Energy-saving control method, electronic equipment and computer readable storage medium

Abstract

The embodiment of the invention provides an energy-saving control method, electronic equipment and a computer readable storage medium, and belongs to the technical field of energy saving. The energy-saving control method comprises the following steps: acquiring flow time sequence data carried by controlled equipment, wherein the flow time sequence data is flow data carrying time sequence characteristics; extracting characteristics of the traffic time sequence data to obtain traffic characteristics; classifying the traffic time sequence data based on the traffic features to obtain traffic scene categories; determining an energy-saving strategy based on the traffic scene category; generating energy-saving control information based on the energy-saving strategy; wherein the control information is used for performing energy-saving control on the controlled equipment. According to the energy-saving control method, the flexibility of the energy-saving control information can be improved, the energy-saving income can be improved, and the business risk can be reduced.

Description

Technical Field

[0001] The present disclosure relates to the field of energy-saving technologies, and in particular, to an energy-saving control method, an electronic device, and a computer-readable storage medium. Background Art

[0002] With the rapid development of the Internet, the scale of networks and data centers has been continuously increasing, and the energy consumption problem has become increasingly prominent. In order to reduce power consumption, it is particularly important to control the power consumption of network devices. When the traffic carried by the device is low, some components can be put into sleep mode, and when the traffic increases, the sleeping components can be woken up, which can achieve the effect of energy conservation and consumption reduction. However, after the components enter the sleep mode, the carrying capacity of the device will decrease, accompanied by an increase in service risks. Considering service security and energy-saving effects, multiple energy-saving strategies are usually set, and the energy-saving strategy is determined according to the traffic carried by the device.

[0003] The energy-saving strategies of related technologies can be determined by manually observing the actual traffic carried by the device. This method has high requirements for operation and maintenance personnel, and there are human misoperation situations, and it also requires a large amount of manpower. The energy-saving strategy can also be determined by a traffic threshold, that is, the energy-saving strategy is determined according to the size of the traffic carried by the device and the threshold. This method requires a preset threshold, and the threshold has a great influence on determining the energy-saving strategy, and it has poor flexibility and cannot adapt to the actual traffic scenario. The energy-saving strategy can also predict the future traffic change situation through historical traffic data, but there are inaccurate prediction situations, resulting in the mismatch between the energy-saving strategy and the traffic carried by the device, failing to achieve the energy-saving effect, or generating service risks such as packet loss. Summary of the Invention

[0004] Embodiments of the present application provide an energy-saving control method, an electronic device, and a computer-readable storage medium, which can flexibly and accurately select an energy-saving strategy, improve energy-saving benefits, and reduce service risks at the same time.

[0005] In a first aspect, embodiments of the present application provide an energy-saving control method, where the method includes:

[0006] Obtain traffic time-series data carried by a controlled device, where the traffic time-series data is traffic data carrying time-series features;

[0007] Extract features of the traffic time-series data to obtain traffic features;

[0008] Classify the traffic time-series data based on the traffic features to obtain a traffic scenario category;

[0009] Determine an energy-saving strategy based on the traffic scenario category;

[0010] Generate energy-saving control information based on the energy-saving strategy; where the control information is used to perform energy-saving control on the controlled device.

[0011] In a second aspect, an embodiment of the present application provides an electronic device, including:

[0012] One or more processors;

[0013] A memory storing one or more programs, which when executed by the one or more processors, cause the one or more processors to implement the energy-saving control method provided by the embodiments of the present application.

[0014] In a third aspect, an embodiment of the present application provides a computer-readable medium storing a computer program, which when executed by a processor, implements the energy-saving control method provided by the embodiments of the present application.

[0015] The energy-saving control method provided by the embodiments of the present application extracts the characteristics of the traffic time-series data carried by the controlled device to obtain traffic characteristics. Compared with the traffic data without time series, the traffic characteristics can better reflect the time-series characteristics of the traffic of the controlled device. Classify the traffic time-series data based on the traffic characteristics of the traffic time-series data, and determine the energy-saving strategy according to the traffic scenario category. Compared with predicting traffic changes based on historical traffic data, it can more accurately reflect the traffic characteristics at the current moment. Therefore, the energy-saving control information determined based on this energy-saving strategy can not only improve the energy-saving benefit, but also reduce the business risk. In addition, since the energy-saving control information is determined based on the traffic time-series data, the energy-saving strategy can be adjusted according to the traffic change situation, thereby improving the flexibility of the energy-saving control information. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a flowchart of an energy-saving control method provided by an embodiment of the present disclosure;

[0017] Figure 2 It is a flowchart of a training classification model provided by an embodiment of the present application;

[0018] Figure 3 It is a block diagram of an energy-saving control device provided by an embodiment of the present application;

[0019] Figure 4 It is a block diagram of an electronic device provided by an embodiment of the present application;

[0020] Figure 5 It is a block diagram of a computer-readable medium provided by an embodiment of the present application. DETAILED DESCRIPTION

[0021] To enable those skilled in the art to better understand the technical solutions of the present disclosure, the energy-saving control method, electronic device, and computer-readable storage medium provided by the present disclosure will be described in detail below with reference to the accompanying drawings.

[0022] Exemplary embodiments will be described more fully hereinafter with reference to the accompanying drawings. However, the exemplary embodiments may be embodied in different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0023] In the case of no conflict, the various embodiments of the present disclosure and the features in the embodiments may be combined with each other.

[0024] As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.

[0025] The terms used herein are only for describing specific embodiments and are not intended to limit the present disclosure. As used herein, the singular forms "a" and "the" are also intended to include the plural forms unless the context clearly indicates otherwise. It will also be understood that when the terms "comprises" and / or "consists of" are used in this specification, the specified features, integers, steps, operations, elements, and / or components are present, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0026] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art. It will also be understood that terms such as those defined in commonly used dictionaries should be interpreted as having a meaning that is consistent with their meaning in the context of the relevant art and the present disclosure, and will not be interpreted as having an idealized or overly formal meaning unless expressly so defined herein.

[0027] The transport network carries communication signals from the base station to the core network to achieve the transmission of various digital information. Moreover, it is necessary to provide corresponding levels of information structures, multiplexing methods, mapping methods, and related synchronization methods for the transmission of digital signals at different speeds. With the development of the Internet and the increasing demand for bandwidth, the transport network combines traditional telecommunication technologies and the multi-label protocol switching technology of the Internet to form the Packet Transport Network (PTN) technology. Since the number of enterprise-oriented services is gradually increasing, it is necessary to provide flexible dedicated line services with different bandwidths and different time delays for different enterprise customers. Therefore, different levels of network resources are isolated in different ways for different customers to use, and finally evolved into the Slicing Packet NewWork (SPN). This slicing packet network can carry a large-capacity bandwidth and achieve large-capacity transmission.

[0028] The transmission network devices adopted by the sliced packet network include devices such as cabinets, main control units, single boards, power management, and heat dissipation modules. In order to support high-capacity bandwidth, the transmission network devices usually set multiple single boards for processing and exchanging data. When the actual traffic carried by the device is low, some devices can be turned off to minimize power consumption on the premise of being able to support services normally.

[0029] In a first aspect, an energy-saving control method provided by an embodiment of the present disclosure.

[0030] Figure 1 It is a flowchart of an energy-saving control method provided by an embodiment of the present disclosure. Refer to Figure 1 , the energy-saving control method provided by an embodiment of the present disclosure includes:

[0031] Step S101, obtain the traffic time-series data carried by the controlled device.

[0032] Among them, the traffic time-series data is traffic data carrying time-series characteristics.

[0033] In some embodiments, the controlled device may be a network element device in a communication transmission network. The network element device includes devices at different levels such as single boards, link ports, modules, and chips. Among them, each single board includes one or more link ports and modules, each module may include one or more chips, and moreover, each single board, link port, module, and chip can be independently controlled to be turned on and off. Connection devices may also be included in the module, and the chip and the connection device jointly implement related functions.

[0034] Step S102, extract the characteristics of the traffic time-series data to obtain traffic characteristics.

[0035] The traffic characteristics in the embodiments of the present application are obtained based on the traffic time-series data. Therefore, the traffic characteristics can reflect characteristics such as the traffic level, smoothness, and periodicity of the network traffic carried by the controlled device.

[0036] In some embodiments, the traffic characteristics include one or more of time-domain characteristics, statistical characteristics, and frequency-domain characteristics.

[0037] The embodiments of the present application do not limit the time-domain characteristics. For example, the time-domain characteristics include but are not limited to moving average, variance of the moving average, autocorrelation coefficient, etc. The embodiments of the present application do not limit the statistical characteristics. For example, the statistical characteristics include but are not limited to mean, variance, peak value, etc. The embodiments of the present application do not limit the frequency-domain characteristics. For example, the frequency-domain characteristics include but are not limited to performing Fourier transform or wavelet transform on the time series, etc.

[0038] In some embodiments, in step S102, extracting the features of the flow time-series data to obtain flow features includes: preprocessing the flow time-series data to obtain the preprocessed flow time-series data; and extracting features based on the preprocessed flow time-series data to obtain flow features.

[0039] Preprocessing the flow time-series data enables the preprocessed flow time-series data to better reflect the flow characteristics of the controlled device, making the subsequent energy-saving strategy determined based on the flow features more accurate.

[0040] In some embodiments, preprocessing the flow time-series data to obtain the preprocessed flow time-series data includes one or more of the following methods:

[0041] Performing outlier filtering on the flow time-series data to obtain the filtered flow time-series data;

[0042] Performing missing value filling on the flow time-series data to obtain the filled flow time-series data;

[0043] Performing normalization on the flow time-series data to obtain the normalized flow time-series data.

[0044] The embodiments of the present disclosure do not limit the method of outlier filtering. For example, the outlier filtering method includes, but is not limited to, box plot detection, 3-sigma screening, isolation forest algorithm, etc.

[0045] The embodiments of the present disclosure do not limit the method of missing value filling. For example, the missing value filling method includes, but is not limited to, mean filling, adjacent value filling, etc.

[0046] The embodiments of the present disclosure do not limit the method of normalization. For example, the normalization method includes, but is not limited to, min-max normalization, Z-score standardization, etc.

[0047] In step S103, classifying the flow time-series data based on the flow features to obtain the flow scenario categories.

[0048] The embodiments of the present disclosure classify the flow time-series data based on the flow features, and the obtained flow scenario categories include one or more of the following:

[0049] Low flow scenario;

[0050] Medium flow and stable flow scenario;

[0051] Medium flow and large flow fluctuation scenario;

[0052] High flow scenario.

[0053] The embodiments of the present application do not limit the specific values of low flow rate, medium flow rate, and high flow rate. Users can set according to the actual situation.

[0054] It should be noted that although only four types of traffic scenario categories are listed here, the embodiments of the present application are not limited thereto. In fact, more traffic scenario categories can be set according to the situation. For example, the low traffic scenario can be further divided into the first traffic scenario, the second traffic scenario,..., the Nth traffic scenario according to the traffic volume. Again, according to the traffic smoothness and the magnitude of traffic fluctuations, more traffic scenario categories can be further divided and set. Also, the high traffic scenario can be further divided according to the traffic volume to set more traffic scenario categories. In short, only listing four types of traffic scenario categories here is only an illustration of the present application, not a limitation on the number of traffic scenario categories.

[0055] When the traffic carried by the controlled device is in the low traffic scenario, a relatively large number of single boards, chips, modules, and link ports can be turned off; when the traffic carried by the controlled device is in the high traffic scenario, a relatively small number of single boards, chips, modules, and link ports can be turned off, or even none of the single boards, chips, modules, and link ports are turned off, that is, all devices of the controlled device are turned on. When the traffic carried by the controlled device is medium traffic, an appropriate number of single boards, chips, modules, and link ports can be selected to be turned on according to the situation.

[0056] In the embodiments of the present disclosure, a single board refers to a switching board in the controlled device that processes the distribution of traffic volume.

[0057] Step S104, determine the energy-saving strategy based on the traffic scenario category.

[0058] In the embodiments of the present application, there is a corresponding relationship between the traffic scenario category and the energy-saving strategy, and the energy-saving strategy can be determined when the traffic scenario category is determined.

[0059] In some embodiments, the energy-saving strategy includes one or more of the following:

[0060] The strategy of turning off the single board in the controlled device;

[0061] The strategy of reducing the modules, link ports, and chips in the controlled device;

[0062] The strategy of reducing the operating frequency of the central processing unit and the power consumption of the programmable array logic in the controlled device;

[0063] The strategy of exiting the energy-saving mode.

[0064] In the embodiments of the present application, turning off some components in the controlled device may be turning off some single boards, and turning off some chips, modules, and link ports in the single boards. Reducing the operating frequency of the controlled device may be reducing the frequency of the Central Processing Unit (CPU), or reducing the frequency of other components. Reducing the power consumption of the internal components of the controlled device may be reducing the power consumption of the Field Programmable Gate Array (FPGA). Exiting the energy-saving mode means that all components in the controlled device exit the energy-saving mode.

[0065] Table 1 is a correspondence table between traffic scenario categories and energy-saving strategies. As can be seen from Table 1, the traffic scenario categories and energy-saving strategies may be in a one-to-one correspondence.

[0066] Table 1

[0067]

[0068]

[0069] In the embodiments of the present application, when the traffic scenario category is a low-traffic scenario, the single-board turning-off strategy is adopted, and the number of single boards that can be turned off or need to be turned on is calculated based on the maximum value of the traffic sequence carried by the controlled device and the maximum bandwidth that the controlled device can currently support. When the traffic volume is medium and the traffic time series fluctuates smoothly with obvious periodic changes, the strategy of turning off chips, modules, and link ports is adopted. When the traffic volume is medium, but the traffic time series fluctuates unevenly and has no obvious periodicity, the frequency reduction and FPGA power consumption reduction strategy is adopted, that is, the frequency of the CPU in the controlled device is reduced, and the power consumption of the FPGA is reduced, so that the CPU and FPGA in the controlled device operate in a low-power state. When the traffic volume is a high-traffic scenario, the strategy of exiting the energy-saving mode is adopted, that is, no energy-saving action is taken, and all components in the controlled device that have taken energy-saving measures are awakened, so that all components are in a normal working state.

[0070] Step S105, determining energy-saving control information based on the energy-saving strategy.

[0071] Among them, the control information is used to perform energy-saving control on the controlled device.

[0072] In some embodiments, step S105, determining energy-saving control information based on the energy-saving strategy, includes: determining the turn-on and turn-off information of each component in the controlled device based on the energy-saving strategy; determining the energy-saving control information based on the turn-on and turn-off information of each component.

[0073] Exemplarily, calculate the number of single boards that can be turned off or need to be turned on according to the energy-saving strategy. Assume that the maximum value of the traffic sequence is A. Select N single boards in sequence. When the total traffic volume that N single boards can handle ≤ the maximum value A of the traffic sequence, but the total traffic volume that N + 1 single boards can handle > the maximum value A of the traffic sequence, then select to turn on N + 1 single boards, and the remaining single boards can all be turned off. It should be noted that when only 1 single board needs to be turned on through calculation, in order to ensure the stability of the service, 2 single boards can be selected to be turned on. That is, in practical applications, the minimum number of single boards to be turned on is 2.

[0074] The method for calculating the number of modules, chips to be turned off or turned on, and the number of single-board link ports is as follows: Select M modules in sequence from all single boards, so that the total traffic volume that M modules can handle ≤ the maximum value A of the traffic sequence, and the total traffic volume that M + 1 modules can handle > the maximum value A of the traffic sequence.

[0075] Record the total traffic volume that M modules can handle as Mn. Select k chips in sequence from the (M + 1)-th module, so that the total traffic volume that k - 1 chips can handle + Mn ≤ the maximum value A of the traffic sequence, and the total traffic volume that k modules can handle + Mn > A. The method for calculating the number of link ports is as follows: Select N links in sequence, so that the total traffic volume that N links can handle ≤ A, and the total traffic volume that N + 1 links can handle > A.

[0076] The specific methods for downclocking the CPU in the controlled device and reducing the power consumption of the FPGA are as follows: Place it in the low-power mode under the condition of low CPU utilization rate, and configure the devices in the controlled device to the low-power mode.

[0077] In some embodiments, in step S103, classify the traffic time-series data based on the traffic characteristics to obtain the traffic scenario category, including: classify the traffic time-series data based on the traffic characteristics using the classification model to obtain the traffic scenario category.

[0078] Among them, the classification model can adopt the already trained BERT classification model, or can be trained using a new training method.

[0079] Figure 2 This is a flowchart of a method for training a classification model provided by an embodiment of the present application. As Figure 2 shown, the classification model is trained through the following steps:

[0080] Step S201, obtain training samples.

[0081] Among them, the training samples are samples annotated with tagging information. The tagging information includes one or more of high / low flow information, flow smoothness information, and flow periodicity information. The high / low flow information is used to represent the change of the magnitude of the flow carried by the controlled device over time, the flow smoothness information is used to represent the smoothness of the flow carried by the controlled device, and the flow periodicity information is used to represent the variation law of the flow carried by the controlled device over time.

[0082] In the embodiments of the present application, in order to improve the robustness of the classification model, the training samples can come from different network elements in the metropolitan area network and the backbone network, as well as the flow time series data of each network element in different time periods.

[0083] Step S202: Use the training samples to train the classification model to be trained.

[0084] Input the training samples into the classification model to be trained. The classification model to be trained outputs a classification result, which may or may not be consistent with the classification result in the tagging information. Different processing methods can be adopted for different classification results.

[0085] Step S203: When the classification result output by the classification model to be trained is inconsistent with the annotated tagging information, after adjusting the parameters in the classification model to be trained, continue to train the classification model to be trained.

[0086] When the classification result output by the classification model to be trained is inconsistent with the annotated tagging information, it indicates that the parameter settings in the classification model to be trained are unreasonable. It is necessary to adjust the parameters in the classification model to be trained, and then continue to iteratively train the classification model to be trained.

[0087] Step S204: When the classification result output by the classification model to be trained is consistent with the annotated tagging information, obtain the trained classification model.

[0088] When the classification result output by the classification model to be trained is consistent with the annotated tagging information, it indicates that the parameter settings in the classification model to be trained reach the ideal state, the classification model training is completed, and the trained classification model is obtained.

[0089] The embodiments of the present application can also set different cut-off conditions, such as setting a threshold for the number of generations. When training the classification model, a threshold for the number of iterations can also be set as the condition for ending the model training, that is, when the number of iterations of the model training reaches the threshold for the number of iterations, the trained classification model is obtained.

[0090] In order to obtain a more accurate traffic scenario category, the embodiments of the present application respectively obtain the traffic scenario categories through multiple classification models, and then determine the final traffic scenario category based on multiple traffic scenario categories.

[0091] In some embodiments, the classification model includes one or more of a tree model, a support vector machine model, an artificial neural network model, and a naive Bayes model.

[0092] In some embodiments, the classification model can not only output the traffic scenario category, but also output the probability value of each traffic scenario category.

[0093] In some embodiments, classifying the traffic time-series data by using a pre-trained classification model based on traffic characteristics to obtain a traffic scenario category includes:

[0094] Classifying the traffic time-series data by using two or more classification models based on traffic characteristics to obtain two or more candidate traffic scenario categories; obtaining the traffic scenario category based on the two or more candidate traffic scenario categories and the weight coefficient of each candidate traffic scenario category.

[0095] The embodiments of the present application do not limit the weight coefficient. For example, the weight coefficient can be set according to the accuracy of each classification model.

[0096] Exemplarily, when the classification model includes a tree model, a support vector machine model, and an artificial neural network model, each classification model is trained by using the training samples in step S201 to obtain a trained classification model. After extracting the traffic characteristics from the traffic time-series data carried by the controlled device, candidate traffic scenario categories are obtained based on the traffic characteristics by using the tree model, the support vector machine model, and the artificial neural network model respectively. Assume that the tree model outputs the probability value S11 of the first traffic scenario category, the probability value S12 of the second traffic scenario category, and the probability value S13 of the third traffic scenario category; the support vector machine model outputs the probability value S21 of the first traffic scenario category, the probability value S22 of the second traffic scenario category, and the probability value S23 of the third traffic scenario category; the artificial neural network model outputs the probability value S31 of the first traffic scenario category, the probability value S32 of the second traffic scenario category, and the probability value S33 of the third traffic scenario category. If the weights of the tree model, the support vector machine model, and the artificial neural network model are set as A1, A2, and A3 respectively, therefore, the probability value H1 corresponding to the first traffic scenario category is A1×S11 + A2×S21 + A3×S31, the probability value H2 corresponding to the second traffic scenario category is A1×S12 + A2×S22 + A3×S32, the probability value H3 corresponding to the second traffic scenario category is A1×S13 + A2×S23 + A3×S33, and the maximum value among H1, H2, and H3 is used as the final traffic scenario category.

[0097] The energy-saving control method provided by the embodiments of the present application extracts the features of the traffic time-series data carried by the controlled device to obtain traffic features. Compared with the traffic data without time series, the traffic features can better reflect the time-series characteristics of the traffic of the controlled device. Classify the traffic time-series data based on the traffic features of the traffic time-series data, and determine the energy-saving strategy according to the traffic scenario category. Compared with predicting traffic changes based on historical traffic data, it can more accurately reflect the traffic features at the current moment. Therefore, the energy-saving control information determined based on this energy-saving strategy can not only improve the energy-saving benefit, but also reduce the business risk. In addition, since this method determines the energy-saving control information based on the traffic time-series data, the energy-saving strategy can be adjusted according to the traffic change situation, thereby improving the flexibility of the energy-saving control information.

[0098] In a second aspect, the embodiments of the present application further provide an energy-saving control device.

[0099] Figure 3 It is a block diagram of an energy-saving control device provided by the embodiments of the present application. As Figure 3 shown, the energy-saving control device 300 includes:

[0100] The acquisition module 301 is configured to obtain the traffic time-series data carried by the controlled device, and the traffic time-series data is traffic data carrying time-series features.

[0101] The extraction module 302 is configured to extract the features of the traffic time-series data to obtain traffic features.

[0102] The classification module 303 is configured to classify the traffic time-series data based on the traffic features to obtain the traffic scenario category.

[0103] The determination module 304 is configured to determine the energy-saving strategy based on the traffic scenario category.

[0104] The generation module 305 is configured to generate energy-saving control information based on the energy-saving strategy; wherein, the control information is used to perform energy-saving control on the controlled device.

[0105] The energy-saving control device provided by the embodiments of the present application can implement any one of the energy-saving control methods provided by the embodiments of the present application. For the sake of saving space, it will not be elaborated here.

[0106] In some embodiments, the energy-saving control device further includes an execution module (not shown in the figure) for executing the energy-saving control information.

[0107] When using the energy-saving control device provided by the embodiments of the present application to perform energy-saving control on the traffic, first, the acquisition module 301 obtains the traffic time-series data carried by the controlled device, then the extraction module 302 and the classification module 303 are used to determine the traffic scenario category, and then the determination module 304 and the generation module 305 are used to generate the energy-saving control information. Finally, the execution module executes the corresponding energy-saving control information.

[0108] For the energy-saving control device provided by the embodiments of the present application, the acquisition module obtains the traffic time-series data carried by the controlled device, the extraction module extracts the features of the traffic time-series data carried by the controlled device to obtain the traffic features. Compared with the traffic data without time series, the traffic features can better reflect the time-series characteristics of the traffic of the controlled device. The classification module classifies the traffic time-series data based on the traffic features of the traffic time-series data. The determination module determines the energy-saving strategy according to the traffic scenario category. Compared with predicting the traffic change based on the historical traffic data, it can more accurately reflect the traffic features at the current moment. The energy-saving control information determined by the generation module based on this energy-saving strategy can not only improve the energy-saving benefit, but also reduce the service risk. In addition, since the device determines the energy-saving control information based on the traffic time-series data, it can adjust the energy-saving strategy according to the traffic change situation, thereby improving the flexibility of the energy-saving control information.

[0109] In a third aspect, referring to Figure 4 , the embodiments of the present disclosure provide an electronic device, which includes:

[0110] One or more processors 401;

[0111] A memory 402, on which one or more programs are stored. When the one or more programs are executed by the one or more processors, the one or more processors implement the energy-saving control method of any one of the above;

[0112] One or more I / O interfaces 403, connected between the processor and the memory, configured to implement information interaction between the processor and the memory.

[0113] Among them, the processor 401 is a device with data processing capabilities, which includes but is not limited to a central processing unit (CPU), etc.; the memory 402 is a device with data storage capabilities, which includes but is not limited to a random access memory (RAM, more specifically such as SDRAM, DDR, etc.), a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), a flash memory (FLASH); the I / O interface (read-write interface) 403 is connected between the processor 401 and the memory 402 and can implement information interaction between the processor 401 and the memory 402, which includes but is not limited to a data bus (Bus), etc.

[0114] In some embodiments, the processor 401, the memory 402, and the I / O interface 403 are interconnected via a bus 404 and, thus, connected to other components of the computing device.

[0115] In a fourth aspect, embodiments of the present disclosure provide a computer-readable medium. Figure 5 FIG. is a block diagram of a computer-readable medium provided for an embodiment of the present application. The computer-readable medium stores a computer program which, when executed by a processor, implements any of the above energy-saving control methods.

[0116] Those of ordinary skill in the art will understand that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, and appropriate combinations thereof. In a hardware implementation, the division of the functional modules / units mentioned above does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or one function or step may be executed by several physical components in cooperation. Some or all of the physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or may be implemented as hardware, or may be implemented as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, which can include a computer storage medium (or a non-transitory medium) and a communication medium (or a transitory medium). As is well known to those of ordinary skill in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information, such as computer-readable instructions, data structures, program modules, or other data. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disks (DVDs) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, as is well known to those of ordinary skill in the art, a communication medium typically includes computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transmission mechanism, and can include any information delivery medium.

[0117] Example embodiments have been disclosed herein, and although specific terms are employed, they are used only and should be interpreted only as general descriptive meanings and not for the purpose of limitation. In some instances, it will be apparent to those skilled in the art that, unless otherwise expressly specified, features, characteristics, and / or elements described in connection with a particular embodiment may be used alone or in combination with features, characteristics, and / or elements described in connection with other embodiments. Accordingly, those skilled in the art will understand that various forms and details of changes may be made without departing from the scope of the present disclosure as set forth by the appended claims.

Claims

1. An energy-saving control method, wherein, The method includes: Obtaining traffic time-series data carried by a controlled device, where the traffic time-series data is traffic data carrying time-series features; Extracting features of the traffic time-series data to obtain traffic features; Classifying the traffic time-series data based on the traffic features to obtain traffic scenario categories; Determining an energy-saving strategy based on the traffic scenario categories; Determining energy-saving control information based on the energy-saving strategy; where the control information is used to perform energy-saving control on the controlled device.

2. The method according to claim 1, wherein, The classifying the traffic time-series data based on the traffic features to obtain traffic scenario categories includes: Classifying the traffic time-series data based on the traffic features using two or more classification models to obtain multiple candidate traffic scenario categories; Obtaining the traffic scenario categories based on the multiple candidate traffic scenario categories and the weight coefficients of each candidate traffic scenario category.

3. The method according to claim 2, wherein, The classification models include one or more of a tree model, a support vector machine model, an artificial neural network model, and a naive Bayes model.

4. The method according to claim 2, wherein, The classification models are trained through the following steps: Obtaining training samples; where the training samples are samples labeled with labeling information; Training a classification model to be trained using the training samples; When the classification result output by the classification model to be trained is inconsistent with the labeled labeling information, adjusting the parameters in the classification model to be trained and then continuing to train the classification model to be trained; When the classification result output by the classification model to be trained is consistent with the labeled labeling information, obtaining the trained classification model.

5. The method according to claim 4, wherein, The training samples include traffic data of multiple network elements in multiple time periods, and / or The labeling information includes one or more of traffic high / low information, traffic smoothness information, and traffic periodicity information.

6. The method according to claim 1, wherein, The extracting features of the traffic time-series data to obtain traffic features includes: Performing preprocessing on the traffic time-series data to obtain preprocessed traffic time-series data; Performing feature extraction based on the preprocessed traffic time-series data to obtain the traffic features.

7. The method according to claim 6, wherein, The performing preprocessing on the traffic time-series data to obtain preprocessed traffic time-series data includes one or more of the following methods: Performing outlier filtering on the traffic time-series data to obtain filtered traffic time-series data; Performing missing value filling on the traffic time-series data to obtain filled traffic time-series data; Performing normalization on the traffic time-series data to obtain normalized traffic time-series data.

8. The method according to any one of claims 1-7, wherein, The traffic features include one or more of time-domain features, statistical features, and frequency-domain features; and / or, the traffic scenario categories include one or more of the following: Low traffic scenario; Medium traffic and stable traffic scenario; Medium traffic and large traffic fluctuation scenario; High traffic scenario; and / or, the energy-saving strategies include one or more of the following: Single-board strategy to turn off the controlled device; Strategies to reduce modules, link ports, and chips in the controlled device; Strategies to reduce the working frequency of the central processing unit and the power consumption of the programmable array logic in the controlled device; Strategy to exit the energy-saving mode.

9. An electronic device, comprising: One or more processors; A memory storing one or more programs which, when executed by the one or more processors, cause the one or more processors to implement the energy-saving control method according to any one of claims 1 to 8.

10. A computer-readable medium having stored thereon a computer program, which when executed by a processor, implements the energy-saving control method according to any one of claims 1 to 8.

Images