Energy consumption detection method and device, electronic equipment and storage medium

Abstract

The application provides an energy consumption detection method and device, electronic equipment and a storage medium. The energy consumption detection method can include: obtaining to-be-detected energy consumption information and inputting the to-be-detected energy consumption information into an energy consumption detection model, wherein the energy consumption detection model is trained by a training sample set with micro-moment features, and the micro-moment features are features about device energy consumption categories; and determining an energy consumption detection result corresponding to the to-be-detected energy consumption information based on an output result of the energy consumption detection model. The energy consumption detection method provided by the application can improve the authenticity and reliability of the energy consumption detection result, and lays a foundation for users to execute an energy-saving scheme based on the energy consumption detection result and track device faults.

Description

Technical Field

[0001] This invention relates to the field of energy consumption detection technology, and in particular to an energy consumption detection method, device, electronic device, and storage medium. Background Technology

[0002] Today, most user behaviors result in high energy costs, from keeping room lights on to watching television all day. Studies show that embedding an individual's occupational profile into a building energy management system could save approximately 10%–40% of electricity in a home. Energy consumption is expected to increase further in the coming years due to improved living conditions and increased use of appliances and electrical equipment.

[0003] According to relevant technologies, real-time detection and analysis of energy usage patterns can not only promote energy conservation but also help track equipment malfunctions by analyzing sudden and unexpected changes in energy use. Therefore, energy consumption monitoring technology is receiving increasing attention due to global energy efficiency considerations. Summary of the Invention

[0004] This invention provides an energy consumption detection method, device, electronic device, and storage medium to solve the problem of lacking a training set with real labels in the training process of energy consumption detection models in the prior art. It realizes the provision of a training set with real labels for energy consumption detection models, thereby improving the accuracy of energy consumption detection results of energy consumption detection models.

[0005] This invention provides an energy consumption detection method, characterized in that the energy consumption detection method includes: acquiring energy consumption information to be detected, and inputting the energy consumption information to be detected into an energy consumption detection model, wherein the energy consumption detection model is trained through a training sample set with micro-moment features, the micro-moment features being features related to the energy consumption category of the device; and determining an energy consumption detection result corresponding to the energy consumption information to be detected based on the output result of the energy consumption detection model.

[0006] According to an energy consumption detection method provided by the present invention, the training sample set with micro-moment features is obtained in the following manner: based on the device energy consumption information, the device micro-moment features corresponding to the device energy consumption information are obtained through a micro-moment energy labeling model, wherein the device energy consumption information includes at least device occupancy rate, device power consumption, device reference consumption rate, device maximum standby time, and device standby energy consumption; the training sample set with micro-moment features is obtained based on the device energy consumption information and the device micro-moment features.

[0007] According to an energy consumption detection method provided by the present invention, the micro-moment energy tag model is provided with an energy consumption information and micro-moment feature mapping table. The step of obtaining the device micro-moment feature corresponding to the device energy consumption information through the micro-moment energy tag model based on the device energy consumption information includes: determining the device micro-moment feature corresponding to the device energy consumption information through the energy consumption information and micro-moment feature mapping table based on the device energy consumption information.

[0008] According to the energy consumption detection method provided by the present invention, the energy consumption information and the micro-moment feature mapping table are determined in the following manner:

[0009] If the energy consumption information is the first energy consumption information, then the micro-moment feature corresponding to the first energy consumption information is the device in good use micro-moment feature, wherein the first energy consumption information is the device power consumption at the current time step being greater than or equal to the minimum device reference consumption rate and less than or equal to a preset multiple of the maximum device reference consumption rate; if the energy consumption information is the second energy consumption information, then the micro-moment feature corresponding to the second energy consumption information is the device on micro-moment feature, wherein the second energy consumption information is the device power consumption at the current time step being greater than or equal to the minimum device reference consumption rate and the device power consumption at the previous time step being less than or equal to the maximum device standby power consumption; if the energy consumption information is the third energy consumption information, then the micro-moment feature corresponding to the third energy consumption information is the device off micro-moment feature, wherein the third energy consumption information... The device power consumption at the current time step is less than or equal to the maximum device standby power consumption, and the device power consumption at the previous time step is greater than or equal to the minimum device reference power consumption rate. If the energy consumption information is the fourth energy consumption information, then the micro-moment feature corresponding to the fourth energy consumption information is the micro-moment feature of excessive device power consumption, wherein the fourth energy consumption information is that the device power consumption at the current time step is greater than or equal to a preset multiple of the maximum device reference power consumption rate, and the device's longest standby time at the current time step is greater than or equal to the maximum device longest standby time. If the energy consumption information is the fifth energy consumption information, then the micro-moment feature corresponding to the fifth energy consumption information is the micro-moment feature of abnormal device power consumption when the device is away, wherein the fifth energy consumption information is that the device is unoccupied indoors, and the device power consumption at the current time step is greater than or equal to a preset multiple of the maximum device standby power consumption.

[0010] According to an energy consumption detection method provided by the present invention, the power consumption of the device is determined in the following manner: within a preset time interval, the power consumption of a plurality of first devices is determined; based on the power consumption of the plurality of first devices, the power consumption of the device is determined.

[0011] According to the energy consumption detection method provided by the present invention, the power consumption of the device is determined based on the power consumption of multiple first devices using the following formula:

[0012]

[0013] Among them, P N P(t) represents the power consumption of the device at time t, P(t) represents the power consumption of the first device at time t, mean(P) represents the average power consumption of multiple first devices within a preset time interval, max(P) represents the maximum power consumption of the first device within a preset time interval, and min(P) represents the minimum power consumption of the first device within a preset time interval.

[0014] According to an energy consumption detection method provided by the present invention, the device energy consumption information is determined by the following method: acquiring first device energy consumption information; performing interpolation processing on the first device energy consumption information, and using the processed first device energy consumption information as the device energy consumption information.

[0015] The present invention also provides an energy consumption detection device, characterized in that the energy consumption detection device includes: an acquisition module, used to acquire energy consumption information to be detected and input the energy consumption information to be detected into an energy consumption detection model, wherein the energy consumption detection model is trained through a training sample set with micro-moment features, the micro-moment features being features related to the energy consumption category of the device; and a processing module, used to determine an energy consumption detection result corresponding to the energy consumption information to be detected based on the output result of the energy consumption detection model.

[0016] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of any of the above-described energy consumption detection methods.

[0017] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the above-described energy consumption detection methods.

[0018] The energy consumption detection method, apparatus, electronic device, and storage medium provided by this invention improve the detection accuracy of the energy consumption detection model by training the model with a training sample set featuring micro-momentary characteristics. Furthermore, obtaining energy consumption detection results corresponding to the energy consumption information to be detected based on the energy consumption detection model enhances the authenticity and reliability of the results, laying the foundation for users to implement energy-saving solutions and track equipment faults based on the energy consumption detection results. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0020] Figure 1 This is one of the flowcharts of the energy consumption detection method provided by the present invention;

[0021] Figure 2 This is one of the flowcharts provided by the present invention for determining a training sample set with micro-time features;

[0022] Figure 3 This is one of the flowcharts for determining device power consumption provided by the present invention;

[0023] Figure 4 This is one of the flowcharts for determining equipment energy consumption information provided by the present invention;

[0024] Figure 5 This is a schematic diagram illustrating an application scenario of the energy consumption detection method provided by the present invention;

[0025] Figure 6 This is a schematic diagram of the energy consumption detection device provided by the present invention;

[0026] Figure 7 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0028] According to relevant technologies, both overall energy efficiency and plug load are closely related to occupancy rate. Furthermore, the presence / absence of people in a building significantly impacts energy consumption, as many users keeping devices on for extended periods without actually being present represents abnormal behavior. For example, keeping televisions, air conditioners, lights, laptops / desktop computers, and fans on for extended periods when no one is present.

[0029] Real-time detection and analysis of energy usage patterns not only promotes energy conservation but also helps track equipment malfunctions by analyzing sudden and unexpected changes in energy use. During application, users can be notified if abnormal energy usage behavior is identified, allowing them to implement appropriate power efficiency measures. Furthermore, the proliferation of wireless sensors and sub-meters makes the detection of abnormal usage in residential buildings increasingly urgent. Therefore, energy consumption detection technology is receiving growing attention due to global energy efficiency considerations.

[0030] Energy consumption detection faces numerous difficulties and challenges. Specifically, one of the main obstacles to developing and evaluating energy consumption detection technologies is the lack of labeled real-world datasets. In short, there is a lack of discussion on how to label energy consumption observations as normal or anomalous, and what kind of anomalousness anomalous observations fall into. To address this, this invention proposes a method for labeling energy consumption events using device occupancy patterns, power consumption footprints, and micro-moment features.

[0031] The energy consumption detection method provided by this invention improves the detection accuracy of the energy consumption detection model by training an energy consumption detection model with a training sample set having micro-moment features. Furthermore, obtaining energy consumption detection results corresponding to the energy consumption information to be detected based on the energy consumption detection model can improve the authenticity and reliability of the energy consumption detection results, laying the foundation for users to implement energy-saving solutions and track equipment faults based on the energy consumption detection results.

[0032] The present invention will describe the process of the energy consumption detection method in conjunction with the following embodiments.

[0033] Figure 1 This is one of the flowcharts of the energy consumption detection method provided by the present invention.

[0034] In an exemplary embodiment of the present invention, such as Figure 1 As shown, the energy consumption detection method may include steps 110 and 120, which will be described in detail below.

[0035] In step 110, the energy consumption information to be detected is obtained and input into the energy consumption detection model, wherein the energy consumption detection model is trained using a training sample set with micro-moment features. The micro-moment features are features related to the energy consumption category of the device.

[0036] In step 120, based on the output of the energy consumption detection model, the energy consumption detection result corresponding to the energy consumption information to be detected is determined.

[0037] In one embodiment, energy consumption information to be detected can be acquired, which may be energy usage information about equipment inside a building. In one example, equipment energy usage information can be collected based on energy sensors or equipment occupancy sensors. The energy consumption information to be detected may include equipment occupancy rate (denoted by O), equipment power consumption (denoted by P), equipment reference consumption rate (denoted by A), equipment maximum standby time (denoted by TM), and equipment standby energy consumption (denoted by S), etc. Furthermore, the energy consumption information to be detected can be input into an energy consumption detection model, and based on the output of the energy consumption detection model, an energy consumption detection result corresponding to the energy consumption information to be detected can be obtained.

[0038] In one embodiment, the energy consumption detection model can be a deep neural network model, trained on a training sample set with micro-moment features. It is understood that micro-moment features are characteristics related to the energy consumption category of the device. In one example, micro-moment features can include two types: normal consumption and abnormal consumption. Further, normal consumption can also include three subcategories: good usage, device on, and device off. Abnormal consumption can include two subcategories: excessive consumption and consumption while away from home. It is understood that the energy consumption detection result corresponding to the energy consumption information to be detected can include detection results related to good usage, device on, device off, excessive consumption, and consumption while away from home.

[0039] The energy consumption detection method provided by this invention improves the detection accuracy of the energy consumption detection model by training it with a training sample set containing micro-moment features. Furthermore, obtaining energy consumption detection results corresponding to the energy consumption information to be detected based on the energy consumption detection model enhances the authenticity and reliability of the results, laying the foundation for users to implement energy-saving solutions and track equipment faults based on the energy consumption detection results.

[0040] To further illustrate the energy consumption detection method provided by this invention, the process of determining a training sample set with micro-moment features will be explained in conjunction with the following embodiments.

[0041] Figure 2 This is one of the flowcharts provided by the present invention for determining a training sample set with micro-time features.

[0042] In an exemplary embodiment of the present invention, such as Figure 2 As shown, determining the training sample set with micro-time features may include steps 210 to 220, which will be described in detail below.

[0043] In step 210, based on the device energy consumption information, the device micro-moment characteristics corresponding to the device energy consumption information are obtained through a micro-moment energy tagging model. The device energy consumption information includes at least the device occupancy rate, device power consumption, device reference consumption rate, device maximum standby time, and device standby energy consumption.

[0044] In step 220, a training sample set with micro-moment features is obtained based on the device energy consumption information and the device micro-moment features.

[0045] In one implementation, based on the device energy consumption information of each device, such as device occupancy rate (represented by O), device power consumption (represented by P), device reference consumption rate (represented by A), device maximum standby time (represented by TM), and device standby energy consumption (represented by S), the device micro-moment features corresponding to the device energy consumption information can be obtained through a micro-moment energy tag model.

[0046] Furthermore, the obtained device micro-moment features can be used as labels for the device's energy consumption information, and together with the device's energy consumption information, they can form a training sample set for the energy consumption detection model, that is, a training sample set with micro-moment features.

[0047] In an exemplary embodiment of the present invention, the micro-moment energy tag model may further include a mapping table between energy consumption information and micro-moment features. In one example, based on the device's energy consumption information, the device's micro-moment features corresponding to the device's energy consumption information can be determined through the mapping table.

[0048] The process of determining the mapping table between energy consumption information and micro-moment features will be explained below.

[0049] In an exemplary embodiment of the present invention, the energy consumption information and the micro-moment feature mapping table are determined in the following manner:

[0050] If the energy consumption information is the first energy consumption information, then the micro-moment feature corresponding to the first energy consumption information is the micro-moment feature of good device usage. The first energy consumption information is a preset multiple by which the device power consumption (P(t)) at the current time step is greater than or equal to the minimum device reference consumption rate (A) and less than or equal to the maximum device reference consumption rate (A). The preset multiple can be adjusted according to actual conditions; for example, the preset multiple can be 0.95 times. In this embodiment, the preset multiple is not specifically limited.

[0051] If the energy consumption information is the second energy consumption information, then the micro-moment feature corresponding to the second energy consumption information is the device turn-on micro-moment feature. Among them, the second energy consumption information is that the device power consumption (P(t)) at the current time step is greater than or equal to the minimum device activity consumption rate (A) and the device power consumption (P(t-1)) at the previous time step is less than or equal to the maximum device standby power consumption (S).

[0052] If the energy consumption information is the third energy consumption information, then the micro-moment feature corresponding to the third energy consumption information is the device shutdown micro-moment feature. Among them, the third energy consumption information is that the device power consumption (P(t)) at the current time step is less than or equal to the maximum device standby power consumption (S) and the device power consumption (P(t-1)) at the previous time step is greater than or equal to the minimum device activity consumption rate (A).

[0053] If the energy consumption information is the fourth energy consumption information, then the micro-moment feature corresponding to the fourth energy consumption information is the micro-moment feature of excessive device energy consumption. Among them, the fourth energy consumption information is that the device power consumption (P(t)) at the current time step is greater than or equal to a preset multiple of the maximum device activity consumption rate (A) and the device's longest standby time (TM(t)) at the current time step is greater than or equal to the maximum device longest standby time (TM).

[0054] If the energy consumption information is the fifth energy consumption information, then the micro-moment feature corresponding to the fifth energy consumption information is the micro-moment feature of abnormal device energy consumption when the device is away. Among them, the fifth energy consumption information is the device power consumption (P(t)) at the current time step when no one is indoors and the device power consumption (P(t)) is greater than or equal to a preset multiple of the maximum device standby power consumption (S).

[0055] The process of extracting device micro-moment features based on the micro-moment energy tag model will be described below with reference to the following embodiments.

[0056] Inputs: device occupancy (O), device power consumption (P), device reference power consumption per device (A), device maximum standby time (TM), and device standby power consumption (S).

[0057] Output: Micro-time eigenvectors (MF)

[0058] Initialize MF

[0059] While t≤N do

[0060] if P(t)≥min(A)and P(t)≤0.95×max(A)then

[0061] MF(t) = 0 (Good usage);

[0062] else if P(t)≥min(A)and P(t-1)≤max(S)then

[0063] MF(t) = 1 (Turn on device);

[0064] else if P(t)≤max(S)and P(t-1)≥min(A)then

[0065] MF(t) = 2 (Turn off device);

[0066] else if P(t)≥0.95×max(A)or TM(t)≥max(T)then

[0067] MF(t)=3(Excessive consumption);

[0068] else

[0069] if O(t)=0and P(t)≥0.95×max(S)then

[0070] MF(t)=4(Consumption while outside);

[0071] end

[0072] end

[0073] The process of determining device power consumption will be explained below with reference to the following embodiments.

[0074] Figure 3 This is one of the flowcharts for determining device power consumption provided by the present invention.

[0075] In an exemplary embodiment of the present invention, such as Figure 3 As shown, the process of determining device power consumption may include steps 310 and 320, which will be described in detail below.

[0076] In step 310, the power consumption of multiple first devices is determined within a preset time interval.

[0077] In step 320, the device power consumption is determined based on the power consumption of multiple first devices.

[0078] Most energy consumption databases collected through experimental activities for energy consumption detection can represent real-world patterns, but some outliers exist, which can affect the accuracy of the model when training based on these outliers. In this embodiment, the accuracy of training samples can be improved by normalizing the device power consumption. In one embodiment, multiple first device power consumptions can be acquired within a preset time period, and the device power consumption can be determined based on these multiple first device power consumptions.

[0079] In an exemplary embodiment of the present invention, the power consumption of a device can be determined by the following formula (Formula 1) based on the power consumption of multiple first devices:

[0080]

[0081] Among them, PN P(t) represents the device power consumption at time t, P(t) represents the power consumption of the first device at time t, mean(P) represents the average power consumption of multiple first devices within a preset time interval, max(P) represents the maximum power consumption of the first device within the preset time interval, and min(P) represents the minimum power consumption of the first device within the preset time interval. This embodiment effectively avoids the adverse effects of individual abnormal data on the accuracy of training samples.

[0082] The present invention will describe the process of determining device energy consumption information in conjunction with the following embodiments.

[0083] Figure 4 This is one of the flowcharts for determining equipment energy consumption information provided by the present invention.

[0084] In an exemplary embodiment of the present invention, such as Figure 4 As shown, the process of determining equipment energy consumption information may include steps 410 and 420, which will be described in detail below.

[0085] In step 410, the energy consumption information of the first device is obtained.

[0086] In step 420, the energy consumption information of the first device is interpolated, and the processed energy consumption information of the first device is used as the device energy consumption information.

[0087] Data collected from various energy and occupancy sensors can first undergo cleaning and preprocessing to remove or correct invalid records. It is understood that the collected footprint is raw or incomplete data, containing missing values, and some attribute information may be lost during collection. The lack of these values ​​is usually attributed to hardware and / or software malfunctions of the measuring equipment. Furthermore, other data is noisy, i.e., it contains errors or outliers. Therefore, a data cleaning process is essential.

[0088] In one embodiment, the attribute average can be used to fill in all missing values ​​in the power dataset. In one example, multiple first device energy consumption information can be obtained, interpolation can be performed based on the first device energy consumption information, and the processed first device energy consumption information can be used as the device energy consumption information. This embodiment effectively fills in all missing values ​​in the power dataset, achieving preprocessing of the collected data.

[0089] To further illustrate the energy consumption detection method provided by this invention, the invention will be described in conjunction with the following embodiments.

[0090] Figure 5 This is a schematic diagram illustrating an application scenario of the energy consumption detection method provided by the present invention.

[0091] In an exemplary embodiment of the present invention, such as Figure 5 As shown, this example illustrates a resident living in a building, where the resident's room includes household appliances. During application, occupancy data can be collected using energy sensors or occupancy sensors. The occupancy data can be information on energy consumption to be detected.

[0092] In another embodiment, the check-in data can also be the raw data used to train the energy consumption detection model. During application, the energy consumption detection model can be trained based on a training sample set with micro-momentary features. Further, based on the output of the energy consumption detection model, the energy consumption detection result corresponding to the energy consumption information to be detected is determined. It is understood that the energy consumption detection result can be the energy consumption detection result for normal device power consumption, device being turned on, device being turned off, excessive device power consumption, device power consumption while the user is away, etc.

[0093] In one example, based on device energy consumption information, a micro-moment energy labeling model can be used to obtain device micro-moment features corresponding to the device energy consumption information. Based on the device energy consumption information and the device micro-moment features, a training sample set with micro-moment features can be obtained. The device energy consumption information can include device occupancy rate (O), device power consumption (P), device reference consumption rate (A), device maximum standby time (TM), and device standby power consumption (S), etc.

[0094] In one example, energy markers at specific micro-moments can be obtained based on the power consumption specifications of different home appliances. Furthermore, the power consumption specifications of different home appliances and their corresponding energy markers at specific micro-moments can form a training sample set for training an energy consumption detection model.

[0095] After labeling the data using a micro-moment energy labeling model (micro-moment features), a dataset for training the energy consumption detection model can be obtained. In one example, a dataset with five micro-moment feature labels can be obtained: normal device power consumption, device on, device off, device excessive power consumption, and device power consumption while away from home. During application, the DNN model (energy consumption detection model) can be trained using training data containing these five micro-moment feature labels. In one example, K-fold cross-validation can be deployed for training and testing. The training process is a statistical analysis process that means dividing the input data and its labels into K subgroups, then supervising the training of (K-1) subgroups, and using the remaining subgroups to evaluate the model's output performance in terms of accuracy and F1 score. During application, the training can be repeated K times, with each subgroup used (K-1) times to train the DNN model and once for DNN model detection. During the training phase, input data vectors, including timestamps, device IDs, device occupancy (O), device power consumption (P), device reference power consumption (A), device maximum standby time (TM), and device standby power consumption (S), along with their corresponding micro-moment features (MF), are fed into the DNN model. The DNN model can contain multiple hidden layers to learn the relationship between normal and abnormal power consumption behaviors. The Rectified Linear Unit (ReLU) activation function can be used at the output.

[0096] As described above, the energy consumption detection method provided by this invention improves the detection accuracy of the energy consumption detection model by training the model with a training sample set containing micro-moment features. Furthermore, obtaining energy consumption detection results corresponding to the energy consumption information to be detected based on the energy consumption detection model enhances the authenticity and reliability of the results, laying the foundation for users to implement energy-saving solutions and track equipment faults based on the energy consumption detection results.

[0097] Based on the same concept, the present invention also provides an energy consumption detection device.

[0098] The energy consumption detection device provided by the present invention is described below. The energy consumption detection device described below can be referred to in correspondence with the energy consumption detection method described above.

[0099] Figure 6 This is a schematic diagram of the energy consumption detection device provided by the present invention.

[0100] In an exemplary embodiment of the present invention, such as Figure 6 As shown, the energy consumption detection device may include an acquisition module 610 and a processing module 620. Each module will be described in detail below.

[0101] The acquisition module 610 can be configured to: acquire the energy consumption information to be detected and input the energy consumption information to be detected into the energy consumption detection model, wherein the energy consumption detection model is trained by a training sample set with micro-moment features, and the micro-moment features are features related to the energy consumption category of the device.

[0102] The processing module 620 can be configured to: determine the energy consumption detection result corresponding to the energy consumption information to be detected based on the output of the energy consumption detection model.

[0103] In an exemplary embodiment of the present invention, the acquisition module 610 may acquire a training sample set with micro-time features in the following manner:

[0104] Based on device energy consumption information, a micro-moment energy labeling model is used to obtain device micro-moment features corresponding to the device energy consumption information. The device energy consumption information includes at least device occupancy rate, device power consumption, device activity consumption rate, device maximum standby time, and device standby energy consumption. Based on the device energy consumption information and device micro-moment features, a training sample set with micro-moment features is obtained.

[0105] In an exemplary embodiment of the present invention, the micro-moment energy tag model is configured with a mapping table between energy consumption information and micro-moment features. The acquisition module 610 can obtain the device micro-moment features corresponding to the device energy consumption information based on the device energy consumption information and through the micro-moment energy tag model in the following manner:

[0106] Based on equipment energy consumption information, the equipment micro-moment features corresponding to the equipment energy consumption information are determined through a mapping table between energy consumption information and micro-moment features.

[0107] In an exemplary embodiment of the present invention, the acquisition module 610 may determine the energy consumption information and micro-moment feature mapping table in the following manner:

[0108] If the energy consumption information is the first energy consumption information, then the micro-moment feature corresponding to the first energy consumption information is the device in good use micro-moment feature, wherein the first energy consumption information is the device power consumption at the current time step being greater than or equal to the minimum device activity consumption rate and less than or equal to a preset multiple of the maximum device reference consumption rate; if the energy consumption information is the second energy consumption information, then the micro-moment feature corresponding to the second energy consumption information is the device on micro-moment feature, wherein the second energy consumption information is the device power consumption at the current time step being greater than or equal to the minimum device reference consumption rate and the device power consumption at the previous time step being less than or equal to the maximum device standby power consumption; if the energy consumption information is the third energy consumption information, then the micro-moment feature corresponding to the third energy consumption information is the device off micro-moment feature, wherein the third energy consumption information is the current... The device power consumption at the current time step is less than or equal to the maximum device standby power consumption, and the device power consumption at the previous time step is greater than or equal to the minimum device reference power consumption rate. If the power consumption information is the fourth power consumption information, then the micro-moment feature corresponding to the fourth power consumption information is the micro-moment feature of excessive device power consumption, wherein the fourth power consumption information is that the device power consumption at the current time step is greater than or equal to a preset multiple of the maximum device reference power consumption rate, and the device's longest standby time at the current time step is greater than or equal to the maximum device longest standby time. If the power consumption information is the fifth power consumption information, then the micro-moment feature corresponding to the fifth power consumption information is the micro-moment feature of abnormal device power consumption when the user is away, wherein the fifth power consumption information is that the user is indoors and the device power consumption at the current time step is greater than or equal to a preset multiple of the maximum device standby power consumption.

[0109] In an exemplary embodiment of the present invention, the acquisition module 610 may determine the device power consumption in the following manner:

[0110] Within a preset time interval, the power consumption of multiple first devices is determined; based on the power consumption of the multiple first devices, the power consumption of the device is determined.

[0111] In an exemplary embodiment of the present invention, the acquisition module 610 may use the following formula (Formula 2) to determine the device power consumption based on the power consumption of multiple first devices:

[0112]

[0113] Among them, P N P(t) represents the power consumption of the device at time t, P(t) represents the power consumption of the first device at time t, mean(P) represents the average power consumption of multiple first devices within a preset time interval, max(P) represents the maximum power consumption of the first device within a preset time interval, and min(P) represents the minimum power consumption of the first device within a preset time interval.

[0114] In an exemplary embodiment of the present invention, the acquisition module 610 may determine the device energy consumption information in the following manner:

[0115] Obtain the energy consumption information of the first device; perform interpolation processing on the energy consumption information of the first device, and use the processed energy consumption information of the first device as the device energy consumption information.

[0116] Figure 7 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 7 As shown, the electronic device may include a processor 710, a communications interface 720, a memory 730, and a communication bus 740. The processor 710, communications interface 720, and memory 730 communicate with each other via the communication bus 740. The processor 710 can call logical instructions in the memory 730 to execute an energy consumption detection method. The energy consumption detection method may include: acquiring energy consumption information to be detected and inputting this information into an energy consumption detection model, wherein the energy consumption detection model is trained using a training sample set with micro-moment features, where the micro-moment features are features related to the energy consumption category of the device; and determining the energy consumption detection result corresponding to the energy consumption information to be detected based on the output of the energy consumption detection model.

[0117] Furthermore, the logical instructions in the aforementioned memory 730 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0118] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the energy consumption detection method provided by the above methods. The energy consumption detection method may include: acquiring energy consumption information to be detected and inputting the energy consumption information to be detected into an energy consumption detection model, wherein the energy consumption detection model is trained through a training sample set with micro-moment features, the micro-moment features being features related to the energy consumption category of the device; and determining the energy consumption detection result corresponding to the energy consumption information to be detected based on the output result of the energy consumption detection model.

[0119] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the energy consumption detection method provided by the above methods. The energy consumption detection method may include: acquiring energy consumption information to be detected and inputting the energy consumption information to be detected into an energy consumption detection model, wherein the energy consumption detection model is trained through a training sample set with micro-moment features, the micro-moment features being features related to the energy consumption category of the device; and determining the energy consumption detection result corresponding to the energy consumption information to be detected based on the output result of the energy consumption detection model.

[0120] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0121] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0122] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An energy consumption detection method, characterized in that, The method includes: The energy consumption information to be detected is obtained and input into the energy consumption detection model. The energy consumption detection model is trained by a training sample set with micro-moment features, which are features related to the energy consumption category of the device. Based on the output of the energy consumption detection model, the energy consumption detection result corresponding to the energy consumption information to be detected is determined, wherein the training sample set with micro-moment features is obtained in the following way: Based on the device energy consumption information, the device micro-moment features corresponding to the device energy consumption information are obtained through the micro-moment energy tag model. The device energy consumption information includes at least the device occupancy rate, device power consumption, device reference consumption rate, device maximum standby time, and device standby energy consumption. Based on the device energy consumption information and the device micro-moment features, the training sample set with micro-moment features is obtained.

2. The energy consumption detection method according to claim 1, characterized in that, The micro-moment energy tagging model is configured with a mapping table between energy consumption information and micro-moment features. The process of obtaining the device micro-moment features corresponding to the device energy consumption information through the micro-moment energy tagging model, based on the device energy consumption information, includes: Based on the device energy consumption information, the device micro-moment features corresponding to the device energy consumption information are determined through the mapping table between the energy consumption information and the micro-moment features.

3. The energy consumption detection method according to claim 2, characterized in that, The mapping table between energy consumption information and micro-moment features is determined in the following way: If the energy consumption information is the first energy consumption information, then the micro-moment feature corresponding to the first energy consumption information is the micro-moment feature of good device use, wherein the first energy consumption information is the device power consumption at the current time step is greater than or equal to the minimum device reference consumption rate and less than or equal to a preset multiple of the maximum device reference consumption rate. If the energy consumption information is the second energy consumption information, then the micro-moment feature corresponding to the second energy consumption information is the device turn-on micro-moment feature, wherein the second energy consumption information is that the device power consumption at the current time step is greater than or equal to the minimum device reference consumption rate and the device power consumption at the previous time step is less than or equal to the maximum device standby power consumption. If the energy consumption information is the third energy consumption information, then the micro-moment feature corresponding to the third energy consumption information is the device shutdown micro-moment feature, wherein the third energy consumption information is the device power consumption at the current time step being less than or equal to the maximum device standby power consumption and the device power consumption at the previous time step being greater than or equal to the minimum device reference power consumption rate. If the energy consumption information is the fourth energy consumption information, then the micro-moment feature corresponding to the fourth energy consumption information is the micro-moment feature of excessive device energy consumption, wherein the fourth energy consumption information is that the device power consumption at the current time step is greater than or equal to a preset multiple of the maximum device reference consumption rate and the device's longest standby time at the current time step is greater than or equal to the maximum device longest standby time. If the energy consumption information is the fifth energy consumption information, then the micro-moment feature corresponding to the fifth energy consumption information is the abnormal energy consumption micro-moment feature of the device when the user is away. The fifth energy consumption information is the device power consumption at the current time step when the user is not indoors and the device power consumption is greater than or equal to a preset multiple of the maximum standby power consumption of the device.

4. The energy consumption detection method according to claim 1, characterized in that, The power consumption of the device is determined in the following way: Within a preset time interval, the power consumption of multiple first devices is determined; The power consumption of the device is determined based on the power consumption of multiple first devices.

5. The energy consumption detection method according to claim 4, characterized in that, The power consumption of the device is determined based on the power consumption of multiple first devices using the following formula: in, This represents the device power consumption at time t. This represents the power consumption of the first device at time t. This represents the average power consumption of multiple first devices within a preset time interval. This indicates the maximum power consumption of the first device within a preset time interval. This represents the minimum power consumption of the first device within a preset time interval.

6. The energy consumption detection method according to claim 1, characterized in that, The device energy consumption information is determined using the following method: Obtain the energy consumption information of the first device; The energy consumption information of the first device is interpolated, and the processed energy consumption information of the first device is used as the energy consumption information of the device.

7. An energy consumption detection device, characterized in that, The apparatus is used to implement the energy consumption detection method according to any one of claims 1 to 6, and the apparatus comprises: An acquisition module is used to acquire energy consumption information to be detected and input the energy consumption information to be detected into an energy consumption detection model, wherein the energy consumption detection model is trained through a training sample set with micro-moment features, and the micro-moment features are features related to the energy consumption category of the device; The processing module is used to determine the energy consumption detection result corresponding to the energy consumption information to be detected based on the output result of the energy consumption detection model.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the energy consumption detection method as described in any one of claims 1 to 6.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the energy consumption detection method as described in any one of claims 1 to 6.

Images