Energy consumption anomaly detection method, device and electronic equipment based on spatiotemporal feature fusion
By acquiring time-series energy consumption data and spatial distribution information of energy-consuming facilities, and combining the spatiotemporal feature fusion method, a time-series graph convolutional network is used to identify energy consumption anomalies. This solves the problem of insufficient accuracy in energy consumption anomaly detection in existing technologies and achieves higher detection accuracy.
Patent Information
- Application Number
- CN202611504403.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-24
- Publication Date
- 2026-11-17
AI Technical Summary
Existing energy consumption anomaly detection methods ignore the influence of the external environment in which the equipment or building is located, resulting in insufficient accuracy in energy consumption anomaly detection.
By acquiring time-series energy consumption data and spatial distribution information of multiple energy-consuming facilities within the target area, and combining the spatiotemporal feature fusion method, differential energy consumption values and the energy consumption deviation amplitude of adjacent facilities are calculated, and energy consumption anomalies are identified using a time-series graph convolutional network.
It improves the accuracy of detecting abnormal energy consumption in energy-consuming facilities, avoids potential errors in judging energy consumption data from a single device, and comprehensively captures changes in energy consumption of energy-consuming facilities.
Smart Images

Figure CN123116098A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of Internet of Things (IoT) technology, and in particular to a method, apparatus, and electronic device for detecting energy consumption anomalies based on spatiotemporal feature fusion. Background Technology
[0002] Energy consumption anomaly detection is a key task in comprehensive building energy conservation, even surpassing the importance of energy-saving algorithms themselves. In recent years, the rapid proliferation of smart meters has spurred research into energy consumption anomaly detection based on smart meter data. Existing energy consumption anomaly detection methods often only consider the time-series energy consumption characteristics of a single device or building, neglecting the influence of the external environment. This can lead to misjudgments of energy consumption anomalies, affecting the accuracy of energy consumption anomaly detection.
[0003] The above problems urgently need to be addressed. Summary of the Invention
[0004] The purpose of this invention is to at least partially solve one of the technical problems existing in the prior art.
[0005] Therefore, one objective of this invention is to provide an energy consumption anomaly detection method based on spatiotemporal feature fusion, which improves the accuracy of energy consumption anomaly detection for energy-consuming facilities.
[0006] Another objective of this invention is to provide an energy consumption anomaly detection device based on spatiotemporal feature fusion.
[0007] To achieve the above-mentioned technical objectives, the technical solutions adopted in the embodiments of the present invention include: On one hand, embodiments of the present invention provide a method for detecting energy consumption anomalies based on spatiotemporal feature fusion, comprising the following steps: Acquire time-series energy consumption data of multiple energy-consuming facilities within a target area during a preset time period, and acquire spatial distribution information of each energy-consuming facility; The differential energy consumption value of each energy-consuming facility in each sampling interval is determined based on the time-series energy consumption data, and the first energy consumption deviation amplitude of each energy-consuming facility in each sampling interval and the previous sampling interval is calculated based on the differential energy consumption value. The second energy consumption deviation amplitude between each energy-consuming facility and its neighboring facilities is determined based on the differential energy consumption value and the spatial distribution information at each sampling interval; The energy consumption of each energy-consuming facility is determined to be abnormal based on the first energy consumption deviation range and the second energy consumption deviation range.
[0008] Further, in one embodiment of the present invention, the step of determining the differential energy consumption value of each energy-consuming facility in each sampling interval based on the time-series energy consumption data, and calculating the first energy consumption deviation amplitude of each energy-consuming facility in each sampling interval compared with the previous sampling interval based on the differential energy consumption value, specifically includes: The cumulative energy consumption value of each energy-consuming facility at each sampling time is determined based on the time-series energy consumption data; The cumulative energy consumption value of the energy-consuming facility at adjacent sampling times is calculated differentially to obtain the differential energy consumption value for the corresponding sampling interval; Calculate the unit energy consumption rate of the corresponding sampling interval based on the differential energy consumption value; The energy consumption rate growth rate of the energy-consuming facility in adjacent sampling intervals is calculated based on the unit energy consumption rate to obtain the first energy consumption deviation amplitude.
[0009] Furthermore, in one embodiment of the present invention, determining the second energy consumption deviation amplitude between each energy-consuming facility and its neighboring facilities at each sampling interval based on the differential energy consumption value and the spatial distribution information specifically includes: The unit energy consumption rate of the corresponding sampling interval is calculated based on the differential energy consumption value, and multiple neighboring facilities of each energy-consuming facility are determined based on the spatial distribution information. For any energy-consuming facility, the average neighboring energy consumption rate of the energy-consuming facility in the corresponding sampling interval is determined based on the average of the unit energy consumption rates of the neighboring facilities in the same sampling interval. A first difference between the unit energy consumption rate and the average adjacent energy consumption rate of the energy-consuming facility in each sampling interval is determined, and a second energy consumption deviation amplitude of the energy-consuming facility in the corresponding sampling interval is determined based on the ratio of the first difference to the unit energy consumption rate.
[0010] Furthermore, in one embodiment of the present invention, determining whether each of the energy-consuming facilities has an energy consumption anomaly based on the first energy consumption deviation amplitude and the second energy consumption deviation amplitude specifically includes: Determine whether the absolute value of the first energy consumption deviation amplitude is less than a preset first threshold; When the absolute value of the first energy consumption deviation amplitude is less than the first threshold, it is determined whether the absolute value of the second energy consumption deviation amplitude of the applied energy facility in the corresponding sampling interval is less than the preset second threshold. If yes, it is determined that there is no energy consumption abnormality of the applied energy facility in the corresponding sampling interval. If no, it is determined that there is an energy consumption abnormality of the applied energy facility in the corresponding sampling interval. When the absolute value of the first energy consumption deviation amplitude is greater than or equal to the first threshold, it is determined whether the absolute value of the second energy consumption deviation amplitude of the applied energy facility in the corresponding sampling interval is less than the preset third threshold. If yes, it is determined that there is no energy consumption abnormality of the applied energy facility in the corresponding sampling interval. If no, it is determined that there is an energy consumption abnormality of the applied energy facility in the corresponding sampling interval. Wherein, the second threshold is greater than the third threshold.
[0011] Furthermore, in one embodiment of the present invention, determining whether each of the energy-consuming facilities has an energy consumption anomaly based on the first energy consumption deviation amplitude and the second energy consumption deviation amplitude specifically includes: The spatial relationships of each energy-consuming facility are determined based on the spatial distribution information. Construct an initial graph structure with each of the energy-consuming facilities as nodes based on the spatial relationships; The node timing characteristics corresponding to each energy-consuming facility are determined based on the first energy consumption deviation amplitude, and the edge timing characteristics between each energy-consuming facility and its neighboring facilities are determined based on the second energy consumption deviation amplitude, thus obtaining the timing graph structure data. The time series diagram structure data is input into a pre-trained energy consumption anomaly identification model, and the model output results are used to determine whether each energy-consuming facility has an energy consumption anomaly.
[0012] Furthermore, in one embodiment of the present invention, the energy consumption anomaly identification model is trained through the following steps: Obtain the first energy consumption deviation time series sample of each sample facility within the sample area between adjacent historical sampling intervals, and obtain the second energy consumption deviation time series sample between the sample facility and the neighboring sample facility; A time series diagram structure sample is constructed based on the first energy consumption deviation time series sample, the second energy consumption deviation time series sample, and the spatial relationship between the sample facilities, and the energy consumption anomaly label corresponding to the time series diagram structure sample is determined by manual annotation; The time-series graph structure samples are input into a pre-constructed time-series graph convolutional network to obtain energy consumption anomaly prediction results; The loss value is determined based on the energy consumption anomaly prediction results and the energy consumption anomaly label; The parameters of the temporal graph convolutional network are updated based on the loss value to obtain the trained energy consumption anomaly recognition model.
[0013] Furthermore, in one embodiment of the present invention, the temporal graph convolutional network includes a spatial graph convolutional layer, an LSTM layer, and an output layer. The step of inputting the temporal graph structure sample into the pre-constructed temporal graph convolutional network to obtain the energy consumption anomaly prediction result specifically includes: The spatial features of the temporal graph structure sample at each time step are extracted by the spatial graph convolutional layer to obtain the temporal spatial features; The hidden state vector is obtained by calculating the hidden state of the temporal spatial features through the LSTM layer. The hidden state vector is mapped to the energy consumption anomaly prediction result through the output layer.
[0014] On the other hand, embodiments of the present invention provide an energy consumption anomaly detection device based on spatiotemporal feature fusion, comprising: The data acquisition module is used to acquire time-series energy consumption data of multiple energy-consuming facilities within a target area during a preset time period, and to acquire spatial distribution information of each energy-consuming facility. The first energy consumption deviation determination module is used to determine the differential energy consumption value of each energy-consuming facility in each sampling interval based on the time-series energy consumption data, and to calculate the first energy consumption deviation amplitude of each energy-consuming facility in each sampling interval and the previous sampling interval based on the differential energy consumption value. The second energy consumption deviation determination module is used to determine the second energy consumption deviation amplitude between each of the energy-consuming facilities and the adjacent facilities in each sampling interval based on the differential energy consumption value and the spatial distribution information; An energy consumption anomaly detection module is used to determine whether each of the energy-consuming facilities has an energy consumption anomaly based on the first energy consumption deviation amplitude and the second energy consumption deviation amplitude.
[0015] On the other hand, embodiments of the present invention provide an electronic device, the electronic device including a memory, a processor, a computer program stored in the memory and executable on the processor, and a data bus for implementing communication between the processor and the memory, wherein when the computer program is executed by the processor, it implements the energy consumption anomaly detection method based on spatiotemporal feature fusion as described above.
[0016] On the other hand, embodiments of the present invention also provide a storage medium, which is a computer-readable storage medium for computer-readable storage. The storage medium stores one or more computer programs, which can be executed by one or more processors to implement the energy consumption anomaly detection method based on spatiotemporal feature fusion as described above.
[0017] On the other hand, embodiments of the present invention also provide a computer program product, including a computer program that, when executed by a processor, implements the energy consumption anomaly detection method based on spatiotemporal feature fusion as described above.
[0018] The advantages and beneficial effects of the present invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention: This invention acquires time-series energy consumption data of multiple energy-consuming facilities within a target area over a preset period, and obtains spatial distribution information of each facility. Based on the time-series energy consumption data, it determines the differential energy consumption value of each facility at each sampling interval. Based on the differential energy consumption value, it calculates the first energy consumption deviation amplitude of each facility between each sampling interval and the previous sampling interval. Based on the differential energy consumption value and spatial distribution information, it determines the second energy consumption deviation amplitude of each facility between each sampling interval and neighboring facilities. Based on the first and second energy consumption deviation amplitudes, it determines whether each facility has an energy consumption anomaly. This invention analyzes the energy consumption deviation of facilities in adjacent sampling intervals and the energy consumption deviation with neighboring facilities in the same sampling interval, using environmental factors of the area where the facility is located as a basis for judgment. This allows for a comprehensive capture of energy consumption changes in both time and space dimensions, avoiding potential errors in judging energy consumption anomalies based on the energy consumption data of a single facility, thereby improving the accuracy of energy consumption anomaly detection. Attached Figure Description
[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the embodiments of the present invention are described below. It should be understood that the drawings described below are only for the convenience of clearly describing some embodiments of the technical solutions of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0020] Figure 1 A flowchart illustrating the steps of an energy consumption anomaly detection method based on spatiotemporal feature fusion provided in an embodiment of the present invention; Figure 2 A flowchart of step S102 provided in an embodiment of the present invention; Figure 3 A flowchart of step S103 provided in an embodiment of the present invention; Figure 4 A flowchart of step S104 provided in an embodiment of the present invention; Figure 5 Another flowchart for step S104 provided in an embodiment of the present invention; Figure 6 A flowchart of a training energy consumption anomaly identification model provided in an embodiment of the present invention; Figure 7 A flowchart of step S203 provided in an embodiment of the present invention; Figure 8 A schematic diagram of the energy consumption anomaly detection device based on spatiotemporal feature fusion provided in an embodiment of the present invention; Figure 9A schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present invention; Figure 10 This is a schematic diagram of the structure of the storage medium provided in an embodiment of the present invention. Detailed Implementation
[0021] The embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this application, and should not be construed as limiting this application. It should be noted that although functional modules are divided in the system schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the system schematic diagram or the order in the flowchart. The step numbers in the following embodiments are only set for ease of explanation and do not limit the order between steps. The execution order of each step in the embodiments can be adaptively adjusted according to the understanding of those skilled in the art.
[0022] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.
[0023] The following describes some of the terms used in the embodiments of the present invention.
[0024] Time-series energy consumption data: Time-series energy consumption data generally refers to the cumulative energy consumption data reported by energy-consuming facilities at certain time intervals. Specific energy consumption types may include electricity consumption, gas consumption, etc. This data has a clear time sequence.
[0025] Temporal Graph Convolutional Network (TGCN) is a deep learning model that extends Graph Convolutional Networks (GCNs) to temporal scenarios. It is specifically designed to handle data that has both spatial graph structure and temporally dynamic features. It models both the spatial dependencies between graph nodes and the temporal dependencies of node features as they evolve over time, overcoming the limitation of traditional static graph convolutions, which can only handle data with fixed structures.
[0026] The energy consumption anomaly detection method based on spatiotemporal feature fusion provided in this application can be applied to terminals, servers, or software running on terminals or servers. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, smart speaker, smartwatch, or vehicle terminal, but is not limited to these. The server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The server can also be a node server in a blockchain network. The software can be an application that implements the above method, but is not limited to the above forms.
[0027] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.
[0028] It should be noted that in all specific embodiments of this application, when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent is obtained first. Furthermore, the collection, use, and processing of this data comply with relevant laws, regulations, and standards of the relevant countries and regions. In addition, when embodiments of this application require access to sensitive personal information of users, separate permission or consent from the user is obtained through pop-ups or redirects to confirmation pages. Only after obtaining the user's separate permission or consent is the necessary user-related data for the proper functioning of the embodiments of this application obtained.
[0029] like Figure 1 The diagram shows a flowchart of a method for detecting energy consumption anomalies based on spatiotemporal feature fusion provided in an embodiment of the present invention. (Refer to...) Figure 1This invention provides a method for detecting energy consumption anomalies based on spatiotemporal feature fusion, specifically including the following steps: S101. Obtain time-series energy consumption data of multiple energy-consuming facilities within the target area during a preset time period, and obtain spatial distribution information of each energy-consuming facility; S102. Determine the differential energy consumption value of each energy-consuming facility in each sampling interval based on the time-series energy consumption data, and calculate the first energy consumption deviation amplitude of each energy-consuming facility in each sampling interval from the previous sampling interval based on the differential energy consumption value. S103. Determine the second energy consumption deviation amplitude between each energy-consuming facility and its neighboring facilities at each sampling interval based on the differential energy consumption value and spatial distribution information; S104. Determine whether there is any energy consumption abnormality in each energy-consuming facility based on the first energy consumption deviation range and the second energy consumption deviation range.
[0030] It should be noted that the energy-consuming facilities in the embodiments of the present invention can be individual energy-consuming devices or buildings with multiple energy-consuming devices. That is, the energy consumption anomaly detection method in the embodiments of the present invention can be applied to the detection of energy consumption anomalies at the level of energy-consuming devices, or to the detection of overall energy consumption anomalies at the level of buildings. In addition, energy consumption in the embodiments of the present invention can refer to the consumption of various types of energy, such as electricity, gas, etc., and is not limited thereto.
[0031] This invention analyzes the energy consumption deviation of energy-consuming facilities in adjacent sampling intervals and the energy consumption deviation with neighboring facilities in the same sampling interval. By using environmental factors of the area where the energy-consuming facility is located as a basis for judgment, it can comprehensively capture the energy consumption changes of energy-consuming facilities in both time and space dimensions. This avoids the potential errors in judging energy consumption anomalies based on the energy consumption data of a single energy-consuming facility, thereby improving the accuracy of energy consumption anomaly detection. Taking the electrical energy consumed by a lamp as an example, if the energy consumption of the target lamp is generally stable over a specified period, considering only the lamp's own energy consumption time-series characteristics, it might be judged that the lamp's energy consumption is normal. However, by comparing it with multiple neighboring lamps, if the lamp's energy consumption rate differs significantly from that of neighboring lamps at a certain time period (e.g., the lamp sensor is triggered for a long time due to the blower, resulting in high energy consumption), the energy consumption anomaly detection method based on spatiotemporal feature fusion of this invention can accurately identify and judge the lamp's energy consumption anomaly. Compared with traditional methods, energy consumption anomaly detection is more accurate.
[0032] like Figure 2 The diagram shown is a flowchart of step S102 provided in an embodiment of the present invention. (Refer to...) Figure 2 As an optional implementation, the differential energy consumption value of each energy-consuming facility in each sampling interval is determined based on the time-series energy consumption data, and the first energy consumption deviation amplitude of each energy-consuming facility in each sampling interval compared with the previous sampling interval is calculated based on the differential energy consumption value, specifically including: S1021. Determine the cumulative energy consumption value of each energy-consuming facility at each sampling time based on the time-series energy consumption data; S1022. Perform differential calculation on the cumulative energy consumption value of the energy-consuming facility at adjacent sampling times to obtain the differential energy consumption value for the corresponding sampling interval; S1023. Calculate the unit energy consumption rate of the corresponding sampling interval based on the differential energy consumption value; S1024. Calculate the energy consumption rate growth rate of the energy-consuming facility in adjacent sampling intervals based on the unit energy consumption rate to obtain the first energy consumption deviation amplitude.
[0033] Specifically, access to energy facilities Energy consumption data during the preset time period The timing format is as follows: , in, express Energy-consuming facilities The cumulative energy consumption value.
[0034] Energy consumption facilities are calculated using differential calculations. Differential energy consumption at adjacent sampling times The timing format is as follows:
[0035] Considering that the time intervals for smart meters to report energy consumption are not entirely consistent, energy-consuming facilities are calculated individually. Unit energy consumption rate in each sampling interval Then, it is compared with the unit energy consumption rate of the previous sampling interval to calculate the energy consumption rate growth rate. as follows:
[0036] Ultimately, this can be calculated based on the energy consumption rate growth rate. A first energy consumption deviation range is determined, which reflects the energy consumption deviation of the energy-consuming facility in adjacent sampling intervals.
[0037] like Figure 3 The diagram shown is a flowchart of step S103 provided in an embodiment of the present invention. (Refer to...) Figure 3 As an optional implementation, the second energy consumption deviation amplitude between each energy-consuming facility and its neighboring facilities at each sampling interval is determined based on the differential energy consumption value and spatial distribution information, specifically including: S1031. Calculate the unit energy consumption rate of the corresponding sampling interval based on the differential energy consumption value, and determine multiple neighboring facilities of each energy-consuming facility based on the spatial distribution information. S1032. For any energy-consuming facility, determine the average neighboring energy consumption rate of the energy-consuming facility in the corresponding sampling interval based on the average unit energy consumption rate of the neighboring facilities in the same sampling interval. S1033. Determine the first difference between the unit energy consumption rate of the energy-consuming facility and the average adjacent energy consumption rate in each sampling interval, and determine the second energy consumption deviation amplitude of the energy-consuming facility in the corresponding sampling interval based on the ratio of the first difference to the unit energy consumption rate.
[0038] Specifically, spatial distribution information of each energy-consuming facility is obtained, and other related energy-consuming facilities are determined based on the spatial distribution information. These facilities are then arranged in order of increasing distance from the current energy-consuming facility or decreasing degree of association, and finally, the top-ranked facilities are selected as neighboring facilities.
[0039] For any energy-consuming facility, the average neighboring energy consumption rate of that facility in the corresponding sampling interval is first determined based on the average unit energy consumption rate of its neighboring facilities in the same sampling interval. Then calculate the unit energy consumption rate of the energy-consuming facility in the corresponding sampling interval. The second energy consumption deviation amplitude corresponding to the sampling interval can be obtained by dividing the first difference between the average adjacent energy consumption rate and the unit energy consumption rate. The timing format is as follows:
[0040] in, express and The sampling interval between them.
[0041] The embodiments of the present invention can realize energy consumption anomaly detection through two methods: threshold judgment and model recognition. The two implementation methods are described below.
[0042] like Figure 4 The diagram shown is a flowchart of step S104 provided in an embodiment of the present invention. (Refer to...) Figure 4 As an optional implementation, it is further determined whether each energy-consuming facility has abnormal energy consumption based on the first energy consumption deviation range and the second energy consumption deviation range, specifically including: S1041. Determine whether the absolute value of the first energy consumption deviation amplitude is less than the preset first threshold. S1042. When the absolute value of the first energy consumption deviation amplitude is less than the first threshold, determine whether the absolute value of the second energy consumption deviation amplitude of the applied energy facility in the corresponding sampling interval is less than the preset second threshold. If yes, determine that there is no energy consumption abnormality of the applied energy facility in the corresponding sampling interval. If no, determine that there is an energy consumption abnormality of the applied energy facility in the corresponding sampling interval. S1043. When the absolute value of the first energy consumption deviation amplitude is greater than or equal to the first threshold, determine whether the absolute value of the second energy consumption deviation amplitude of the energy application facility in the corresponding sampling interval is less than the preset third threshold. If yes, determine that there is no energy consumption abnormality of the energy application facility in the corresponding sampling interval. If no, determine that there is an energy consumption abnormality of the energy application facility in the corresponding sampling interval. The second threshold is greater than the third threshold.
[0043] Specifically, in this embodiment of the invention, anomaly detection of energy consumption is performed by threshold judgment.
[0044] When the absolute value of the first energy consumption deviation amplitude is greater than or equal to the first threshold, it indicates that the time-series energy consumption characteristics of the energy-consuming facility are abnormal. If the absolute value of the second energy consumption deviation amplitude between the facility and the adjacent facility in the same period is greater than or equal to the third threshold, it indicates that the energy-consuming facility also has abnormal spatial energy consumption characteristics. It can be determined that the energy-consuming facility has an energy consumption abnormality in the corresponding period. If the absolute value of the second energy consumption deviation amplitude between the facility and the adjacent facility in the same period is less than the third threshold, it indicates that the energy consumption fluctuation of the energy-consuming facility is a collective phenomenon and the energy-consuming facilities are relatively consistent. It can be determined that the energy-consuming facility does not have an energy consumption abnormality in the corresponding period.
[0045] When the absolute value of the first energy consumption deviation amplitude is less than the first threshold, it indicates that the time-series energy consumption characteristics of the energy-consuming facility are not abnormal (no obvious fluctuations). If the absolute value of the second energy consumption deviation amplitude between the facility and the adjacent facility in the same period is greater than or equal to the second threshold (the second threshold is greater than the third threshold), it indicates that the energy-consuming facility has significant abnormalities in spatial energy consumption characteristics, and it can be determined that the energy-consuming facility has energy consumption abnormalities in the corresponding period. If the absolute value of the second energy consumption deviation amplitude between the facility and the adjacent facility in the same period is less than the second threshold, it can be determined that the energy-consuming facility does not have energy consumption abnormalities in the corresponding period.
[0046] This invention, through the design of a second threshold and a third threshold, uses a larger second threshold to determine whether there is an abnormality in the energy consumption deviation between the power-consuming facility and neighboring facilities when the energy consumption of the power-consuming facility itself does not fluctuate significantly, and uses a smaller third threshold to determine whether there is an abnormality in the energy consumption deviation between the power-consuming facility and neighboring facilities when the energy consumption of the power-consuming facility itself fluctuates significantly. This fully utilizes the energy consumption changes of the power-consuming facility in the time and space dimensions, and improves the accuracy of energy consumption anomaly detection.
[0047] like Figure 5 The diagram shown is another flowchart of step S104 provided in an embodiment of the present invention. (Refer to...) Figure 5 As an optional implementation, it is further determined whether each energy-consuming facility has abnormal energy consumption based on the first energy consumption deviation range and the second energy consumption deviation range, specifically including: S1044. Determine the spatial relationship between various energy-consuming facilities based on spatial distribution information; S1045. Construct an initial graph structure with each energy-consuming facility as a node based on spatial relationships; S1046. Determine the node time sequence characteristics corresponding to each energy-consuming facility based on the first energy consumption deviation amplitude, and determine the edge time sequence characteristics between each energy-consuming facility and its neighboring facilities based on the second energy consumption deviation amplitude, to obtain the time sequence diagram structure data. S1047. Input the time series diagram structure data into the pre-trained energy consumption anomaly identification model, and determine whether there is an energy consumption anomaly in each energy-consuming facility based on the model output results.
[0048] Specifically, in this embodiment of the invention, energy consumption anomaly detection is performed through model recognition.
[0049] First, the spatial relationships of each energy-consuming facility are determined based on spatial distribution information, and the neighboring facilities of each energy-consuming facility are identified. An initial graph structure is constructed based on these spatial relationships, with each energy-consuming facility serving as a node, and two nodes that are neighboring facilities are designated as adjacent nodes. The temporal characteristics of the nodes corresponding to each energy-consuming facility are determined based on the first energy consumption deviation amplitude, and the edge temporal characteristics between each energy-consuming facility and its neighboring facilities are determined based on the second energy consumption deviation amplitude, resulting in temporal graph structure data. This temporal graph structure data is then input into a pre-trained energy consumption anomaly identification model. This model captures the energy consumption changes of each node in both the time and spatial dimensions, and finally, the model output results are used to determine whether any energy consumption anomalies exist in each energy-consuming facility.
[0050] like Figure 6 The diagram shown is a flowchart of a training energy consumption anomaly identification model provided in an embodiment of the present invention. (Refer to...) Figure 6 As an optional implementation, the energy consumption anomaly identification model is trained through the following steps: S201. Obtain the first energy consumption deviation time series sample of each sample facility within the sample area between adjacent historical sampling intervals, and obtain the second energy consumption deviation time series sample between the sample facility and the neighboring sample facility. S202. Construct a time series diagram structure sample based on the first energy consumption deviation time series sample, the second energy consumption deviation time series sample, and the spatial relationship between the sample facilities, and determine the energy consumption anomaly label corresponding to the time series diagram structure sample through manual annotation. S203. Input the temporal graph structure samples into the pre-constructed temporal graph convolutional network to obtain the energy consumption anomaly prediction results; S204. Determine the loss value based on the energy consumption anomaly prediction results and energy consumption anomaly labels; S205. Update the parameters of the temporal graph convolutional network according to the loss value to obtain the trained energy consumption anomaly recognition model.
[0051] Specifically, the first energy consumption deviation time-series sample of each sample facility within the sample area between adjacent historical sampling intervals is obtained, and the second energy consumption deviation time-series sample sample between the sample facility and its neighboring sample facilities is obtained. The calculation process of the energy consumption deviation time-series sample is similar to the aforementioned calculation process of the energy consumption deviation amplitude, and will not be repeated here. A time-series graph structure sample is constructed based on the first energy consumption deviation time-series sample, the second energy consumption deviation time-series sample, and the spatial relationship between the sample facilities. The energy consumption anomaly label corresponding to the time-series graph structure sample is determined by manual annotation. The construction process of the time-series graph structure sample is similar to the aforementioned construction process of the time-series graph structure data, and will not be repeated here. The time-series graph structure sample is input into a pre-constructed time-series graph convolutional network to obtain the energy consumption anomaly prediction result. The loss value is determined based on the energy consumption anomaly prediction result and the energy consumption anomaly label. The parameters of the time-series graph convolutional network are updated based on the loss value to complete one iteration of training. When the number of iterations reaches a preset threshold, or the loss value is lower than the preset threshold, training is stopped, and the trained energy consumption anomaly recognition model is obtained.
[0052] like Figure 7 The diagram shown is a flowchart of step S203 provided in an embodiment of the present invention. (Refer to...) Figure 7 As an optional implementation, the temporal graph convolutional network includes spatial graph convolutional layers, LSTM layers, and an output layer. Temporal graph structure samples are input into the pre-constructed temporal graph convolutional network to obtain energy consumption anomaly prediction results, specifically including: S2031. Extract the spatial features of the temporal graph structure samples at each time step through the spatial graph convolutional layer to obtain the temporal spatial features; S2032. Hidden state vectors are obtained by calculating the hidden state of the temporal spatial features through LSTM layers. S2033. The hidden state vector is mapped to the energy consumption anomaly prediction result through the output layer.
[0053] Specifically, the temporal graph convolutional network of this embodiment includes an input layer, a spatial graph convolutional layer, an LSTM layer, and an output layer. The input layer is used to input the normalized temporal graph structure samples, including an adjacency matrix describing the graph structure connectivity at the current time step, as well as node feature matrices and edge feature matrices. The spatial graph convolutional layer extracts spatial features from the graph structure data at each time step, which can be implemented using Chebyshev Graph Convolution (Chebyshev GCN). The LSTM is used to capture the temporal dependence of the spatial feature sequence, forming a hidden state vector. The output layer maps the hidden state vector to an energy consumption anomaly prediction results.
[0054] After training the energy consumption anomaly identification model through the above steps, the time series graph structure data of the target area is constructed based on the first energy consumption deviation amplitude and the second energy consumption deviation amplitude, and then input into the energy consumption anomaly identification model. Finally, the model output results are used to determine whether each energy-consuming facility has energy consumption anomalies, the time period of energy consumption anomalies, and the corresponding confidence level.
[0055] The implementation process of the present invention has been described above. It is understood that the embodiments of the present invention, by analyzing the energy consumption deviation of energy-consuming facilities in adjacent sampling intervals and the energy consumption deviation with neighboring facilities in the same sampling interval, and using environmental factors of the area where the energy-consuming facility is located as the basis for judgment, can comprehensively capture the energy consumption changes of energy-consuming facilities in both time and space dimensions. This avoids potential errors in judging energy consumption anomalies based on the energy consumption data of a single energy-consuming facility, thereby improving the accuracy of energy consumption anomaly detection.
[0056] like Figure 8 The diagram shown is a structural schematic of the energy consumption anomaly detection device based on spatiotemporal feature fusion provided in an embodiment of the present invention. (Refer to...) Figure 8 This invention provides an energy consumption anomaly detection device based on spatiotemporal feature fusion, comprising: The data acquisition module is used to acquire time-series energy consumption data of multiple energy-consuming facilities within the target area during a preset time period, and to acquire spatial distribution information of each energy-consuming facility; The first energy consumption deviation determination module is used to determine the differential energy consumption value of each energy-consuming facility in each sampling interval based on the time-series energy consumption data, and to calculate the first energy consumption deviation amplitude of each energy-consuming facility in each sampling interval and the previous sampling interval based on the differential energy consumption value. The second energy consumption deviation determination module is used to determine the magnitude of the second energy consumption deviation between each energy-consuming facility and its neighboring facilities at each sampling interval based on the differential energy consumption value and spatial distribution information. The energy consumption anomaly detection module is used to determine whether there is an energy consumption anomaly in each energy-consuming facility based on the first energy consumption deviation range and the second energy consumption deviation range.
[0057] It is understood that the content of the above method embodiments is applicable to the present device embodiments. The specific functions implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.
[0058] This invention also provides an electronic device, comprising: a memory, a processor, a program stored in the memory and executable on the processor, and a data bus for communication between the processor and the memory. When the program is executed by the processor, it implements the aforementioned energy consumption anomaly detection method based on spatiotemporal feature fusion. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.
[0059] like Figure 9 The diagram shown is a hardware structure schematic of an electronic device provided in an embodiment of the present invention. (Refer to...) Figure 9 This invention provides an electronic device, comprising: The processor 901 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present invention. The memory 902 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 902 can store the operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 902 and is called and executed by the processor 901 to implement the energy consumption anomaly detection method based on spatiotemporal feature fusion of the embodiments of this invention. The 903 input / output interface is used to implement information input and output. The communication interface 904 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.). Bus 905 transmits information between various components of the device (e.g., processor 901, memory 902, input / output interface 903, and communication interface 904); The processor 901, memory 902, input / output interface 903, and communication interface 904 are connected to each other within the device via bus 905.
[0060] It is understood that the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.
[0061] like Figure 10 The diagram shown is a structural schematic of the storage medium provided in an embodiment of the present invention. (Refer to...) Figure 10The present invention also provides a storage medium, which is a computer-readable storage medium for computer-readable storage. The storage medium stores one or more programs 1001, which can be executed by one or more processors to implement the above-mentioned energy consumption anomaly detection method based on spatiotemporal feature fusion.
[0062] It is understood that the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.
[0063] This invention also discloses a computer program product, including a computer program that, when executed by a processor, implements the above-described energy consumption anomaly detection method based on spatiotemporal feature fusion.
[0064] It is understood that the content of the above method embodiments is applicable to the embodiments of this program product. The specific functions implemented by the embodiments of this program product are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.
[0065] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.
[0066] The embodiments described in this invention are for the purpose of more clearly illustrating the technical solutions of the embodiments of this invention, and do not constitute a limitation on the technical solutions provided by the embodiments of this invention. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this invention are also applicable to similar technical problems.
[0067] The terms "first," "second," "third," "fourth," etc. (if present) in the specification and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.
[0068] In some alternative embodiments, the functions / operations mentioned in the block diagrams may not occur in the order shown in the operation diagrams. For example, depending on the functions / operations involved, two consecutively shown blocks may actually be executed substantially simultaneously, or the aforementioned blocks may sometimes be executed in reverse order. Furthermore, the embodiments presented and described in the flowcharts of this invention are provided by way of example to provide a more comprehensive understanding of the technology. The disclosed methods are not limited to the operations and logic flows presented herein. Alternative embodiments are contemplated in which the order of various operations is changed and sub-operations described as part of a larger operation are executed independently.
[0069] Furthermore, although the invention has been described in the context of functional modules, it should be understood that, unless otherwise stated, one or more of the aforementioned functions and / or features may be integrated into a single physical device and / or software module, or one or more functions and / or features may be implemented in a separate physical device or software module. It is also understood that a detailed discussion of the actual implementation of each module is unnecessary for understanding the invention. Rather, given the properties, functions, and internal relationships of the various functional modules in the apparatus disclosed herein, the actual implementation of the module will be understood within the scope of conventional skill of an engineer. Therefore, those skilled in the art can implement the invention as set forth in the claims using ordinary techniques without excessive experimentation. It is also understood that the specific concepts disclosed are merely illustrative and not intended to limit the scope of the invention, which is determined by the full scope of the appended claims and their equivalents.
[0070] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a portion 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 this 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.
[0071] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.
[0072] More specific examples (a non-exhaustive list) of computer-readable media include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the aforementioned program can be printed, because the aforementioned program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.
[0073] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.
[0074] In the foregoing description of this specification, references to terms such as "one embodiment," "another embodiment," or "some embodiments" indicate that a specific feature, structure, material, or characteristic described in connection with an embodiment or example is included in at least one embodiment or example of the present invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.
[0075] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.
[0076] The above is a detailed description of the preferred embodiments of the present invention. However, the present invention is not limited to the above embodiments. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present invention. All such equivalent modifications or substitutions are included within the scope defined by the claims of this application.
Claims
1. A method for detecting energy consumption anomalies based on spatiotemporal feature fusion, characterized in that, Includes the following steps: Acquire time-series energy consumption data of multiple energy-consuming facilities within a target area during a preset time period, and acquire spatial distribution information of each energy-consuming facility; The differential energy consumption value of each energy-consuming facility in each sampling interval is determined based on the time-series energy consumption data, and the first energy consumption deviation amplitude of each energy-consuming facility in each sampling interval and the previous sampling interval is calculated based on the differential energy consumption value. The second energy consumption deviation amplitude between each energy-consuming facility and its neighboring facilities is determined based on the differential energy consumption value and the spatial distribution information at each sampling interval; The energy consumption of each energy-consuming facility is determined to be abnormal based on the first energy consumption deviation range and the second energy consumption deviation range.
2. The energy consumption anomaly detection method based on spatiotemporal feature fusion according to claim 1, characterized in that, The step of determining the differential energy consumption value of each energy-consuming facility in each sampling interval based on the time-series energy consumption data, and calculating the first energy consumption deviation amplitude of each energy-consuming facility in each sampling interval compared with the previous sampling interval based on the differential energy consumption value, specifically includes: The cumulative energy consumption value of each energy-consuming facility at each sampling time is determined based on the time-series energy consumption data; The cumulative energy consumption value of the energy-consuming facility at adjacent sampling times is calculated differentially to obtain the differential energy consumption value for the corresponding sampling interval; Calculate the unit energy consumption rate of the corresponding sampling interval based on the differential energy consumption value; The energy consumption rate growth rate of the energy-consuming facility in adjacent sampling intervals is calculated based on the unit energy consumption rate to obtain the first energy consumption deviation amplitude.
3. The energy consumption anomaly detection method based on spatiotemporal feature fusion according to claim 1, characterized in that, The step of determining the second energy consumption deviation amplitude between each energy-consuming facility and its neighboring facilities at each sampling interval based on the differential energy consumption value and the spatial distribution information specifically includes: The unit energy consumption rate of the corresponding sampling interval is calculated based on the differential energy consumption value, and multiple neighboring facilities of each energy-consuming facility are determined based on the spatial distribution information. For any energy-consuming facility, the average neighboring energy consumption rate of the energy-consuming facility in the corresponding sampling interval is determined based on the average of the unit energy consumption rates of the neighboring facilities in the same sampling interval. A first difference between the unit energy consumption rate and the average adjacent energy consumption rate of the energy-consuming facility in each sampling interval is determined, and a second energy consumption deviation amplitude of the energy-consuming facility in the corresponding sampling interval is determined based on the ratio of the first difference to the unit energy consumption rate.
4. A method for detecting energy consumption anomalies based on spatiotemporal feature fusion according to any one of claims 1 to 3, characterized in that, The step of determining whether each energy-consuming facility has an energy consumption anomaly based on the first energy consumption deviation amplitude and the second energy consumption deviation amplitude specifically includes: Determine whether the absolute value of the first energy consumption deviation amplitude is less than a preset first threshold; When the absolute value of the first energy consumption deviation amplitude is less than the first threshold, it is determined whether the absolute value of the second energy consumption deviation amplitude of the applied energy facility in the corresponding sampling interval is less than the preset second threshold. If yes, it is determined that there is no energy consumption abnormality of the applied energy facility in the corresponding sampling interval. If no, it is determined that there is an energy consumption abnormality of the applied energy facility in the corresponding sampling interval. When the absolute value of the first energy consumption deviation amplitude is greater than or equal to the first threshold, it is determined whether the absolute value of the second energy consumption deviation amplitude of the applied energy facility in the corresponding sampling interval is less than the preset third threshold. If yes, it is determined that there is no energy consumption abnormality of the applied energy facility in the corresponding sampling interval. If no, it is determined that there is an energy consumption abnormality of the applied energy facility in the corresponding sampling interval. Wherein, the second threshold is greater than the third threshold.
5. A method for detecting energy consumption anomalies based on spatiotemporal feature fusion according to any one of claims 1 to 3, characterized in that, The step of determining whether each energy-consuming facility has an energy consumption anomaly based on the first energy consumption deviation amplitude and the second energy consumption deviation amplitude specifically includes: The spatial relationships of each energy-consuming facility are determined based on the spatial distribution information. Construct an initial graph structure with each of the energy-consuming facilities as nodes based on the spatial relationships; The node timing characteristics corresponding to each energy-consuming facility are determined based on the first energy consumption deviation amplitude, and the edge timing characteristics between each energy-consuming facility and its neighboring facilities are determined based on the second energy consumption deviation amplitude, thus obtaining the timing graph structure data. The time series diagram structure data is input into a pre-trained energy consumption anomaly identification model, and the model output results are used to determine whether each energy-consuming facility has an energy consumption anomaly.
6. The energy consumption anomaly detection method based on spatiotemporal feature fusion according to claim 5, characterized in that, The energy consumption anomaly identification model is trained through the following steps: Obtain the first energy consumption deviation time series sample of each sample facility within the sample area between adjacent historical sampling intervals, and obtain the second energy consumption deviation time series sample between the sample facility and the neighboring sample facility; A time series diagram structure sample is constructed based on the first energy consumption deviation time series sample, the second energy consumption deviation time series sample, and the spatial relationship between the sample facilities, and the energy consumption anomaly label corresponding to the time series diagram structure sample is determined by manual annotation; The time-series graph structure samples are input into a pre-constructed time-series graph convolutional network to obtain energy consumption anomaly prediction results; The loss value is determined based on the energy consumption anomaly prediction results and the energy consumption anomaly label; The parameters of the temporal graph convolutional network are updated based on the loss value to obtain the trained energy consumption anomaly recognition model.
7. The energy consumption anomaly detection method based on spatiotemporal feature fusion according to claim 6, characterized in that, The temporal graph convolutional network includes spatial graph convolutional layers, LSTM layers, and an output layer. The step of inputting the temporal graph structure samples into the pre-constructed temporal graph convolutional network to obtain energy consumption anomaly prediction results specifically includes: The spatial features of the temporal graph structure sample at each time step are extracted by the spatial graph convolutional layer to obtain the temporal spatial features; The hidden state vector is obtained by calculating the hidden state of the temporal spatial features through the LSTM layer. The hidden state vector is mapped to the energy consumption anomaly prediction result through the output layer.
8. An energy consumption anomaly detection device based on spatiotemporal feature fusion, characterized in that, include: The data acquisition module is used to acquire time-series energy consumption data of multiple energy-consuming facilities within a target area during a preset time period, and to acquire spatial distribution information of each energy-consuming facility. The first energy consumption deviation determination module is used to determine the differential energy consumption value of each energy-consuming facility in each sampling interval based on the time-series energy consumption data, and to calculate the first energy consumption deviation amplitude of each energy-consuming facility in each sampling interval and the previous sampling interval based on the differential energy consumption value. The second energy consumption deviation determination module is used to determine the second energy consumption deviation amplitude between each of the energy-consuming facilities and the adjacent facilities in each sampling interval based on the differential energy consumption value and the spatial distribution information; An energy consumption anomaly detection module is used to determine whether each of the energy-consuming facilities has an energy consumption anomaly based on the first energy consumption deviation amplitude and the second energy consumption deviation amplitude.
9. An electronic device, characterized in that, The electronic device includes a memory, a processor, a computer program stored in the memory and executable on the processor, and a data bus for enabling communication between the processor and the memory. When the computer program is executed by the processor, it implements the energy consumption anomaly detection method based on spatiotemporal feature fusion as described in any one of claims 1 to 7.
10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the energy consumption anomaly detection method based on spatiotemporal feature fusion as described in any one of claims 1 to 7.
