Anomaly Detection Method, Device, Electronic Device and Medium for a Data Center

Abstract

An embodiment of the present invention provides an abnormal detection method, device, electronic device and medium for a data center. The method includes: obtaining target energy consumption information collected for each power-consuming device; then, predicting the energy consumption increase or decrease of each power-consuming device in a preset time period according to the target energy consumption information; when receiving an energy consumption abnormal warning information for the data center in the preset time period, determining an abnormal power-consuming device from multiple power-consuming devices according to the energy consumption increase or decrease situation. Through the embodiment of the present invention, it is realized to predict in advance the energy consumption increase or decrease situation of the power-consuming devices in the data center. When the data center has an abnormal energy consumption in the preset time period, the abnormal power-consuming device can be quickly determined based on the pre-obtained energy consumption increase or decrease situation, thereby improving the efficiency of locating the abnormal power-consuming device when the data center has an abnormal energy consumption, and there is no need to check one by one, saving manpower and material resources.

Description

Technical Field

[0001] The present invention relates to the technical field of anomaly detection, and particularly to an anomaly detection method, device, electronic device and medium for a data center. Background Art

[0002] A data center may refer to a building site where electronic information devices are centrally placed to provide an operating environment, and various power-consuming devices may be deployed therein, such as air conditioners, IT (Internet Technology) devices, lighting devices, etc.

[0003] With the continuous development of computer technology, the services provided by data centers are increasing; correspondingly, their energy consumption is also increasing; in order to reduce the energy consumption of data centers, an energy consumption energy-saving system can be deployed in the data centers to precisely control the power-consuming devices in the data centers, thereby reducing the energy consumption of each power-consuming device.

[0004] In practical applications, due to various reasons, the energy consumption of each power-consuming device may be abnormally high, such as a sudden increase / decrease in energy consumption; in the prior art, only the total energy consumption or partial energy consumption of the data center can be detected.

[0005] When an abnormality in the total energy consumption or partial energy consumption of the data center is detected, it is necessary to check each device to determine which power-consuming device has an abnormal energy consumption, and then repair the device with abnormal energy consumption.

[0006] However, this method of detecting abnormal energy consumption is inefficient and requires a large amount of manpower and material resources. Summary of the Invention

[0007] In view of the above problems, an anomaly detection method, device, electronic device and medium for a data center are proposed to overcome the above problems or at least partially solve the above problems, including:

[0008] An anomaly detection method for a data center, in which a plurality of power-consuming devices are deployed in the data center, and the method includes:

[0009] Obtain target energy consumption information collected for each power-consuming device;

[0010] Predict the energy consumption increase or decrease of each power-consuming device in a preset time period according to the target energy consumption information;

[0011] At the preset time period, if an energy consumption anomaly warning message for the data center is received, determine the abnormal power-consuming device from the plurality of power-consuming devices according to the energy consumption increase or decrease situation.

[0012] Optionally, the target energy consumption information includes current energy consumption information and historical energy consumption information. Predicting the energy consumption increase or decrease of each electrical device within a preset time period according to the target energy consumption information includes:

[0013] Determining a first energy consumption transfer probability of the multiple electrical devices in the previous time period and a second energy consumption transfer probability of the multiple electrical devices in the current time period according to the historical energy consumption information and the current energy consumption information;

[0014] Determining a third energy consumption transfer probability of the multiple electrical devices within the preset time period according to the first energy consumption transfer probability and the second energy consumption transfer probability;

[0015] Predicting the energy consumption increase or decrease of each electrical device within the preset time period according to the third energy consumption transfer probability.

[0016] Optionally, determining the energy consumption increase or decrease of each electrical device within the preset time period according to the third energy consumption transfer probability includes:

[0017] Determining a first power consumption of each electrical device in the current time period according to the current energy consumption information;

[0018] Predicting a second power consumption of each electrical device within the preset time period according to the third energy consumption transfer probability and the first power consumption;

[0019] Predicting the energy consumption increase or decrease of each electrical device within the preset time period according to the first power consumption and the second power consumption.

[0020] Optionally, the data center includes multiple regions. Determining abnormal electrical devices from the multiple electrical devices according to the energy consumption increase or decrease includes:

[0021] Determining an abnormal occurrence region from the multiple regions according to the energy consumption anomaly warning information;

[0022] Determining abnormal electrical devices from the electrical devices in the abnormal occurrence region according to the energy consumption increase or decrease.

[0023] Optionally, the energy consumption increase or decrease includes an energy consumption increase ratio or an energy consumption decrease ratio. The method further includes:

[0024] When an energy consumption increase ratio exceeds a first threshold, giving an abnormal warning prompt for the electrical device corresponding to the energy consumption increase ratio;

[0025] Or, when an energy consumption decrease ratio exceeds a second threshold, giving an abnormal warning prompt for the electrical device corresponding to the energy consumption decrease ratio.

[0026] An embodiment of the present invention further provides an abnormal detection device for a data center, where a plurality of power-consuming devices are deployed in the data center, and the device includes:

[0027] An acquisition module, configured to acquire target energy consumption information collected for each power-consuming device;

[0028] A prediction module, configured to predict the energy consumption increase or decrease situation of each power-consuming device within a preset time period according to the target energy consumption information;

[0029] An alarm module, configured to determine an abnormal power-consuming device from the plurality of power-consuming devices according to the energy consumption increase or decrease situation when receiving an energy consumption abnormal alarm information for the data center within a preset time period.

[0030] Optionally, the target energy consumption information includes current energy consumption information and historical energy consumption information, and the prediction module includes:

[0031] A first transition probability determination sub-module, configured to determine a first energy consumption transition probability of the plurality of power-consuming devices in the previous time period and a second energy consumption transition probability of the plurality of power-consuming devices in the current time period according to the historical energy consumption information and the current energy consumption information;

[0032] A second transition probability determination sub-module, configured to determine a third energy consumption transition probability of the plurality of power-consuming devices within a preset time period according to the first energy consumption transition probability and the second energy consumption transition probability;

[0033] An increase or decrease situation determination sub-module, configured to predict the energy consumption increase or decrease situation of each power-consuming device within a preset time period according to the third energy consumption transition probability.

[0034] Optionally, the increase or decrease situation determination sub-module is configured to determine a first power consumption of each power-consuming device in the current time period according to the current energy consumption information; determine a second power consumption of each power-consuming device within a preset time period according to the third energy consumption transition probability and the first power consumption; and determine the energy consumption increase or decrease situation of each power-consuming device within a preset time period according to the first power consumption and the second power consumption.

[0035] Optionally, the data center includes a plurality of regions, and the alarm module includes:

[0036] A region determination sub-module, configured to determine an abnormal occurrence region from the plurality of regions according to the energy consumption abnormal alarm information;

[0037] An abnormal device determination sub-module, configured to determine an abnormal power-consuming device from the power-consuming devices in the abnormal occurrence region according to the energy consumption increase or decrease situation.

[0038] Optionally, the energy consumption increase or decrease situation includes the energy consumption increase ratio or the energy consumption decrease ratio, and the device further includes:

[0039] A first prompting module, configured to perform an abnormal alarm prompt for the electrical equipment corresponding to the energy consumption increase ratio when an energy consumption increase ratio exceeds a first threshold;

[0040] A second prompting module, configured to perform an abnormal alarm prompt for the electrical equipment corresponding to the energy consumption decrease ratio when an energy consumption decrease ratio exceeds a second threshold.

[0041] An embodiment of the present invention further provides an electronic device, including a processor, a memory, and a computer program stored on the memory and capable of running on the processor. When the computer program is executed by the processor, the abnormal detection method of the data center as described above is implemented.

[0042] An embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the abnormal detection method of the data center as described above is implemented.

[0043] The embodiments of the present invention have the following advantages:

[0044] In the embodiments of the present invention, the target energy consumption information collected for each electrical equipment can be obtained first; then, according to the target energy consumption information, the energy consumption increase or decrease situation of each electrical equipment in a preset time period can be predicted; at the preset time period, if an abnormal energy consumption alarm information for the data center is received, then according to the energy consumption increase or decrease situation, the abnormal electrical equipment can be determined from multiple electrical equipment. Through the embodiments of the present invention, the prediction of the energy consumption increase or decrease situation of the electrical equipment in the data center in advance is realized. When the data center has abnormal energy consumption in the preset time period, the abnormal electrical equipment can be quickly determined based on the pre-obtained energy consumption increase or decrease situation, thereby improving the efficiency of locating the abnormal electrical equipment when the data center has abnormal energy consumption, and there is no need to check one by one, saving manpower and material resources. Description of the Drawings

[0045] In order to more clearly illustrate the technical solutions of the present invention, the drawings required for the description of the present invention will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0046] Figure 1 is a flowchart of the steps of an abnormal detection method for a data center according to an embodiment of the present invention;

[0047] Figure 2 is a schematic diagram of the energy consumption situation in a data according to an embodiment of the present invention;

[0048] Figure 3 is a flowchart of steps of another method for detecting anomalies in a data center according to an embodiment of the present invention;

[0049] Figure 4 is a block diagram of the structure of an anomaly detection device for a data center according to an embodiment of the present invention. Detailed implementation manners

[0050] To make the above objects, features, and advantages of the present invention more obvious and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific implementation manners. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0051] Refer to Figure 1 , which shows a flowchart of steps of a method for detecting anomalies in a data center according to an embodiment of the present invention. There are multiple power-consuming devices deployed in the data center, and the following steps may be included:

[0052] Step 101, obtain target energy consumption information collected for each power-consuming device.

[0053] Among them, the power-consuming device may refer to a device deployed in the data center that generates energy consumption, such as Figure 2 , the power-consuming devices may include chillers, humidifiers, precision air conditioners, IT devices, PDUs (Power Distribution Units), UPSs (Uninterruptible Power Supplies), lighting and auxiliary devices, switching devices / generators, etc.

[0054] When power is input, it can be respectively delivered to different power-consuming devices so that each power-consuming device can operate normally.

[0055] As an example, 24% of the power input to the data center can be delivered to the chiller, 3% to the humidifier, 15% to the precision air conditioner, 47% to the IT device, 3% to the PDU, 6% to the UPS, 2% to the lighting and auxiliary devices, and 1% to the switching device / generator.

[0056] For IT devices, they will also generate heat during operation, thus forming the heat in the indoor data center; for the data center, there may also be heat output generated based on the operation of each power-consuming device.

[0057] In practical applications, an energy-saving system can be deployed for a data center. The energy-saving system may include modules for different power-consuming devices. For example, an intelligent air-conditioning energy-saving management module for energy-saving of precision air conditioners, an intelligent hot-wind transfer energy-saving management module for chillers, humidifiers, lighting devices, switch devices / motors, an intelligent IT device energy-saving management module for IT devices, and an intelligent power distribution system energy-saving management module for UPS and PDU.

[0058] Each module in the energy-saving system can not only perform energy consumption management on power-consuming devices during operation, but also collect and store the energy consumption of power-consuming devices in real time.

[0059] When it is necessary to obtain the target energy consumption information of each power-consuming device, a preset program can be run to inspect each module in the energy-saving system to obtain the target energy consumption information of each power-consuming device; the target energy consumption information can refer to information related to energy consumption during the historical operation and / or current operation of each power-consuming device, such as electricity consumption, etc.

[0060] Step 102: Predict the energy consumption increase or decrease of each power-consuming device in a preset time period according to the target energy consumption information.

[0061] Among them, the preset time period can refer to a future time period relative to the current time period. For example, if the current time period is 00:00:00-23:59:59 on July 1, 2022, the preset time period can be 00:00:00-23:59:59 on July 2, 2022. The embodiments of the present invention do not limit this.

[0062] The energy consumption increase or decrease situation can refer to the increase or decrease ratio of the energy consumption of a power-consuming device in a time period relative to the energy consumption in the previous time period. For example, if the electricity consumption in a time period is 11 kwh and the electricity consumption in the previous time period is 10 kwh, the energy consumption increase or decrease situation can be: an increase of 10%.

[0063] After obtaining the target energy consumption information of each power-consuming device, the energy consumption increase or decrease of each power-consuming device can be predicted in a future preset time period according to the target energy consumption information corresponding to each power-consuming device.

[0064] Step 103: At the preset time period, if an energy consumption anomaly warning information for the data center is received, determine the abnormal power-consuming device from multiple power-consuming devices according to the energy consumption increase or decrease situation.

[0065] After determining the energy consumption increase or decrease of each electrical device in a preset time period in the future, when the system time of the data center reaches the preset time period, it can be detected whether an energy consumption anomaly warning message for the data center is received; the energy consumption anomaly warning message can be generated when the total energy consumption of the data center is abnormal or when the local energy consumption of the data center is abnormal.

[0066] In practical applications, the energy consumption anomaly warning message can only indicate that the total energy consumption of the data center is abnormal or the local energy consumption of the data center is abnormal; based on this energy consumption anomaly warning message, it is impossible to know which electrical device is abnormal.

[0067] Therefore, when receiving the energy consumption anomaly warning message for the data center, the energy consumption increase or decrease of each electrical device obtained by prior prediction can be combined to determine the electrical device with abnormal energy consumption.

[0068] Specifically, the electrical device with the largest increase in energy consumption can be used as the abnormal electrical device, or the electrical device with the largest decrease in energy consumption can be used as the abnormal electrical device; after determining the abnormal electrical device, it can be reported for the abnormal electrical device so that the administrator or the energy-saving management system can perform energy-saving management or inspection on the abnormal electrical device.

[0069] In the embodiment of the present invention, the target energy consumption information collected for each electrical device can be obtained first; then, according to the target energy consumption information, the energy consumption increase or decrease of each electrical device in the preset time period can be predicted; at the preset time period, if an energy consumption anomaly warning message for the data center is received, the abnormal electrical device can be determined from multiple electrical devices according to the energy consumption increase or decrease. Through the embodiment of the present invention, the energy consumption increase or decrease of the electrical devices in the data center is predicted in advance. When the energy consumption of the data center is abnormal in the preset time period, the abnormal electrical device can be quickly determined based on the previously obtained energy consumption increase or decrease, thereby improving the efficiency of locating the abnormal electrical device when the energy consumption of the data center is abnormal, and there is no need to check one by one, saving manpower and material resources.

[0070] Refer to Figure 3 , which shows the flowchart of the steps of another method for detecting anomalies in a data center according to an embodiment of the present invention, and may include the following steps:

[0071] Step 301, obtain the target energy consumption information collected for each electrical device, where the target energy consumption information includes the current energy consumption information and the historical energy consumption information.

[0072] When it is necessary to obtain the target energy consumption information of each electrical device, the modules in the energy-saving system can be inspected by running a preset program to obtain the target energy consumption information of each electrical device. Among them, the historical energy consumption information can be obtained from the storage sub-module in each module, and the current historical energy consumption information can be directly obtained from the sensors deployed at each electrical device through each module.

[0073] As an example, the energy consumption information of a precision air conditioner mainly includes the electricity consumed by the precision air conditioner; for a precision air conditioner, it can be used to lower the temperature of the computer room in a data center; reducing the energy consumption of the precision air conditioner can reduce the PUE (Power Usage Effectiveness) value of the computer room.

[0074] As another example, the energy consumption information of a chiller, a humidifier, lighting and auxiliary equipment, and a switch device / generator can also include the electricity consumed by the corresponding electrical devices; a chiller, a humidifier, lighting and auxiliary equipment, and a switch device / generator can form the air transfer heat hardware in a computer room; for the intelligent air transfer heat energy-saving management module, reducing the energy consumption of the air transfer heat hardware can reduce the PUE value of the computer room.

[0075] As yet another example, the energy consumption information of IT equipment can also include the electricity consumed by the IT equipment; for IT equipment, the CPU temperature data and the percentage unit of the process-level server load energy consumption data of the IT equipment in this period can be obtained through inspection; the main components of the energy consumption index of IT equipment in a computer room are the CPU temperature of the IT equipment and the process-level server load, so reducing these two parts can effectively reduce the PUE value of the IT equipment in the computer room.

[0076] As still another example, the energy consumption information of UPS and PDU can also include the electricity consumed by UPS and PDU; for UPS and PDU, the energy consumption data of the power distribution system in this period can be obtained through inspection. Secondly, the energy consumption index of the previous period is obtained by analyzing the historical energy consumption data of the power distribution system. The main components of the power distribution system in a computer room are UPS and PDU, so reducing the power consumption of UPS and PDU in the computer room can effectively reduce the PUE value of the computer room.

[0077] Among them, PUE is the abbreviation of Power Usage Effectiveness, which is the ratio of all the energy consumed by a data center to the energy consumed by the IT load, and is one of the most basic and effective indicators for evaluating the energy efficiency of a data center. The closer the PUE value is to 1, the higher the degree of greenness of a data center. When the above value exceeds 1, it means that the data center needs additional power overhead to support the IT load. Therefore, the higher the PUE value, the lower the overall efficiency of the data center.

[0078] Step 302: Determine the first energy consumption transfer probability of multiple electrical devices in the previous time period and the second energy consumption transfer probability of multiple electrical devices in the current time period according to the historical energy consumption information and the current energy consumption information.

[0079] Among them, the first energy consumption transfer probability may refer to the proportion of the energy consumption of multiple electrical devices in the previous time period transferred relative to the energy consumption in the period before the previous one; the first energy consumption transfer probability can be represented in the form of a transfer probability matrix. For example, the first energy consumption transfer probability can be [0.3, 0.7], which can indicate that 30% of the energy consumption used by the target electrical device in the period before the previous one is transferred to the target electrical device in the previous time period, and 70% is transferred to other devices in the data center except the target electrical device.

[0080] The second energy consumption transfer probability may refer to the proportion of the energy consumption of multiple electrical devices in the current time period transferred relative to the energy consumption in the previous time period.

[0081] The first energy consumption transfer probability can be determined according to a preset transfer probability model; when it is necessary to determine the first energy consumption transfer probability, the historical energy consumption information can be input into the preset transfer probability model to obtain the first energy consumption transfer probability; among them, the preset transfer probability model can be trained based on the historical energy consumption data of each electrical device.

[0082] At the same time, the historical energy consumption information and the current energy consumption information can also be input into the transfer probability model to obtain the second energy consumption transfer probability of multiple electrical devices in the current time period; the second energy consumption transfer probability may refer to the proportion of the energy consumption of multiple electrical devices in the current time period transferred relative to the energy consumption in the previous time period.

[0083] As an example, a preset transfer probability model can be obtained based on a Markov chain; the obtained transfer probability model can be expressed as follows:

[0084] X(k + 1) = X(k) × P;

[0085] In the formula: X(k) represents the state vector of the trend analysis and prediction object at time t = k, P represents the one-step transfer probability matrix, and X(k + 1) represents the state vector of the trend analysis and prediction object at time t = k + 1. A data set can be generated using a two-step transfer matrix.

[0086] It should be noted that the above first energy consumption transfer probability and second energy consumption transfer probability can include both the proportion of the energy consumption of an electrical device transferred to itself and the proportion of the energy consumption transferred to other electrical devices, and can also include the proportion of the energy consumption of other electrical devices transferred to this electrical device and the proportion of the energy consumption transferred to other electrical devices. The embodiments of the present invention do not limit this.

[0087] Step 303: Determine the third energy consumption transfer probability of multiple electrical devices within a preset time period according to the first energy consumption transfer probability and the second energy consumption transfer probability.

[0088] The third energy consumption transfer probability may refer to the ratio of the energy consumption of multiple electrical devices within a preset time period to the energy consumption transfer during the current time period.

[0089] After determining the first energy consumption transfer probability and the second energy consumption transfer probability, the third energy consumption transfer probability of multiple electrical devices within a future preset time period can be predicted according to the first energy consumption transfer probability and the second energy consumption transfer probability.

[0090] For example: the energy consumption transfer probability of the air conditioner in the previous time period is [0.3, 0.7], the energy consumption transfer probability of the air conditioner in the current time period is [0.6, 0.4], and the probability of other energy consumption transferring to the air conditioner in the current time period is [0.3, 0.7].

[0091] Calculate the occurrence probability of the energy consumption of the inspected air conditioner transferring to others in the next time period:

[0092] The probability of the air conditioner's energy consumption transferring in in the next time period is 0.3×0.6 + 0.3×0.7 = 0.39;

[0093] The probability of the air conditioner's energy consumption transferring out in the next time period is 0.3×0.4 + 0.7×0.7 = 0.61;

[0094] The energy consumption transfer probability of the air conditioner in the next time period is [0.39, 0.61].

[0095] In an embodiment of the present invention, the energy consumption increase and decrease situation can be predicted through the following sub-steps:

[0096] Sub-step 11: Determine the first power consumption of each electrical device within the current time period according to the current energy consumption information.

[0097] First, the first power consumption of each electrical device within the current time period can be determined from the current energy consumption information.

[0098] Sub-step 12: Predict the second power consumption of each electrical device within the preset time period according to the third energy consumption transfer probability and the first power consumption.

[0099] After determining the first power consumption, the second power consumption of each electrical device within the preset time period can be calculated according to the third energy consumption transfer probability and the first power consumption.

[0100] For example, the electrical equipment includes A and B; for equipment A, the third energy consumption transfer probability is [0.3, 0.7], and the first power consumption is 10 kwh; for equipment B, the third energy consumption transfer probability is [0.6, 0.4], and the first power consumption is 20 kwh. Then, the second power consumption of equipment A can be calculated as 10 kwh * 0.3 + 20 kwh * 0.4 = 11 kwh, and the second power consumption of equipment B is 10 kwh * 0.7 + 20 kwh * 0.6 = 19 kwh.

[0101] Sub-step 13: Predict the energy consumption increase or decrease of each electrical equipment within a preset time period based on the first power consumption and the second power consumption.

[0102] After determining the second power consumption, the energy consumption increase or decrease of each electrical equipment within the preset time period can be determined according to the increase or decrease ratio of the second power consumption relative to the first power consumption.

[0103] For example, for equipment A, the first power consumption is 10 kwh and the second power consumption is 11 kwh. Then, the energy consumption increase or decrease of equipment A can be determined as: the energy consumption increases by 10%.

[0104] In practical applications, the energy consumption increase or decrease includes the energy consumption increase ratio or the energy consumption decrease ratio; after determining the energy consumption abnormal probability of all electrical equipment within the preset time period, the embodiments of the present invention may further include the following steps:

[0105] When an energy consumption increase ratio exceeds the first threshold, an abnormal alarm prompt is given for the electrical equipment corresponding to the energy consumption increase ratio; or, when an energy consumption decrease ratio exceeds the second threshold, an abnormal alarm prompt is given for the electrical equipment corresponding to the energy consumption decrease ratio.

[0106] Specifically, the energy consumption increase or decrease can indicate whether the energy consumption usage of the electrical equipment within the preset time period is normal; when the energy consumption increase ratio of an electrical equipment exceeds the first threshold, it can indicate that the electrical equipment may have abnormal energy consumption in the future preset time period. At this time, an abnormal alarm prompt can be given in advance for the electrical equipment whose energy consumption increase ratio exceeds the first threshold. For example: an abnormal report is made for the electrical equipment so that the administrator or the energy-saving system can pre-control or check the electrical equipment in advance.

[0107] Or, when the energy consumption decrease ratio of an electrical equipment exceeds the second threshold, it can also indicate that the electrical equipment may have abnormal energy consumption in the future preset time period. For example: it may be that the electrical equipment fails to work properly, resulting in a decrease in energy consumption.

[0108] At this time, an abnormal alarm prompt can be given in advance for the electrical equipment whose energy consumption decrease ratio exceeds the second threshold.

[0109] Step 304: The data center includes multiple regions. During a preset time period, if an energy consumption anomaly warning message for the data center is received, then based on the energy consumption anomaly warning message, determine the region where the anomaly occurs from the multiple regions.

[0110] In practical applications, the energy consumption anomaly warning message may indicate that there is an energy consumption anomaly in the electrical equipment in a certain region of the data center; at this time, the region where the energy consumption anomaly occurs, that is, the anomaly occurrence region, can be determined from the multiple regions of the data center first according to the energy consumption anomaly warning message.

[0111] Step 305: Determine the abnormal electrical equipment from the electrical equipment in the anomaly occurrence region according to the energy consumption increase and decrease conditions.

[0112] Then, determine one or more electrical equipment in the anomaly occurrence region; and then, based on the energy consumption increase and decrease conditions of these one or more electrical equipment, determine the abnormal electrical equipment. For example: when the energy consumption anomaly warning message indicates that there is an abnormal increase in energy consumption, the electrical equipment with the highest increase in energy consumption among one or more electrical equipment can be used as the abnormal electrical equipment. Or, when the energy consumption anomaly warning message indicates that there is an abnormal decrease in energy consumption, the electrical equipment with the largest decrease in energy consumption among one or more electrical equipment can be used as the abnormal electrical equipment. The embodiments of the present invention are not limited thereto.

[0113] After determining the abnormal electrical equipment, it can be reported for the abnormal electrical equipment, so that the administrator or the energy-saving management system can perform energy-saving management or inspection on the abnormal electrical equipment.

[0114] In an embodiment of the present invention, the fusion computer (cloud operating system software) technology can be used to construct a virtual data center for the data center; the virtual data center can map an energy-saving system, modules in the energy-saving system, and the electrical equipment managed by each module. The servers in the data center can complete the information interaction with the virtual data center through the open api (open platform).

[0115] For each data center, it can be connected to the central server respectively to receive the transfer probability model issued by the central server; specifically, the central server can send a flag (issued instruction) to the servers corresponding to each local data center. After receiving the flag, the local servers access the historical database identification field flag in the local deployed server to check whether the transfer probability model is deployed locally. flag = 1 indicates that it is deployed, and flag = 0 indicates that it is not deployed; if it is not deployed, the program sends an instruction to notify the central server to request the transfer probability model to be issued. After receiving the request instruction, the central server sends the transfer probability model to the local server.

[0116] After receiving the transfer probability model, the local server can perform privacy data training; that is, utilize the distributed virtual switching characteristics of the virtual data center constructed by fusioncomputer to obtain the energy consumption information for each electrical device mapped in the historical database corresponding to the local virtual data center.

[0117] When a local anomaly occurs in the energy consumption link of the data center, the identifier corresponding to the abnormal electrical device determined based on the above method can be transmitted to the virtual data center through the open api; the virtual data center performs energy consumption management on the corresponding abnormal electrical device through the module corresponding to the mapped energy-saving system.

[0118] As an example, the energy-saving system of the data center in the embodiments of the present invention can perform energy consumption control and energy conservation and emission reduction on each electrical device based on AI (Artificial Intelligence) technology, and the embodiments of the present invention do not limit this.

[0119] In the embodiments of the present invention, the target energy consumption information collected for each electrical device can be obtained first, and the target energy consumption information includes the current energy consumption information and the historical energy consumption information; then, based on the historical energy consumption information and the current energy consumption information, the first energy consumption transfer probability of multiple electrical devices in the previous time period and the second energy consumption transfer probability of multiple electrical devices in the current time period are determined; based on the first energy consumption transfer probability and the second energy consumption transfer probability, the third energy consumption transfer probability of multiple electrical devices in a preset time period is determined; the data center includes multiple regions, and when an energy consumption anomaly warning information for the data center is received during the preset time period, the anomaly occurrence region is determined from the multiple regions according to the energy consumption anomaly warning information; according to the energy consumption increase and decrease situation, the abnormal electrical device is determined from the electrical devices in the anomaly occurrence region. Through the embodiments of the present invention, the prediction of the energy consumption increase and decrease situation of the electrical devices in the data center is realized in advance. When an energy consumption anomaly occurs in the data center during the preset time period, the abnormal electrical device can be quickly determined based on the previously obtained energy consumption increase and decrease situation, thereby improving the efficiency of locating the abnormal electrical device when the data center has an energy consumption anomaly, and there is no need to check one by one, saving human and material resources.

[0120] It should be noted that for the method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that the embodiments of the present invention are not limited by the described action sequence, because according to the embodiments of the present invention, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily essential for the embodiments of the present invention.

[0121] Refer toFigure 4 , which shows a schematic structural diagram of an abnormal detection device for a data center according to an embodiment of the present invention. A plurality of power-consuming devices are deployed in the data center, and it may include the following modules:

[0122] An acquisition module 401, configured to acquire target energy consumption information collected for each power-consuming device;

[0123] A prediction module 402, configured to predict the energy consumption increase or decrease situation of each power-consuming device in a preset time period according to the target energy consumption information;

[0124] An alarm module 403, configured to determine an abnormal power-consuming device from a plurality of power-consuming devices according to the energy consumption increase or decrease situation when receiving an abnormal energy consumption alarm message for the data center at a preset time period.

[0125] In an optional embodiment of the present invention, the target energy consumption information includes current energy consumption information and historical energy consumption information. The prediction module 402 includes:

[0126] A first transition probability determination sub-module, configured to determine a first energy consumption transition probability of a plurality of power-consuming devices in the previous time period and a second energy consumption transition probability of a plurality of power-consuming devices in the current time period according to the historical energy consumption information and the current energy consumption information;

[0127] A second transition probability determination sub-module, configured to determine a third energy consumption transition probability of a plurality of power-consuming devices in a preset time period according to the first energy consumption transition probability and the second energy consumption transition probability;

[0128] An increase or decrease situation determination sub-module, configured to predict the energy consumption increase or decrease situation of each power-consuming device in a preset time period according to the third energy consumption transition probability.

[0129] In an optional embodiment of the present invention, the increase or decrease situation determination sub-module is configured to determine a first power consumption of each power-consuming device in the current time period according to the current energy consumption information; determine a second power consumption of each power-consuming device in a preset time period according to the third energy consumption transition probability and the first power consumption; and determine the energy consumption increase or decrease situation of each power-consuming device in a preset time period according to the first power consumption and the second power consumption.

[0130] In an optional embodiment of the present invention, the data center includes a plurality of regions. The alarm module 403 includes:

[0131] A region determination sub-module, configured to determine an abnormal occurrence region from a plurality of regions according to the abnormal energy consumption alarm message;

[0132] An abnormal device determination sub-module, configured to determine an abnormal power-consuming device from the power-consuming devices in the abnormal occurrence region according to the energy consumption increase or decrease situation.

[0133] In an alternative embodiment of the present invention, the energy consumption increase and decrease situation includes the energy consumption increase ratio or the energy consumption decrease ratio, and the device further includes:

[0134] A first prompt module, configured to perform an abnormal alarm prompt for the electrical equipment corresponding to an energy consumption increase ratio when an energy consumption increase ratio exceeds a first threshold;

[0135] A second prompt module, configured to perform an abnormal alarm prompt for the electrical equipment corresponding to an energy consumption decrease ratio when an energy consumption decrease ratio exceeds a second threshold.

[0136] In an embodiment of the present invention, the target energy consumption information collected for each electrical equipment can be obtained first; then, according to the target energy consumption information, the energy consumption increase and decrease situation of each electrical equipment in a preset time period can be predicted; at the preset time period, if an abnormal energy consumption alarm information for the data center is received, then according to the energy consumption increase and decrease situation, the abnormal electrical equipment can be determined from multiple electrical equipment. Through the embodiment of the present invention, the prediction of the energy consumption increase and decrease situation of the electrical equipment in the data center is realized in advance. When the data center has abnormal energy consumption in the preset time period, the abnormal electrical equipment can be quickly determined based on the previously obtained energy consumption increase and decrease situation, thereby improving the efficiency of locating the abnormal electrical equipment when the data center has abnormal energy consumption, and there is no need to check one by one, saving manpower and material resources.

[0137] An embodiment of the present invention further provides an electronic device, including a processor, a memory, and a computer program stored on the memory and capable of running on the processor. When the computer program is executed by the processor, the above abnormal detection method of the data center is implemented.

[0138] An embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by the processor, the above abnormal detection method of the data center is implemented.

[0139] For the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple. For the relevant parts, refer to the partial description of the method embodiment.

[0140] Each embodiment in this specification is described in a progressive manner. The key point of each embodiment is to illustrate the differences from other embodiments. The same or similar parts among the embodiments can be referred to each other.

[0141] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, apparatus, or computer program product. Therefore, the embodiments of the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the embodiments of the present invention can take the form of a computer program product 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.

[0142] The embodiments of the present invention are described with reference to the flowcharts and / or block diagrams of methods, terminal devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing terminal devices generate a device for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0143] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing terminal device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device that implements the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0144] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device, such that a series of operation steps are executed on the computer or other programmable terminal device to generate a computer-implemented process, so that the instructions executed on the computer or other programmable terminal device provide steps for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0145] Although the preferred embodiments of the embodiments of the present invention have been described, those skilled in the art can make additional changes and modifications to these embodiments once they know the basic creative concept. Therefore, the appended claims are intended to be construed as including the preferred embodiments and all changes and modifications falling within the scope of the embodiments of the present invention.

[0146] Finally, it should also be noted that in this text, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or terminal device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or terminal device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or terminal device comprising the said element.

[0147] The above has introduced in detail an anomaly detection method, device, electronic device and medium for a data center. In this text, specific examples are used to elaborate on the principle and implementation manner of the present invention. The description of the above embodiments is only used to help understand the method and its core idea of the present invention; at the same time, for those of ordinary skill in the art, according to the idea of the present invention, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to the present invention.

Claims

1. An abnormal detection method for a data center, characterized in that, a plurality of power-consuming devices are deployed in the data center, and the method includes: acquiring target energy consumption information collected for each power-consuming device; predicting the energy consumption increase or decrease of each power-consuming device in a preset time period according to the target energy consumption information; when receiving an energy consumption abnormal alarm message for the data center in the preset time period, determining an abnormal power-consuming device from the plurality of power-consuming devices according to the energy consumption increase or decrease situation; wherein, the target energy consumption information includes current energy consumption information and historical energy consumption information, and predicting the energy consumption increase or decrease of each power-consuming device in a preset time period according to the target energy consumption information includes: determining a first energy consumption transfer probability of the plurality of power-consuming devices in the previous time period and a second energy consumption transfer probability of the plurality of power-consuming devices in the current time period according to the historical energy consumption information and the current energy consumption information, the first energy consumption transfer probability being the ratio of the energy consumption of the plurality of power-consuming devices in the previous time period to the energy consumption in the period before the previous period, and the second energy consumption transfer probability being the ratio of the energy consumption of the plurality of power-consuming devices in the current time period to the energy consumption in the previous time period; determining a third energy consumption transfer probability of the plurality of power-consuming devices in the preset time period according to the first energy consumption transfer probability and the second energy consumption transfer probability, the third energy consumption transfer probability being the ratio of the energy consumption of the plurality of power-consuming devices in the preset time period to the energy consumption in the current time period; predicting the energy consumption increase or decrease of each power-consuming device in the preset time period according to the third energy consumption transfer probability.

2. The method according to claim 1, characterized in that, determining the energy consumption increase or decrease of each power-consuming device in the preset time period according to the third energy consumption transfer probability includes: determining a first power consumption of each power-consuming device in the current time period according to the current energy consumption information; predicting a second power consumption of each power-consuming device in the preset time period according to the third energy consumption transfer probability and the first power consumption; predicting the energy consumption increase or decrease of each power-consuming device in the preset time period according to the first power consumption and the second power consumption.

3. The method according to claim 1, characterized in that, the data center includes a plurality of regions, and determining an abnormal power-consuming device from the plurality of power-consuming devices according to the energy consumption increase or decrease situation includes: determining an abnormal occurrence region from the plurality of regions according to the energy consumption abnormal alarm message; determining an abnormal power-consuming device from the power-consuming devices in the abnormal occurrence region according to the energy consumption increase or decrease situation.

4. The method according to claim 1, characterized in that, the energy consumption increase or decrease situation includes an energy consumption increase ratio or an energy consumption decrease ratio, and the method further includes: when an energy consumption increase ratio exceeds a first threshold, performing an abnormal alarm prompt for the power-consuming device corresponding to the energy consumption increase ratio; or, when an energy consumption decrease ratio exceeds a second threshold, performing an abnormal alarm prompt for the power-consuming device corresponding to the energy consumption decrease ratio.

5. An abnormal detection device for a data center, characterized in that, A plurality of power-consuming devices are deployed in the data center, and the device includes: An acquisition module, configured to acquire target energy consumption information collected for each power-consuming device; A prediction module, configured to predict the energy consumption increase or decrease of each power-consuming device in a preset time period according to the target energy consumption information; An alarm module, configured to, when receiving an energy consumption anomaly alarm information for the data center in a preset time period, determine an abnormal power-consuming device from the plurality of power-consuming devices according to the energy consumption increase or decrease; Wherein, the target energy consumption information includes current energy consumption information and historical energy consumption information, and the prediction module includes: A first transition probability determination sub-module, configured to determine a first energy consumption transition probability of the plurality of power-consuming devices in the previous time period and a second energy consumption transition probability of the plurality of power-consuming devices in the current time period according to the historical energy consumption information and the current energy consumption information, where the first energy consumption transition probability is the ratio of the energy consumption of the plurality of power-consuming devices in the previous time period to the energy consumption in the time period before the previous one, and the second energy consumption transition probability is the ratio of the energy consumption of the plurality of power-consuming devices in the current time period to the energy consumption in the previous time period; A second transition probability determination sub-module, configured to determine a third energy consumption transition probability of the plurality of power-consuming devices in the preset time period according to the first energy consumption transition probability and the second energy consumption transition probability, where the third energy consumption transition probability is the ratio of the energy consumption of the plurality of power-consuming devices in the preset time period to the energy consumption in the current time period; An increase or decrease situation determination sub-module, configured to predict the energy consumption increase or decrease of each power-consuming device in the preset time period according to the third energy consumption transition probability.

6. The device according to claim 5, wherein, the increase or decrease situation determination sub-module is configured to determine a first power consumption of each power-consuming device in the current time period according to the current energy consumption information; determine a second power consumption of each power-consuming device in the preset time period according to the third energy consumption transition probability and the first power consumption; and determine the energy consumption increase or decrease of each power-consuming device in the preset time period according to the first power consumption and the second power consumption.

7. An electronic device, wherein, it includes a processor, a memory, and a computer program stored on the memory and capable of running on the processor, and when the computer program is executed by the processor, it implements the abnormal detection method of the data center according to any one of claims 1 to 4.

8. A computer-readable storage medium, wherein, a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, it implements the abnormal detection method of the data center according to any one of claims 1 to 4.

Images