data to obtain differential operation data; The equipment coefficient diagram acquisition unit is used to input the differential operating data and the power consumption of the benchmark equipment corresponding to each type of equipment into the time-varying parameter random volatility vector autoregressive model to obtain the time-varying influence coefficient diagram and the impulse response diagram corresponding to each type of equipment. The power consumption characteristic model library construction module includes: The power consumption characteristic data acquisition unit is used to acquire power consumption characteristic data of various types of equipment in the target computer room throughout the day based on the variable relationship. The power consumption characteristic model construction unit is used to estimate the time-varying parameter values ​​in the time-varying parameter random volatility vector autoregressive model based on the power consumption characteristic data using the Monte Carlo Markov chain algorithm, and to analyze the influence between different time-varying vectors in the time-varying parameter random volatility vector autoregressive model using the impulse response function, thereby constructing a power consumption characteristic model based on the application scenario for each type of equipment in the target computer room. The power consumption characteristic model library generation unit is used to generate the power consumption characteristic model library based on the power consumption characteristic models corresponding to each type of device.

7. An electronic device, characterized in that, include: A processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, implements the method for establishing a power consumption characteristic model library as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, When the instructions in the storage medium are executed by the processor of the electronic device, the electronic device is able to perform the power consumption characteristic model library establishment method according to any one of claims 1 to 5.

Images