Machine room energy consumption management method and device

Abstract

The application discloses a kind of machine room energy consumption management method and device.Therein, the method includes: obtaining the first equipment energy consumption data of multiple equipment in machine room in the first time period;First equipment energy consumption data is converted into the question set of the preset data format matched with energy consumption prediction model input;Using the energy consumption prediction model trained based on historical equipment energy consumption data, the second equipment energy consumption data in the second time period is obtained by analyzing the question set, and the second time period is the time period after the first time period;Using the preset energy consumption anomaly threshold condition, the second equipment energy consumption data is checked, and if the checking result indicates that there is energy consumption anomaly, an abnormal alarm information is generated;First equipment energy consumption data, second equipment energy consumption data and checking result are recorded to the energy consumption data wide table corresponding to machine room.The application solves the technical problem that energy consumption data of a large number of machine room equipment is difficult to be efficiently and accurately controlled in the machine room operation and maintenance scene.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of machine room operation and maintenance, in particular to a machine room energy consumption management method and device. BACKGROUND

[0002] The monitoring of machine room dynamic environment energy consumption is usually realized through a machine room dynamic environment monitoring system. The system can monitor the energy consumption of various devices in the machine room in real time, including voltage, current, power and other factors, and provide energy consumption optimization suggestions through data analysis. In addition, the machine room dynamic environment monitoring system can also monitor the environmental parameters of the machine room, such as temperature and humidity, to ensure that these environmental factors do not affect the operating efficiency and service life of the equipment, thereby indirectly affecting the energy consumption. These problems not only increase the energy consumption of the machine room, but also may affect the normal operation and service life of the machine room equipment, so effective management and technical measures need to be taken to solve them. In the current field of machine room dynamic environment monitoring and energy consumption management, fixed threshold early warning and preliminary energy consumption data analysis are relied on, but there are challenges such as insufficient energy consumption prediction accuracy, low data processing efficiency, superficial data value evaluation, limitations of artificial intelligence algorithms, and weak data security and privacy protection. These technical bottlenecks limit the accurate identification and management optimization of energy consumption anomalies, especially in dealing with the complex and variable operating environment of data centers, it is difficult to maximize the efficiency, thereby affecting the energy utilization efficiency and the realization of energy saving and emission reduction targets.

[0003] At present, there is no effective solution to the above problems. SUMMARY

[0004] The embodiments of the present application provide a machine room energy consumption management method and device to at least solve the technical problem that it is difficult to efficiently and accurately control the energy consumption data of a large number of machine room devices in the machine room operation and maintenance scenario.

[0005] According to an aspect of an embodiment of the present application, a machine room energy consumption management method is provided, comprising: obtaining first device energy consumption data of a plurality of devices in a machine room in a first time period; converting the first device energy consumption data into a question set in a preset data format matching the input of an energy consumption prediction model; analyzing the question set using the energy consumption prediction model to obtain second device energy consumption data in a second time period output by the energy consumption prediction model, wherein the energy consumption prediction model is trained based on historical device energy consumption data, and the second time period is a time period after the first time period; verifying the second device energy consumption data using a preset energy consumption anomaly threshold condition, and generating an abnormal alarm information in the case that the verification result indicates that there is an energy consumption anomaly; recording the first device energy consumption data, the second device energy consumption data and the verification result to a corresponding energy consumption data wide table of the machine room.

[0006] Optionally, the types of the devices in the machine room include air conditioning devices, air-moving heat devices, information technology devices, and power distribution devices, wherein the types of the air-moving heat devices include water chillers, humidifiers, lighting auxiliary devices, and switch devices, and the types of the power distribution devices include uninterruptible power supplies and power distribution units, and the device energy consumption data of the air conditioning devices at least includes power consumption, and the device energy consumption data of the air-moving heat devices at least includes temperature and humidity data and power consumption, and the device energy consumption data of the information technology devices at least includes central processing unit temperature and process-level server load, and the device energy consumption data of the power distribution devices at least includes power consumption.

[0007] Optionally, the converting the first device energy consumption data into the question set in the preset data format matching the input of the energy consumption prediction model comprises: performing data preprocessing on the first device energy consumption data, wherein the data preprocessing manner comprises at least one of the following: outlier deletion, null value interpolation, and data standardization processing; determining a first total energy consumption of each type of device according to the first device energy consumption data, and determining a sum of all the first total energy consumptions as a second total energy consumption of the machine room; determining a first energy consumption proportion of each type of device according to a ratio of the first total energy consumption of the type of device to the second total energy consumption, and determining a first energy use efficiency of the machine room according to a ratio of the second total energy consumption to the first total energy consumption corresponding to the information technology devices; generating question data of each type of device according to the first total energy consumption and the first energy consumption proportion of the type of device, and generating the question set according to all the question data and the first energy use efficiency.

[0008] Optionally, the training process of the energy consumption prediction model comprises: constructing an initial prediction model; obtaining a plurality of sets of historical device energy consumption data of a plurality of devices in the machine room in a plurality of continuous historical time periods, wherein one historical time period corresponds to one set of historical device energy consumption data; for each historical time period, determining a third total energy consumption and a second energy consumption proportion of each type of device and a second energy use efficiency of the machine room according to the historical device energy consumption data corresponding to the historical time period, and generating a training sample by combining the third total energy consumption, the second energy consumption proportion, and the second energy use efficiency, and taking the third total energy consumption, the second energy consumption proportion, and the second energy use efficiency corresponding to a next historical time period of the historical time period as a sample label of the training sample; and iteratively training the initial prediction model according to the plurality of training samples and the sample labels to obtain the energy consumption prediction model.

[0009] Optionally, the second device energy consumption data comprises: fourth total energy consumption of each type of device in the machine room in a future second time period, a third energy consumption proportion, a third energy use efficiency of the machine room in the future second time period, and the second device energy consumption data is checked by using a preset energy consumption anomaly threshold condition, which comprises: comparing the third energy use efficiency with a preset efficiency threshold; in the case that the third energy use efficiency is greater than the preset efficiency threshold, it is determined that the machine room has energy consumption anomaly, and the energy consumption growth rate of each type of device is determined according to the fourth total energy consumption and the first total energy consumption of each type of device, or the third energy consumption proportion and the first total energy consumption of each type of device; and it is determined that the device with an energy consumption growth rate greater than a preset amplitude threshold has energy consumption anomaly.

[0010] Optionally, after the abnormal alarm information is generated, the method further comprises: obtaining an inspection result of the energy consumption abnormal device; in the case that the inspection result indicates that the energy consumption abnormal device has anomaly, using the first device energy consumption data as a positive sample to update the energy consumption prediction model; and in the case that the inspection result indicates that the energy consumption abnormal device has no anomaly, adjusting the preset amplitude threshold, and using the first device energy consumption data as a negative sample to update the energy consumption prediction model.

[0011] Optionally, the energy consumption data wide table is used to record energy consumption information of each device in the machine room, wherein each device corresponds to at least the following fields: device identifier, device type, device installation location, time period identifier, first device energy consumption data of the corresponding time period, question set, second device energy consumption data, energy consumption anomaly threshold condition, checking result, and abnormal inspection result.

[0012] Optionally, in response to a data query instruction of a target object, a target field matched with a keyword in the data query instruction is retrieved from the energy consumption data wide table, and a target device corresponding to the target field is determined; and energy consumption information of the target device is fed back to the target object.

[0013] Optionally, the method further comprises: scoring the first device energy consumption data from multiple dimensions to obtain a sub-score of each dimension, wherein the multiple dimensions at least comprise: data integrity dimension, data specification dimension, data accuracy dimension, data timeliness dimension, data richness dimension, and data coverage dimension; determining a weight coefficient of each dimension, and performing weighted summation on the sub-scores of the multiple dimensions according to the weight coefficients to obtain a comprehensive score of the first device energy consumption data; and recording the comprehensive score to the energy consumption data wide table.

[0014] According to another aspect of the embodiments of the present application, a device room energy consumption management device is also provided, which comprises: an acquisition module configured to acquire first device energy consumption data of a plurality of devices in a device room in a current first time period; a conversion module configured to convert the first device energy consumption data into a question set in a preset data format matched with an input of an energy consumption prediction model; a prediction module configured to analyze the question set by using the energy consumption prediction model to obtain second device energy consumption data in a second time period output by the energy consumption prediction model, wherein the energy consumption prediction model is obtained based on historical device energy consumption data, and the second time period is a time period after the first time period; an analysis module configured to check the second device energy consumption data by using a preset energy consumption anomaly threshold condition, and generate an anomaly alarm information in a case where a check result indicates that there is an energy consumption anomaly; and a statistical module configured to record the first device energy consumption data, the second device energy consumption data and the check result to a wide table of energy consumption data corresponding to the device room.

[0015] According to another aspect of the embodiments of the present application, a computer program product is also provided, which comprises a computer program, wherein the computer program is executed by a processor to implement the device room energy consumption management method described above.

[0016] According to another aspect of the embodiments of the present application, an electronic device is also provided, which comprises a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the device room energy consumption management method described above by using the computer program.

[0017] In the embodiments of the present application, the energy consumption data of the device is acquired in real time, including temperature, humidity, power and other key indicators. The first device energy consumption data is converted into a question set format that can be understood by the energy consumption prediction model. This step ensures the standardization and structuring of the model input data, which helps to improve the prediction accuracy of the model. The energy consumption prediction model trained based on historical device energy consumption data can predict the second device energy consumption data in the future second time period. This prediction function provides forward-looking decision support for the management of the computer room, enabling managers to plan energy use in advance and optimize device operation strategies, thereby achieving the purpose of energy saving and emission reduction. The second device energy consumption data predicted is checked using a pre-set energy consumption anomaly threshold condition, which can identify energy consumption anomalies in a timely manner. Once an energy consumption anomaly is detected, the system will automatically generate an abnormal alarm information to notify relevant personnel for processing. The first device energy consumption data, the second device energy consumption data and the checking result of the anomaly detection are recorded in the energy consumption data wide table corresponding to the computer room. This not only facilitates long-term storage and historical comparison of data, but also provides a basis for subsequent deep data analysis. Through the wide table, energy consumption data can be analyzed in multiple dimensions, including but not limited to device health status analysis, energy consumption trend prediction, fault mode identification, etc. This further optimizes the energy management and device maintenance of the computer room, and thus solves the technical problem of difficult efficient and accurate management and control of a large number of computer room devices in the computer room operation and maintenance scenario. BRIEF DESCRIPTION OF DRAWINGS

[0018] The accompanying drawings, which are included to provide a further understanding of the present application, constitute a part of this application and help to explain the present application, but do not limit the present application in any way. In the drawings:

[0019] Figure 1 is a flow diagram of an optional computer room energy consumption management method according to an embodiment of the present application;

[0020] Figure 2 is a structural diagram of an optional computer room energy consumption management device according to an embodiment of the present application;

[0021] Figure 3 is a structural diagram of an optional electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0022] In order for those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should be within the scope of protection of the present application.

[0023] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the drawings are used to distinguish similar objects, and do not necessarily describe a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0024] In order to better understand the embodiments of the present application, first, the part of the nouns or terms appearing in the description of the embodiments of the present application are translated and explained as follows:

[0025] Query Set: is the first device energy consumption data after preprocessing and conversion, which is formatted to match the structure of the energy consumption prediction model input. Specifically, the query set is a series of feature vectors or data points constructed to enable the model to understand and process, each query data point usually contains multiple dimensions of information reflecting the energy consumption of various devices in the computer room and their related attributes in a specific time period.

[0026] The information collected in the embodiments of the present application is information and data authorized by the user or authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of related data comply with relevant laws, regulations and standards of relevant countries and regions, necessary security measures are taken, do not violate public order and good customs, and provide corresponding operation portal for user to choose authorization or refusal.

[0027] Embodiment 1

[0028] According to the embodiments of the present application, a computer room energy consumption management method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described herein can be executed in an order different from that herein.

[0029] Figure 1 is a flowchart of a computer room energy consumption management method according to an embodiment of the present application, as shown in Figure 1 The method comprises the following steps:

[0030] Step S102, obtaining first device energy consumption data of a plurality of devices in a machine room in a first time period;

[0031] Step S104, converting the first device energy consumption data into a question set in a preset data format matched with an input of an energy consumption prediction model;

[0032] Step S106, analyzing the question set by using the energy consumption prediction model to obtain second device energy consumption data in a second time period output by the energy consumption prediction model, wherein the energy consumption prediction model is trained based on historical device energy consumption data, and the second time period is a time period after the first time period;

[0033] Step S108, verifying the second device energy consumption data by using a preset energy consumption anomaly threshold condition, and generating an abnormal alarm information in a case where a verification result indicates that there is an energy consumption anomaly;

[0034] Step S110, recording the first device energy consumption data, the second device energy consumption data and the verification result to a corresponding energy consumption data wide table of the machine room.

[0035] The steps of the machine room energy consumption management method will be described in detail below in combination with a specific implementation process.

[0036] Obtaining first device energy consumption data of a plurality of devices in a machine room in a first time period;

[0037] As an optional implementation, the types of devices in the machine room include air conditioning devices, air moving heat devices, information technology devices and power distribution devices, wherein the types of the air moving heat devices include water chillers, humidifiers, lighting auxiliary devices and switch devices, the types of the power distribution devices include uninterruptible power supplies and power distribution units, the device energy consumption data of the air conditioning devices at least includes consumed power, the device energy consumption data of the air moving heat devices at least includes temperature and humidity data and consumed power, the device energy consumption data of the information technology devices at least includes central processing unit temperature and process level server load, and the device energy consumption data of the power distribution devices at least includes consumed power. The data collection of each device is to understand its running state and energy use, thereby providing basic information for subsequent energy consumption prediction and optimization.

[0038] For the collection of first device energy consumption data, corresponding sensors can be deployed on these devices, such as power sensors to monitor power consumption, temperature and humidity sensors to monitor environmental conditions, central processor temperature sensors and server load monitoring tools to assess the running state of the devices. Within a set time period (first time period), data is automatically collected by the deployed sensors on a regular basis. For air conditioning devices, the main information collected is the power consumption. For air-moving heat devices, temperature and humidity data and power consumption can be collected. For information technology devices, the temperature of the central processor and the load of the process-level server can be collected. For power distribution equipment, power consumption can be collected.

[0039] After obtaining the first device energy consumption data, the first device energy consumption data is converted into a question set in a preset data format matching the input of the energy consumption prediction model. The process can take the following steps:

[0040] The first device energy consumption data is preprocessed, and the data preprocessing methods include at least one of the following: outlier deletion (such as statistical outliers), null value interpolation (such as using a specific algorithm or value to fill in missing data), and data standardization (such as scaling the data to the same range for model processing).

[0041] The first total energy consumption of each type of device is determined based on the first device energy consumption data, and the sum of all first total energy consumptions is determined as the second total energy consumption of the machine room. It should be noted that each type of device mentioned here is not necessarily limited to one large category, but can be further subdivided according to actual needs and analysis purposes. For example, air-moving heat devices can be considered as one large category, but the internal water chillers and humidifiers can be considered as more detailed subcategories.

[0042] The ratio of the first total energy consumption to the second total energy consumption of each type of device is determined as the first energy consumption ratio of that type of device, and the ratio of the second total energy consumption to the first total energy consumption corresponding to the information technology devices is determined as the first energy usage efficiency of the machine room (i.e., PUE (Power Usage Effectiveness)).

[0043] The question data for each type of device is generated based on the first total energy consumption and the first energy consumption ratio of that type of device, and the question set is generated based on all question data and the first energy usage efficiency. The question set is designed specifically for the input format of the energy consumption prediction model. It can be understood that question data is constructed for each type of device, which can include device type, time period, total energy consumption, and energy consumption ratio. The question data is aggregated into a question set, and the structure and format of the question set need to match the input of the energy consumption prediction model to ensure that the model can correctly analyze the question data.

[0044] After obtaining the question set, the energy consumption prediction model is used to analyze the question set to obtain second device energy consumption data in a second time period output by the energy consumption prediction model, the second time period being a time period after the first time period, wherein the energy consumption prediction model is trained based on historical device energy consumption data, and can predict the energy consumption of each type of device in a future time period, i.e., the total energy consumption, energy consumption proportion of each device, and the overall PUE of the machine room.

[0045] As an optional implementation, the training process of the energy consumption prediction model can include the following steps:

[0046] S1, constructing an initial prediction model;

[0047] The initial prediction model can be a model suitable for processing time series data, such as LSTM (Long Short-Term Memory) and GRU (Gate Recurrent Unit), which can effectively capture long-term dependencies in time series and is very suitable for predicting the energy consumption of machine room devices.

[0048] S2, obtaining a plurality of sets of historical device energy consumption data of a plurality of devices in a machine room in a plurality of consecutive historical time periods, wherein one historical time period corresponds to one set of historical device energy consumption data;

[0049] S3, for each historical time period, determining the third total energy consumption and the second energy consumption proportion of each type of device, and the second energy use efficiency of the machine room according to the historical device energy consumption data corresponding to the historical time period, combining the third total energy consumption, the second energy consumption proportion, and the second energy use efficiency to form a training sample, and taking the third total energy consumption, the second energy consumption proportion, and the second energy use efficiency corresponding to the next historical time period of the historical time period as the sample label of the training sample;

[0050] S4, iteratively training the initial prediction model according to the plurality of training samples and sample labels to obtain the energy consumption prediction model.

[0051] After obtaining the second device energy consumption data, the second device energy consumption data is verified using a preset energy consumption anomaly threshold condition, and in the case where the verification result indicates that there is an energy consumption anomaly, an abnormal alarm information is generated.

[0052] As an optional implementation, the result of the prediction, i.e., the second device energy consumption data, includes the fourth total energy consumption and the third energy consumption proportion of each type of device in the machine room in the future second time period, and the third energy use efficiency of the machine room in the future second time period.

[0053] Specifically, as an optional implementation, the second device energy consumption data is checked by using a preset energy consumption anomaly threshold condition, including: comparing the third energy use efficiency with a preset efficiency threshold, where the preset efficiency threshold can be determined based on the average value of the average energy consumption level historical data of each type of device under normal operating conditions, or can be determined in other ways, and in the case where the third energy use efficiency is greater than the preset efficiency threshold, it may indicate that the energy use efficiency of the machine room is lower than expected, and it is determined that the machine room has energy consumption anomaly. And according to the fourth total energy consumption of each type of device and the first total energy consumption, or the third energy consumption proportion of each type of device and the first total energy consumption, the energy consumption growth rate of the device is determined, and the device with energy consumption growth rate greater than the preset amplitude threshold is determined to have energy consumption anomaly, that is, if the predicted energy consumption of a certain type of device has a significant increase compared to its historical average energy consumption, and the growth rate exceeds the preset amplitude threshold, it also indicates that the device has an anomaly.

[0054] If an anomaly is detected, an anomaly alarm information will be generated, and after receiving the alarm, the maintenance personnel will inspect the corresponding device to check whether there is an actual fault or perform other inspections.

[0055] As an optional implementation, the inspection result can be further used to update the energy consumption prediction model, and the specific process is as follows: obtaining the inspection result for the energy consumption anomaly device; in the case where the inspection result indicates that the energy consumption anomaly device has an anomaly, the first device energy consumption data is used as a positive sample to update the energy consumption prediction model to improve the diagnostic accuracy of the model in similar situations; in the case where the inspection result indicates that the energy consumption anomaly device does not have an anomaly, the preset amplitude threshold is adjusted, and the first device energy consumption data is used as a negative sample to update the energy consumption prediction model, which helps to adjust the prediction threshold of the model to avoid future false positives.

[0056] The first device energy consumption data, the second device energy consumption data and the checking result obtained above are recorded to the energy consumption data wide table corresponding to the machine room.

[0057] The energy consumption data wide table is a table containing multiple fields, used to record the energy consumption information of each device in the machine room, wherein each device corresponds to at least the following fields: device identifier, device type, device installation location, time period identifier, first device energy consumption data corresponding to the time period, question set, second device energy consumption data, energy consumption anomaly threshold condition, checking result, anomaly inspection result. It should be noted that the data calculated for a type of device is shared by the device.

[0058] In addition to the above steps, further consider the quality assessment of data and optimization of data query, including: scoring the first device energy consumption data from multiple dimensions, obtaining a sub-score for each dimension, wherein the multiple dimensions at least include: data integrity dimension, if the data is complete without missing, the sub-score of this dimension may be full score, otherwise it will be deducted according to the degree of missing. Among them, the data specification dimension, such as whether the temperature data is Celsius (℃), whether the energy consumption data is kilowatt-hour (kWh), etc., the data that does not meet the specification will be deducted; the data accuracy dimension, such as whether the energy consumption data is reasonable, whether it matches the rated power of the device, whether the temperature reading is consistent with the external conditions, etc., inaccurate data will get a lower score; the data timeliness dimension, such as whether it is within the latest monitoring period, outdated data will be given a lower score to reflect its reduced reference value to the current situation; the data richness dimension, such as high load, low load, different environmental temperature, etc., the data coverage is extensive and the score is high, otherwise it is low; the data coverage dimension, check whether the data contains the energy consumption information of all monitored devices, ensure that there is no omission, and the data coverage is complete, the dimension score is high; determine the weight coefficient of each dimension, and sum the weighted scores of each dimension according to the weight coefficient to obtain the comprehensive score of the first device energy consumption data; record the comprehensive score to the energy consumption data wide table.

[0059] When the target object needs to query data, the target field matching the keyword in the data query instruction of the target object can also be retrieved from the energy consumption data wide table, and the target device corresponding to the target field is determined; the energy consumption information of the target device is fed back to the target object. For example, if the query instruction requires to view the energy consumption data of all air conditioning devices in a specific time period, the system will locate all records containing air conditioning and the specified time period, extract the corresponding specific device information (such as device ID, type, location, etc.) from the retrieved target field, determine the target device, and return the energy consumption information of the target device to the target object initiating the query, including the total energy consumption of the device in the query time period, abnormal state (if any), and possible PUE value and other key indicators.

[0060] In the embodiments of the present application, the energy consumption data of the device is acquired in real time, including temperature, humidity, power and other key indicators. The first device energy consumption data is converted into a question set format that can be understood by the energy consumption prediction model. This step ensures the standardization and structuring of the model input data, which is beneficial to improve the prediction accuracy of the model. The energy consumption prediction model trained based on historical device energy consumption data can predict the second device energy consumption data in the second time period. This prediction function provides forward-looking decision support for the management of the computer room, enabling managers to plan energy use in advance and optimize device operation strategies, thereby achieving the purpose of energy saving and emission reduction. The second device energy consumption data predicted is checked using a preset energy consumption anomaly threshold condition, which can timely identify energy consumption anomalies. Once an energy consumption anomaly is detected, the system will automatically generate an abnormal alarm information to notify relevant personnel for processing. The first device energy consumption data, the second device energy consumption data and the verification result of the anomaly detection are recorded in the energy consumption data wide table corresponding to the computer room. This not only facilitates long-term storage and historical comparison of data, but also provides a basis for subsequent in-depth data analysis. Through the wide table, multi-dimensional analysis of energy consumption data can be performed, including but not limited to device health status analysis, energy consumption trend prediction, fault mode identification, etc. This further optimizes the energy management and device maintenance of the computer room, and thus solves the technical problem that it is difficult to efficiently and accurately control the energy consumption data of a large number of computer room devices in the computer room operation and maintenance scenario.

[0061] Embodiment 2

[0062] According to the embodiments of the present application, a computer room energy consumption management device for implementing the computer room energy consumption management method in Embodiment 1 is also provided, as shown in Figure 2 The computer room energy consumption management device at least includes an acquisition module 21, a conversion module 22, a prediction module 23, an analysis module 24, and a statistics module 25, wherein:

[0063] The acquisition module 21 is configured to acquire first device energy consumption data of a plurality of devices in a computer room in a current first time period;

[0064] The conversion module 22 is configured to convert the first device energy consumption data into a question set in a preset data format matching the input of an energy consumption prediction model;

[0065] The prediction module 23 is configured to analyze the question set using the energy consumption prediction model to obtain second device energy consumption data in a second time period output by the energy consumption prediction model, wherein the energy consumption prediction model is trained based on historical device energy consumption data, and the second time period is a time period after the first time period;

[0066] The analysis module 24 is configured to check the second device energy consumption data using a preset energy consumption anomaly threshold condition, and generate an abnormal alarm information in the case that the checking result indicates that there is an energy consumption anomaly.

[0067] a statistical module 25, configured to record the first device energy consumption data, the second device energy consumption data and the check result to the corresponding energy consumption data wide table of the machine room.

[0068] The functions of the modules of the machine room energy consumption management device will be described below in combination with specific implementation processes.

[0069] The acquisition module acquires first device energy consumption data of a plurality of devices in the machine room in a first time period;

[0070] As an optional implementation, the types of the devices in the machine room include air conditioning devices, air-moving heat devices, information technology devices and power distribution devices, wherein the types of the air-moving heat devices include water chillers, humidifiers, lighting auxiliary devices and switching devices, and the types of the power distribution devices include uninterruptible power supplies and power distribution units; the device energy consumption data of the air conditioning devices at least includes power consumption; the device energy consumption data of the air-moving heat devices at least includes temperature and humidity data and power consumption; the device energy consumption data of the information technology devices at least includes central processing unit temperature and process-level server load; and the device energy consumption data of the power distribution devices at least includes power consumption.

[0071] After obtaining the first device energy consumption data, the conversion module converts the first device energy consumption data into a question set in a preset data format matching the input of the energy consumption prediction model, which can take the following steps:

[0072] The data preprocessing of the first device energy consumption data includes at least one of the following: outlier deletion, null value interpolation and data standardization processing; the first total energy consumption of each type of device is determined according to the first device energy consumption data, and the sum of all the first total energy consumptions is determined as the second total energy consumption of the machine room; the first energy consumption ratio of each type of device is determined as the ratio of the first total energy consumption of the device to the second total energy consumption, and the first energy consumption ratio of the information technology device is determined as the first energy consumption ratio of the machine room; the question data of each type of device is generated according to the first total energy consumption and the first energy consumption ratio of the device, and the question set is generated according to all the question data and the first energy consumption ratio.

[0073] After obtaining the question set, the prediction module analyzes the question set by using the energy consumption prediction model to obtain second device energy consumption data in a second time period output by the energy consumption prediction model, wherein the energy consumption prediction model is trained based on historical device energy consumption data, and the second time period is a time period after the first time period.

[0074] As an optional implementation, the training process of the energy consumption prediction model can include the following steps:

[0075] S1, constructing an initial prediction model;

[0076] S2, obtaining a plurality of sets of historical device energy consumption data of a plurality of devices in a machine room in a plurality of continuous historical time periods, wherein one historical time period corresponds to one set of historical device energy consumption data;

[0077] S3, for each historical time period, determining a third total energy consumption and a second energy consumption proportion of each type of device and a second energy use efficiency of the machine room according to the historical device energy consumption data corresponding to the historical time period, grouping the third total energy consumption, the second energy consumption proportion and the second energy use efficiency into a training sample, and taking the third total energy consumption, the second energy consumption proportion and the second energy use efficiency corresponding to a next historical time period of the historical time period as a sample label of the training sample;

[0078] S4, iteratively training an initial prediction model according to the plurality of training samples and the sample labels to obtain an energy consumption prediction model.

[0079] After obtaining the second device energy consumption data, the analysis module verifies the second device energy consumption data using a preset energy consumption anomaly threshold condition, and generates an abnormal alarm information in a case where the verification result indicates that there is an energy consumption anomaly.

[0080] As an optional implementation, the second device energy consumption data includes: fourth total energy consumption and third energy consumption proportions of each type of device in the machine room in a future second time period, and a third energy use efficiency of the machine room in the future second time period, and verifying the second device energy consumption data using the preset energy consumption anomaly threshold condition includes: comparing the third energy use efficiency with a preset efficiency threshold; in a case where the third energy use efficiency is greater than the preset efficiency threshold, determining that the machine room has an energy consumption anomaly, and determining an energy consumption growth rate of each type of device according to the fourth total energy consumption and the first total energy consumption of each type of device, or the third energy consumption proportion and the first total energy consumption of each type of device; determining that a device with an energy consumption growth rate greater than a preset amplitude threshold has an energy consumption anomaly.

[0081] As an optional implementation, after generating the abnormal alarm information, the following process can be performed: obtaining an inspection result for the energy consumption anomaly device; in a case where the inspection result indicates that the energy consumption anomaly device has an anomaly, using the first device energy consumption data as a positive sample to update the energy consumption prediction model; in a case where the inspection result indicates that the energy consumption anomaly device does not have an anomaly, adjusting the preset amplitude threshold, and using the first device energy consumption data as a negative sample to update the energy consumption prediction model.

[0082] The statistical module records the first device energy consumption data, the second device energy consumption data and the verification result obtained above to an energy consumption data wide table corresponding to the machine room.

[0083] The energy consumption data wide table is used to record the energy consumption information of each device in the machine room, wherein each device corresponds to at least the following fields: device identifier, device type, device installation location, time period identifier, first device energy consumption data corresponding to the time period, question set, second device energy consumption data, energy consumption anomaly threshold condition, verification result, and abnormal inspection result.

[0084] Optionally, the method further comprises: scoring the first device energy consumption data from multiple dimensions to obtain a sub-score of each dimension, wherein the multiple dimensions at least include: data integrity dimension, data specification dimension, data accuracy dimension, data timeliness dimension, data richness dimension, and data coverage dimension; determining a weight coefficient of each dimension, and performing weighted summation on the sub-score of each dimension according to the weight coefficient to obtain a comprehensive score of the first device energy consumption data; and recording the comprehensive score to the energy consumption data wide table.

[0085] In response to a data query instruction of a target object, the target field matched with the keyword in the data query instruction is retrieved from the energy consumption data wide table, and the target device corresponding to the target field is determined; and the energy consumption information of the target device is fed back to the target object.

[0086] It should be noted that each module in the machine room energy consumption management device in the embodiment of the present application corresponds to each implementation step of the machine room energy consumption management method in Embodiment 1. Since Embodiment 1 has been described in detail, the details not embodied in this embodiment can be referred to Embodiment 1, and will not be described in detail here.

[0087] Embodiment 3

[0088] According to the embodiments of the present application, a computer program product is also provided, which includes a computer program. When the computer program is executed by a processor, the machine room energy consumption management method in Embodiment 1 is implemented.

[0089] According to the embodiments of the present application, a non-volatile storage medium is also provided, which includes a stored computer program. The device in which the non-volatile storage medium is located executes the machine room energy consumption management method in Embodiment 1 by running the computer program.

[0090] According to the embodiments of the present application, a processor is also provided, which is used to run a computer program. When the computer program is run, the machine room energy consumption management method in Embodiment 1 is executed.

[0091] According to the embodiments of the present application, an electronic device is also provided, which includes a memory and a processor. The memory stores a computer program, and the processor is configured to execute the machine room energy consumption management method in Embodiment 1 by the computer program.

[0092] Specifically, the computer program runs to implement the following steps: obtaining first device energy consumption data of a plurality of devices in a machine room in a first time period; converting the first device energy consumption data into a question set in a preset data format matching an input of an energy consumption prediction model; analyzing the question set by using the energy consumption prediction model to obtain second device energy consumption data in a second time period output by the energy consumption prediction model, wherein the energy consumption prediction model is trained based on historical device energy consumption data, and the second time period is a time period after the first time period; checking the second device energy consumption data by using a preset energy consumption anomaly threshold condition, and generating an abnormal alarm information in a case where a checking result indicates that there is an energy consumption anomaly; and recording the first device energy consumption data, the second device energy consumption data and the checking result to a corresponding energy consumption data wide table of the machine room.

[0093] As an optional implementation, the electronic device can exist in the form of a mobile terminal, a computer terminal or a similar computing device. Figure 3 A hardware structure block diagram of an electronic device for implementing a machine room energy consumption management method is shown. As shown in the figure, Figure 3 The electronic device 30 can include one or more (shown in the figure as 302a, 302b, …, 302n) processors 302 (the processor 302 can include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 304 for storing data, and a transmission device 306 for communication functions. In addition, it can also include a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which can be included as one of the ports in the BUS bus), a network interface, a power supply and / or a camera. Those skilled in the art can understand, Figure 3 The structure shown in the figure is only a schematic, which does not limit the structure of the above-mentioned electronic device. For example, the electronic device 30 can include more or fewer components than Figure 3 shown in the figure, or have a different configuration than Figure 3 shown in the figure.

[0094] It should be noted that the one or more processors 302 and / or other data processing circuits described above can be referred to herein as "data processing circuits" in general. The data processing circuit can be embodied in whole or in part as software, hardware, firmware or any other combination. In addition, the data processing circuit can be a single independent processing module, or all or part of any one of the other elements combined into the electronic device 30. As referred to in the embodiments of the present application, the data processing circuit serves as a processor control (for example, selection of a variable resistance terminal path connected to an interface).

[0095] The memory 304 can be used to store software programs of application software and modules, such as program instructions / data storage device corresponding to the machine room energy consumption management method in the embodiments of the present application, and the processor 302 executes various functional applications and data processing by running the software programs and modules stored in the memory 304, that is, implements the vulnerability detection method of the application program described above. The memory 304 can include a high-speed random access memory, and can also include a non-volatile memory, such as one or more magnetic storage devices, flash memories, or other non-volatile solid-state memories. In some examples, the memory 304 can further include a memory remotely arranged with respect to the processor 302, which can be connected to the electronic device 30 through a network. Examples of the above-mentioned network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0096] The transmission device 306 is used to receive or send data via a network. The specific examples of the above-mentioned network can include a wireless network provided by the communication provider of the electronic device 30. In one example, the transmission device 306 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 306 can be a radio frequency (Radio Frequency, RF) module, which is used to communicate with the Internet in a wireless manner.

[0097] The display can be, for example, a touch screen type liquid crystal display (LCD), which can enable the user to interact with the user interface of the electronic device 30.

[0098] The above-mentioned embodiment numbers are only for description, and do not represent the advantages and disadvantages of the embodiments.

[0099] In the above-mentioned embodiments of the present application, the description of each embodiment has its own emphasis, and the parts not described in detail in a certain embodiment can be referred to the related description of other embodiments.

[0100] In the several embodiments of the present application, it should be understood that the disclosed technology can be implemented in other ways. Of course, the above-mentioned device embodiments are only schematic, for example, the division of units can be a logical function division, and in actual implementation, other division manners can be adopted, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interfaces, units or modules, and can be electrical or other forms.

[0101] The units described as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, i.e. may be located in one place, or may be distributed to multiple units. Part or all of the units may be selected according to actual needs to achieve the purpose of the embodiment.

[0102] In addition, the functional units in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0103] If the integrated unit is realized in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, including a number of instructions to make a computer device (which can be a personal computer, a server or a network device, etc.) execute all or part of the steps of the embodiments of the present application. The aforementioned storage medium includes: a U disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a mobile hard disk, a magnetic disk or an optical disk, and various program code storage media.

[0104] The above is only the preferred embodiment of the present application, and it should be pointed out that for those skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, which should be considered as the protection scope of the present application.

Claims

1. A method for managing energy consumption in a machine room, characterized by, The method comprises: obtaining first device energy consumption data of a plurality of devices in a machine room in a first time period; converting the first device energy consumption data into a question set in a preset data format matched with the input of an energy consumption prediction model; analyzing the question set by using the energy consumption prediction model to obtain second device energy consumption data in a second time period output by the energy consumption prediction model, wherein the energy consumption prediction model is trained based on historical device energy consumption data, and the second time period is a time period after the first time period; checking the second device energy consumption data by using a preset energy consumption anomaly threshold condition, and generating an anomaly alarm information in the case that the checking result indicates that there is an energy consumption anomaly; recording the first device energy consumption data, the second device energy consumption data and the checking result to a corresponding energy consumption data wide table of the machine room; wherein converting the first device energy consumption data into a question set in a preset data format matched with the input of an energy consumption prediction model comprises: data preprocessing of the first device energy consumption data, wherein the data preprocessing mode comprises at least one of the following: outlier deletion, null value interpolation, data standardization processing; determining a first total energy consumption of each type of device according to the first device energy consumption data, and determining the sum of all first total energy consumptions as a second total energy consumption of the machine room; determining the ratio of the first total energy consumption of each type of device to the second total energy consumption as the first energy consumption proportion of the device, and determining the ratio of the second total energy consumption to the first total energy consumption corresponding to the information technology device as the first energy use efficiency of the machine room; generating the question data of each type of device according to the first total energy consumption and the first energy consumption proportion of each type of device, and generating the question set according to all the question data and the first energy use efficiency.

2. The method of claim 1, wherein: the types of devices in the machine room include air conditioning devices, air-moving heat devices, information technology devices and power distribution devices, wherein the types of air-moving heat devices include water chillers, humidifiers, lighting auxiliary devices and switching devices, and the types of power distribution devices include uninterruptible power supplies and power distribution units; the device energy consumption data of the air conditioning devices at least includes power consumption; the device energy consumption data of the air-moving heat devices at least includes temperature and humidity data and power consumption; the device energy consumption data of the information technology devices at least includes central processing unit temperature and process-level server load; the device energy consumption data of the power distribution devices at least includes power consumption.

3. The method of claim 1, wherein, The training process of the energy consumption prediction model comprises: building an initial prediction model; obtaining a plurality of sets of historical device energy consumption data of a plurality of devices in the machine room in a plurality of continuous historical time periods, wherein one historical time period corresponds to one set of historical device energy consumption data; For each historical time period, a third total energy consumption and a second energy consumption proportion of each type of device and a second energy use efficiency of the machine room are determined according to historical device energy consumption data corresponding to the historical time period, the third total energy consumption, the second energy consumption proportion and the second energy use efficiency form a training sample, and the third total energy consumption, the second energy consumption proportion and the second energy use efficiency corresponding to a next historical time period of the historical time period are taken as sample labels of the training sample; The initial prediction model is iteratively trained according to a plurality of training samples and sample labels to obtain the energy consumption prediction model.

4. The method of claim 1, wherein, The second device energy consumption data includes fourth total energy consumption and third energy consumption proportions of each type of device in the machine room in a future second time period, and a third energy use efficiency of the machine room in the future second time period. The second device energy consumption data is verified using a preset energy consumption anomaly threshold condition, including: Comparing the third energy use efficiency with a preset efficiency threshold; In the case where the third energy use efficiency is greater than the preset efficiency threshold, it is determined that the machine room has energy consumption anomalies, and the energy consumption growth rate of each type of device is determined according to the fourth total energy consumption and the first total energy consumption of each type of device, or the third energy consumption proportion and the first total energy consumption of each type of device; It is determined that the device with an energy consumption growth rate greater than a preset amplitude threshold has energy consumption anomalies.

5. The method of claim 4, wherein, After generating the anomaly alarm information, the method further includes: Obtaining an inspection result for the energy consumption anomaly device; In the case where the inspection result indicates that the energy consumption anomaly device has anomalies, the first device energy consumption data is used as a positive sample to update the energy consumption prediction model; In the case where the inspection result indicates that the energy consumption anomaly device does not have anomalies, the preset amplitude threshold is adjusted, and the first device energy consumption data is used as a negative sample to update the energy consumption prediction model.

6. The method of claim 1, wherein The energy consumption data wide table is used to record energy consumption information of each device in the machine room, wherein each device corresponds to at least the following fields: device identifier, device type, device installation location, time period identifier, first device energy consumption data corresponding to the time period, question set, second device energy consumption data, energy consumption anomaly threshold condition, verification result, and anomaly inspection result.

7. The method of claim 6, wherein, The method further includes: In response to a data query instruction of a target object, retrieving a target field matching a keyword in the data query instruction from the energy consumption data wide table, and determining a target device corresponding to the target field; Feeding back energy consumption information of the target device to the target object.

8. The method of claim 6, wherein, The method further includes: Scoring the first device energy consumption data from multiple dimensions to obtain a sub-score of each dimension, wherein the multiple dimensions include at least data integrity dimension, data specification dimension, data accuracy dimension, data timeliness dimension, data richness dimension, and data coverage dimension. determining a weight coefficient of each dimension, and performing weighted summation on the sub-scores of the respective dimensions according to the weight coefficient to obtain a comprehensive score of the first device energy consumption data; recording the comprehensive score to the energy consumption data wide table.

9. A machine room energy consumption management device, characterized by, The method comprises the steps of: obtaining first device energy consumption data of a plurality of devices in a computer room in a current first time period; converting the first device energy consumption data into a question set in a preset data format matched with an input of an energy consumption prediction model, comprising: performing data preprocessing on the first device energy consumption data, wherein the data preprocessing mode comprises at least one of the following: outlier deletion, null value interpolation, and data standardization processing; determining a first total energy consumption of each type of device according to the first device energy consumption data, and determining a sum of all first total energy consumptions as a second total energy consumption of the computer room; determining a first energy consumption proportion of each type of device as a ratio of the first total energy consumption of the device to the second total energy consumption, and determining a first energy use efficiency of the computer room as a ratio of the second total energy consumption to the first total energy consumption corresponding to information technology devices; generating question data of each type of device according to the first total energy consumption and the first energy consumption proportion of the device, and generating the question set according to all the question data and the first energy use efficiency; predicting, by using the energy consumption prediction model, the question set to obtain second device energy consumption data in a second time period output by the energy consumption prediction model, wherein the energy consumption prediction model is trained based on historical device energy consumption data, and the second time period is a time period after the first time period; checking, by using a preset energy consumption anomaly threshold condition, the second device energy consumption data, and generating an abnormal alarm information in a case where a checking result indicates that there is an energy consumption anomaly; recording the first device energy consumption data, the second device energy consumption data, and the checking result to an energy consumption data wide table corresponding to the computer room.

10. A computer program product, characterised in that, The computer program is executed by a processor to implement the computer room energy consumption management method in any one of claims 1 to 8. The memory stores a computer program, and the processor is configured to execute the computer room energy consumption management method in any one of claims 1 to 8 by using the computer program.

11. An electronic device, comprising: The memory stores a computer program, and the processor is configured to execute the computer room energy consumption management method in any one of claims 1 to 8 by using the computer program. ​

Images