A method for predicting device power consumption and related components

Abstract

The present invention discloses a method for predicting device power consumption and related components thereof, and relates to the field of power consumption management. After obtaining the current power consumption of a device at the current time, the historical actual power consumption and the historical predicted power consumption corresponding to N historical time points before the current time are determined, and a power consumption error is determined based on the historical actual power consumption and the historical predicted power consumption. Then, the predicted power consumption at a preset time point after the current time is determined based on the current power consumption, the power consumption error and the N historical actual power consumptions. By determining the predicted power consumptions of the device in the future, the future power consumption trend of the device can be predicted, which is convenient for preventing possible power consumption problems of the device in advance, avoiding abnormal power outages or component damage caused by the power consumption of the device exceeding the limit, and avoiding economic losses caused by affecting the key business of the device.

Description

Technical Field

[0001] The present invention relates to the field of power consumption management, and in particular to a method for predicting device power consumption and related components thereof. Background Art

[0002] A data center is equipped with multiple computer rooms, each of which is equipped with multiple cabinets, and each cabinet is equipped with multiple electronic devices such as servers, switches, and routers. According to the inclusion relationship, the data center, computer room, cabinet, and electronic equipment can be divided into different levels of equipment. Each upper-level device contains multiple lower-level devices, and the electronic equipment is the lowest-level device. These electronic devices need to consume power resources when working. When the user needs to manage a certain level of equipment, by detecting the total power consumption of the device at the current time, the power consumption of the device can be managed according to the detected power consumption. However, the method of only detecting the power consumption at the current time cannot predict the power consumption of the device at a future time point, which is not conducive to predicting the future power consumption trend of the device, and cannot prevent the power consumption problems that may occur in the device in advance, which is easy to cause power supply interruption to the device. Summary of the invention

[0003] The purpose of the present invention is to provide a method for predicting device power consumption and its related components, which can predict the power consumption of the device at various time points in the future, and can also predict the future power consumption trend of the device, so as to prevent power consumption problems that may occur in the device in advance, avoid abnormal power outages or component damage caused by the power consumption of the device exceeding the limit, and avoid economic losses caused by affecting the key business of the device.

[0004] In order to solve the above technical problems, the present invention provides a method for predicting device power consumption, comprising:

[0005] Get the current power consumption of the device at the current time;

[0006] Determine the historical actual power consumption and the historical predicted power consumption of the device corresponding to N historical time points before the current time, where N is a positive integer;

[0007] Determine a power consumption error according to each of the historical predicted power consumptions and each of the historical actual power consumptions;

[0008] The predicted power consumption of the device at a preset time point after the current time is determined according to the current power consumption, the power consumption error and the N historical actual power consumptions.

[0009] Preferably, determining the power consumption error according to each of the historical predicted power consumptions and each of the historical actual power consumptions includes:

[0010] Determine an absolute value of a difference between each of the historical actual power consumptions and the historical predicted power consumption corresponding to the historical actual power consumption;

[0011] A ratio between the sum of all the absolute values ​​and the number of the historical time points is determined as the power consumption error.

[0012] Preferably, when N is greater than 2, after determining the power consumption error according to each of the historical predicted power consumptions and each of the historical actual power consumptions, the method further includes:

[0013] Determine the difference between two adjacent historical actual power consumptions according to the time sequence of the historical time points where the historical actual power consumptions are located;

[0014] Determine the trend of the historical actual power consumption according to each of the differences;

[0015] When the trend is a monotonically increasing trend, determining that the power consumption error is a positive number;

[0016] When the trend is a monotonically decreasing trend, it is determined that the power consumption error is a negative number.

[0017] Preferably, when the trend is not a monotonically increasing trend or a monotonically decreasing trend, it further includes:

[0018] Determining, among the differences, a first number of the differences having positive values ​​and a second number of the differences having negative values;

[0019] When the first number is greater than the second number, determining that the power consumption error is a positive number;

[0020] When the first number is less than the second number, determining that the power consumption error is a negative number;

[0021] When the first number is equal to the second number, the power consumption error is determined to be a preset power consumption error.

[0022] Preferably, when the device is not the lowest level device, obtaining the current power consumption of the device at the current time includes:

[0023] Obtaining the power consumption of each of the lowest level devices in the device at the current time;

[0024] The sum of the power consumption of all the lower-level devices at the current time is determined as the current power consumption of the device.

[0025] Preferably, when the current power consumption acquired at the current time is the first power consumption acquired, determining the historical actual power consumption and the historical predicted power consumption of the device corresponding to N historical time points before the current time includes:

[0026] The current power consumption is determined as the historical actual power consumption and the historical predicted power consumption respectively.

[0027] Preferably, after determining the predicted power consumption of the device at a preset time point after the current time according to the current power consumption, the power consumption error and the N historical actual power consumptions, the method further includes:

[0028] Determining whether the predicted power consumption is greater than a preset power consumption value;

[0029] If it is greater than the preset power consumption value, an alarm message is generated.

[0030] Preferably, before determining the predicted power consumption of the device at a preset time point after the current time according to the current power consumption, the power consumption error and the N historical actual power consumptions, the method further includes:

[0031] Determining whether the power consumption error is less than a preset error value;

[0032] If it is not less than the preset error value, then entering the step of determining the predicted power consumption of the device at a preset time point after the current time according to the current power consumption, the power consumption error and X historical actual power consumptions, where the X historical actual power consumptions are all the historical actual power consumptions at N historical time points, and X is a positive integer and less than N;

[0033] If it is less than the preset error value, obtain Y new historical actual power consumptions at the historical time points, and enter the step of determining the predicted power consumption of the device at a preset time point after the current time based on the current power consumption, the power consumption error and Y+N historical actual power consumptions, where Y is a positive integer.

[0034] The present application also provides a device for predicting power consumption of a device, comprising:

[0035] Memory for storing computer programs;

[0036] The processor is used to implement the steps of the method for predicting device power consumption as described above when executing the computer program.

[0037] The present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method for predicting device power consumption as described above are implemented.

[0038] The present application provides a method for predicting device power consumption and related components thereof, which relate to the field of power consumption management. After obtaining the current power consumption of a device at the current time, the historical actual power consumption and the historical predicted power consumption corresponding to N historical time points before the current time are determined, and a power consumption error is determined based on these historical actual power consumption and historical predicted power consumption. Then, the predicted power consumption at a preset time point after the current time is determined based on the current power consumption, the power consumption error and the N historical actual power consumptions. By determining the predicted power consumptions of the device in the future, the future power consumption trend of the device can be predicted, which is convenient for preventing possible power consumption problems of the device in advance, avoiding abnormal power outages or component damage caused by the power consumption of the device exceeding the limit, and avoiding economic losses caused by affecting the key business of the device. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the prior art and the drawings required for use in the embodiments are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0040] Figure 1 A flowchart of a method for predicting device power consumption provided in this application;

[0041] Figure 2 A schematic diagram of the structure of a device power consumption prediction apparatus provided in the present application. DETAILED DESCRIPTION

[0042] The core of the present invention is to provide a method for predicting device power consumption and its related components, which can predict the power consumption of the device at various time points in the future, and can also predict the future power consumption trend of the device, so as to prevent possible power consumption problems of the device in advance, avoid abnormal power outages or component damage caused by the power consumption of the device exceeding the limit, and avoid economic losses caused by affecting the key business of the device.

[0043] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] Please refer to Figure 1 , Figure 1 A flowchart of a method for predicting device power consumption provided in this application includes:

[0045] S1: Get the current power consumption of the device at the current time;

[0046] S2: Determine the historical actual power consumption and historical predicted power consumption of the device corresponding to N historical time points before the current time, where N is a positive integer;

[0047] S3: determining a power consumption error according to each historical predicted power consumption and each historical actual power consumption;

[0048] S4: Determine the predicted power consumption of the device at a preset time point after the current time according to the current power consumption, the power consumption error and N historical actual power consumptions.

[0049] At present, the existing technology only obtains the power consumption of the device at the current time when managing and controlling power consumption, and then displays the current power consumption of the device on the display screen so that users can view and manage and control it according to the current power consumption. However, the method of only obtaining the current power consumption can only obtain the current power consumption data, and cannot predict the future power consumption, which is not conducive to predicting the power consumption trend and preventing possible problems in the future, and is prone to cause problems such as abnormal power supply of the device and further business interruption.

[0050] In order to predict the future power consumption of the device, in this application, after obtaining the current power consumption of the device at the current time, first determine the historical actual power consumption and historical predicted power consumption corresponding to the N historical time points before the current time. The historical time point is the time before the current time when the step of obtaining the power consumption of the device is executed. The N historical time points can be N consecutive time points with the time before the current time when the power consumption of the device is obtained as the end point. The historical actual power consumption is the power consumption actually measured at the historical time point. The historical predicted power consumption is the power consumption obtained when predicting the historical time point at a certain time point before the historical time point. Considering that the power consumption may have short-term sudden abnormalities and cause power consumption abnormalities, in order to ensure that the predicted power consumption is accurate and in line with the actual working state, the number of historical time points determined needs to be sufficient, that is, the N value needs to be large enough to ensure that the trend of power consumption is reasonable in the long term and in line with the actual working state, and avoid being affected by short-term sudden abnormalities. When determining the predicted power consumption, there are:

[0051] P=A+MAD

[0052] Among them, P is the predicted power consumption, A is the average value between each historical actual power consumption and the current power consumption, MAD is the power consumption error, and the MAD value may be positive or negative, which needs to be judged according to the trend of historical actual power consumption. For example, when each historical actual power consumption shows a monotonically decreasing trend, MAD is negative, and if it shows a monotonically increasing trend, MAD is positive. After obtaining the predicted power consumption, that is, the P value, the current power consumption and the predicted power consumption can be stored, and the storage method includes but is not limited to storing these power consumptions in a database, memory, cache, or writing files.

[0053] In addition, when obtaining the current power consumption of the device, the power consumption of the device can be obtained periodically. For example, the power consumption of the device can be obtained once every minute through a sensor or a power consumption interface. The data can be collected according to the protocol corresponding to the specific model of the device. Different devices need to be collected through different protocols, such as SNMP (Simple Network Management Protocol) or Redfish protocol. It can also be collected according to relevant information and methods provided by the manufacturer of the device. This application does not limit how to obtain the current power consumption of the device.

[0054] In summary, after obtaining the current power consumption of the device at the current time, determine the historical actual power consumption and the historical predicted power consumption corresponding to the N historical time points before the current time, determine the power consumption error based on these historical actual power consumption and historical predicted power consumption, and then determine the predicted power consumption at a preset time point after the current time based on the current power consumption, the power consumption error and the N historical actual power consumptions. By determining the various predicted power consumptions of the device in the future, the future power consumption trend of the device can be predicted, which is convenient for preventing possible power consumption problems of the device in advance, avoiding abnormal power outages or component damage caused by the power consumption of the device exceeding the limit, and avoiding economic losses caused by affecting the key business of the device.

[0055] Based on the above embodiments:

[0056] As a preferred embodiment, determining the power consumption error according to each historical predicted power consumption and each historical actual power consumption includes:

[0057] Determine the absolute value of the difference between each historical actual power consumption and the historical predicted power consumption corresponding to the historical actual power consumption;

[0058] The ratio between the sum of all absolute values ​​and the number of historical time points is determined as the power consumption error.

[0059] In order to determine the power consumption error, in this application, the power consumption error is assumed to be MAD, i is a historical time point, and the historical actual power consumption measured at the i time point is X i, the historical predicted power consumption measured at time point i is Y i , the number of historical time points is N, then the power consumption error is:

[0060]

[0061] Based on this, a power consumption error can be determined, so that the predicted power consumption of the device at a preset time point after the current time can be determined later according to the power consumption error, the current power consumption and N historical actual power consumptions.

[0062] As a preferred embodiment, when N is greater than 2, after determining the power consumption error according to each historical predicted power consumption and each historical actual power consumption, the method further includes:

[0063] Determine the difference between two adjacent historical actual power consumptions according to the time sequence of the historical time points where the historical actual power consumptions are located;

[0064] Determine the trend of historical actual power consumption based on each difference;

[0065] When the trend is a monotonically increasing trend, the power consumption error is determined to be a positive number;

[0066] When the trend is a monotonically decreasing trend, it is determined that the power consumption error is a negative number.

[0067] In order to determine the specific direction of the power consumption error, in this application, when determining the difference between two historical actual power consumptions, it is necessary to determine it according to the time sequence. If it is not determined according to the time sequence, the trend of the historical actual power consumption of the device cannot be determined. For example, there are a total of 3 historical actual power consumptions obtained on the same day, namely, 100W power consumption obtained at 8 o'clock, 150W power consumption obtained at 9 o'clock, and 200W power consumption obtained at 10 o'clock. According to the time sequence, it can be known that the difference between the power consumption at 9 o'clock and the power consumption at 8 o'clock is 50W, and the power consumption at 10 o'clock is 200W. The difference between the power consumption at 8 o'clock and the power consumption at 8 o'clock is 50W, and its trend is 50W—>50W, which shows that it is a monotonically increasing trend; but if it is not in the time sequence, the difference between the power consumption at 10 o'clock and the power consumption at 8 o'clock is taken as the first difference, and the power consumption at 8 o'clock and the power consumption at 9 o'clock is taken as the second difference, the trend is 100W—>-50W, which shows that it has become a monotonically decreasing trend, which does not conform to the actual power consumption of the device. Therefore, it is necessary to determine the difference between two adjacent historical actual power consumptions according to the time sequence of the historical time points corresponding to the historical actual power consumption. In determining the direction of the power consumption error, when the trend of the historical actual power consumption is a monotonically increasing trend, it can be known that the power consumption of the device has been increasing, and it can be foreseen that the device may continue to increase its power consumption in the future, so the power consumption error can be determined as a positive number to increase the value of the predicted power consumption. Similarly, when the trend is a monotonically decreasing trend, the power consumption error is a negative number. Based on this, the direction of the power consumption error can be determined, and the predicted power consumption can be determined more accurately.

[0068] As a preferred embodiment, when the trend is not a monotonically increasing trend or a monotonically decreasing trend, the method further includes:

[0069] Among the differences, determining a first number of differences having positive values ​​and a second number of differences having negative values;

[0070] When the first number is greater than the second number, the power consumption error is determined to be a positive number;

[0071] When the first number is less than the second number, determining that the power consumption error is a negative number;

[0072] When the first number is equal to the second number, it is determined that the power consumption error is a preset power consumption error.

[0073] Considering that the changes in historical actual power consumption are relatively complex, in this application, in order to accurately determine the actual trend of historical actual power consumption, after determining that the trend of historical actual power consumption is not a monotonically increasing trend or a monotonically decreasing trend, the trend of historical actual power consumption can be determined according to each difference. When the first number is greater than the second number, it means that although the historical actual power consumption occasionally decreases, since most of them are still positive numbers, it means that the historical actual power consumption is still increasing overall. It can be predicted that the future power consumption of the device will also be higher than the current time, so the power consumption error is determined to be a positive number; similarly, when the first number is less than the second number, it means that the historical actual power consumption is decreasing overall, so the power consumption error can be determined to be a negative number; when the first number is equal to the second number, it means that the historical actual power consumption remains relatively stable during this period and does not show an increasing or decreasing trend. At this time, the power consumption error can be determined to be a preset power consumption error, for example, it can be 0. Based on this, the actual trend of historical actual power consumption can be accurately determined, which is conducive to determining the direction of the power consumption error.

[0074] As a preferred embodiment, when the device is not the lowest level device, obtaining the current power consumption of the device at the current time includes:

[0075] Get the power consumption of each lowest level device in the device at the current time;

[0076] The sum of the power consumption of all the lowest level devices at the current time is determined as the current power consumption of the device.

[0077] In order to obtain the current power consumption of the device at the current time, in this application, there is an inclusion relationship between the devices at each level. The data center is the upper-level device of the computer room, cabinet and electronic equipment, the computer room is the upper-level device of the cabinet and electronic equipment, the cabinet is the upper-level device of the electronic equipment, and the electronic equipment is the lowest-level device. Considering that the upper-level device is not a device that can consume power resources by itself, the device that mainly consumes power resources is the lowest-level device, that is, electronic devices such as servers, switches and routers are the main devices that consume power resources. If the device for which power consumption needs to be predicted is not the lowest-level device, the power consumption of all the lowest-level devices in the device at the current time can be obtained first.

[0078] For example, when the device is D, the device D contains j lowest-level devices, and the power consumption of the jth lowest-level device at the current time is M j , then:

[0079]

[0080] Based on this, the sum of the power consumption of all the lowest level devices in the device at the current time may be used as the current power consumption of the device itself, so as to subsequently determine the predicted power consumption of the device according to the current power consumption.

[0081] As a preferred embodiment, when the current power consumption acquired at the current time is the first power consumption acquired, determining the historical actual power consumption and the historical predicted power consumption of the device corresponding to N historical time points before the current time includes:

[0082] The current power consumption is determined as the historical actual power consumption and the historical predicted power consumption respectively.

[0083] Because historical actual power consumption and historical predicted power consumption need to be used for prediction when predicting power consumption at the current time, and when the current time is the first time to obtain power consumption, there is no historical actual power consumption and historical predicted power consumption before the current time, and it lacks relevant parameters for determining power consumption errors. In order to enable the first time to obtain power consumption to correctly execute the steps of predicting power consumption, the current power consumption value obtained at this time point can be used as the historical actual power consumption and historical predicted power consumption, which can ensure the correctness of the historical actual power consumption and historical predicted power consumption.

[0084] As a preferred embodiment, after determining the predicted power consumption of the device at a preset time point after the current time according to the current power consumption, the power consumption error and N historical actual power consumptions, the method further includes:

[0085] Determine whether the predicted power consumption is greater than the preset power consumption value;

[0086] If it is greater than the preset power consumption value, an alarm message is generated.

[0087] In order to allow users to intuitively determine the power consumption of the device, in this application, after discovering that the future predicted power consumption of the device is greater than the preset power consumption value, if the predicted power consumption exceeds the upper limit of the power consumption of the device, an alarm message will be generated to prompt the user that there is a problem with the power consumption of the device. Specifically, a power consumption chart can be generated based on the current power consumption of the device, the sum of the current power consumption and each historical actual power consumption, and the predicted power consumption. The horizontal axis of the chart is the time point and the vertical axis is the power consumption value. Through the chart, the user can intuitively understand the actual power consumption trend of the device and the predicted future power consumption trend.

[0088] As a preferred embodiment, before determining the predicted power consumption of the device at a preset time point after the current time according to the current power consumption, the power consumption error and N historical actual power consumptions, the method further includes:

[0089] Determine whether the power consumption error is less than a preset error value;

[0090] If it is not less than the preset error value, then enter the step of determining the predicted power consumption of the device at a preset time point after the current time according to the current power consumption, the power consumption error and X historical actual power consumptions, where the X historical actual power consumptions are all historical actual power consumptions at N historical time points, and X is a positive integer and less than N;

[0091] If it is less than the preset error value, obtain the historical actual power consumption of Y new historical time points, and enter the step of determining the predicted power consumption of the device at a preset time point after the current time based on the current power consumption, power consumption error and Y+N historical actual power consumptions, where Y is a positive integer.

[0092] In order to accurately determine the power consumption error of the device, in the present application, before determining the predicted power consumption of the device, it is first determined whether the power consumption error is less than the preset error value. If it is not less than, it means that the power consumption error is large at this time, and the large reason may be that it is affected by the power consumption anomaly caused by the short-term sudden abnormality of the device. In order to avoid the influence of this anomaly, when determining the predicted power consumption, the number of historical actual power consumptions participating in the determination process will be reduced; and when it is less than, it means that the power consumption error is small, but the small reason may be that the number of historical actual power consumptions is small and cannot reflect the actual working conditions of the device, so when determining the predicted power consumption, the number of historical actual power consumptions participating in the determination process will be increased. The number of historical actual power consumption of the determination process; for example, if N is 5, and the number of historical actual power consumption ranges from 1 to 10, the power consumption error is MAD, and the preset error value is 1. When MAD is greater than or equal to 1, the value after 5 minus 1 is used as the X value, and the X value is used as the new N value when the power consumption is predicted next time; and when MAD is less than 1, the value after 5 plus 1 is used as the Y value, and the Y value is used as the new N value when the power consumption is predicted next time, that is, the N value is increased or decreased by 1 at most per cycle to avoid excessive changes in the N value, which leads to a decrease in the accuracy of the power consumption error. In addition, when MAD is equal to 0, the N value may not be changed. Based on this, the power consumption error can be accurately determined.

[0093] Please refer to Figure 2 , Figure 2 A schematic diagram of a device power consumption prediction apparatus provided in this application includes:

[0094] A memory 21, used for storing computer programs;

[0095] The processor 22 is used to implement the steps of the method for predicting device power consumption as described above when executing a computer program.

[0096] For a detailed introduction to a device for predicting power consumption of a device provided in the present application, please refer to the embodiment of the method for predicting power consumption of a device, and the present application will not go into details here.

[0097] The present application also provides a computer-readable storage medium, characterized in that a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the steps of the method for predicting device power consumption as described above are implemented.

[0098] For a detailed introduction to a computer-readable storage medium provided in the present application, please refer to the embodiment of the method for predicting device power consumption, and the present application will not go into details here.

[0099] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0100] It should also be noted that, in this specification, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the statement "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device including the element.

Claims

1. A method for predicting device power consumption, characterized in that: include: Get the current power consumption of the device at the current time; Determine the historical actual power consumption and the historical predicted power consumption of the device corresponding to N historical time points before the current time, where N is a positive integer; Determine a power consumption error according to each of the historical predicted power consumptions and each of the historical actual power consumptions; Determine the predicted power consumption of the device at a preset time point after the current time according to the current power consumption, the power consumption error and the N historical actual power consumptions; Wherein, when N is greater than 2, after determining the power consumption error according to each of the historical predicted power consumptions and each of the historical actual power consumptions, the method further includes: Determine the difference between two adjacent historical actual power consumptions according to the time sequence of the historical time points where the historical actual power consumptions are located; Determine the trend of the historical actual power consumption according to each of the differences; When the trend is a monotonically increasing trend, determining that the power consumption error is a positive number; When the trend is a monotonically decreasing trend, determining that the power consumption error is a negative number; When the trend is not a monotonically increasing trend or a monotonically decreasing trend, determining, among the differences, a first number of the differences having positive values ​​and a second number of the differences having negative values; When the first number is greater than the second number, determining that the power consumption error is a positive number; When the first number is less than the second number, determining that the power consumption error is a negative number; When the first number is equal to the second number, determining that the power consumption error is a preset power consumption error; Before determining the predicted power consumption of the device at a preset time point after the current time according to the current power consumption, the power consumption error and the N historical actual power consumptions, the method further includes: Determining whether the power consumption error is less than a preset error value; If it is not less than the preset error value, then entering the step of determining the predicted power consumption of the device at a preset time point after the current time according to the current power consumption, the power consumption error and X historical actual power consumptions, where the X historical actual power consumptions are all the historical actual power consumptions at N historical time points, and X is a positive integer and less than N; If it is less than the preset error value, obtain Y new historical actual power consumptions at the historical time points, and enter the step of determining the predicted power consumption of the device at a preset time point after the current time based on the current power consumption, the power consumption error and Y+N historical actual power consumptions, where Y is a positive integer.

2. The method for predicting device power consumption according to claim 1, characterized in that: Determining a power consumption error according to each of the historical predicted power consumptions and each of the historical actual power consumptions includes: Determine an absolute value of a difference between each of the historical actual power consumptions and the historical predicted power consumption corresponding to the historical actual power consumption; A ratio between the sum of all the absolute values ​​and the number of the historical time points is determined as the power consumption error.

3. The method for predicting device power consumption according to claim 1, characterized in that: When the device is not the lowest level device, obtaining the current power consumption of the device at the current time includes: Obtaining the power consumption of each of the lowest level devices in the device at the current time; The sum of the power consumptions of all the lowest level devices at the current time is determined as the current power consumption of the device.

4. The method for predicting device power consumption according to claim 1, characterized in that: When the current power consumption acquired at the current time is the first power consumption acquired, determining the historical actual power consumption and the historical predicted power consumption of the device corresponding to N historical time points before the current time, including: The current power consumption is determined as the historical actual power consumption and the historical predicted power consumption respectively.

5. The method for predicting device power consumption according to claim 1, characterized in that: After determining the predicted power consumption of the device at a preset time point after the current time according to the current power consumption, the power consumption error and the N historical actual power consumptions, the method further includes: Determining whether the predicted power consumption is greater than a preset power consumption value; If it is greater than the preset power consumption value, an alarm message is generated.

6. A device for predicting equipment power consumption, characterized in that: include: Memory for storing computer programs; A processor, configured to implement the steps of the method for predicting device power consumption as described in any one of claims 1 to 5 when executing the computer program.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method for predicting device power consumption according to any one of claims 1 to 5 are implemented.

Images