Method, system, device and equipment for identifying dangerous power consumption behavior and medium

Abstract

The invention discloses a dangerous power consumption behavior identification method, system, device and equipment and a medium. The identification method comprises the following steps: monitoring whether a dangerous electrical load enters the area, and when the dangerous electrical load enters the area, determining the area attribution of the dangerous electrical load; for the dangerous electrical load belonging to the area, monitoring whether the dangerous electrical load enters the shed for charging; determining the charging state of the dangerous electrical load entering the shed for charging; for the dangerous electrical load which does not enter the shed for charging, whether the dangerous electrical load enters the building electronic fence or not is determined; for the dangerous electrical load which does not enter the building electronic fence, whether the dangerous electrical load enters the elevator or not is determined according to the monitoring result; and for the dangerous electrical load which does not enter the elevator, whether the dangerous electrical load enters a home to be charged or not is determined according to the electrical characteristics. The recognition system and device, the electronic equipment and the computer readable storage medium all realize the recognition method.

Description

Technical Field

[0001] The present invention relates to the technical field of electricity consumption identification, and particularly to a method, system, device, equipment and medium for identifying dangerous electricity consumption behaviors. Background Art

[0002] In recent years, as an environmentally friendly and convenient means of transportation, electric bicycles have been widely popularized and applied globally. However, with the rapid increase in the number of electric bicycles, electrical fire accidents caused by their charging batteries have occurred frequently, posing significant safety hazards to society. Such fire accidents are often closely related to the load characteristics of the batteries and the non-standard operation of the equipment. If not effectively prevented and controlled, they will pose a serious threat to the lives and property safety of residents.

[0003] In the existing technical system, there are deficiencies in the monitoring and management of electric bicycle charging equipment. The missed detection rate of single equipment monitoring is high, and relying on manual inspections or static threshold alarms is prone to lag in response, with poor effects and inability to adapt to complex scenarios, resulting in some potential safety hazards not being discovered and processed in a timely manner. Summary of the Invention

[0004] The purpose of the present invention is to provide a method, system, device, equipment and medium for identifying dangerous electricity consumption behaviors to solve at least one of the problems such as high missed detection rate and lag in response mentioned in the above background art.

[0005] To achieve the above purpose, the present invention adopts the following technical solutions: In the first aspect of the present invention, a method for identifying dangerous electricity consumption behaviors is provided, including: Monitoring whether there is a dangerous electricity consumption load entering this area. When a dangerous electricity consumption load enters this area, determining the area attribution of the dangerous electricity consumption load; for a dangerous electricity consumption load that does not belong to this area, generating an alarm message for rejecting entry; For a dangerous electricity consumption load that belongs to this area, monitoring whether the dangerous electricity consumption load enters the shed for charging; for a dangerous electricity consumption load that enters the shed for charging, determining the charging state of the dangerous electricity consumption load. When the charging state is abnormal, generating a corresponding alarm message; For a dangerous electricity consumption load that does not enter the shed for charging, determining whether the dangerous electricity consumption load enters the building electronic fence; for a dangerous electricity consumption load that enters the building electronic fence, generating a corresponding alarm message; For a dangerous electricity consumption load that does not enter the building electronic fence, obtaining the monitoring result of the elevator AI camera for the dangerous electricity consumption load, and determining whether the dangerous electricity consumption load enters the elevator according to the monitoring result; for a dangerous electricity consumption load that enters the elevator, generating a corresponding alarm message; For dangerous electrical loads that do not enter the elevator, obtain the electrical usage characteristics of the user identified and analyzed by the intelligent electricity meter, and determine whether the dangerous electrical load enters the household for charging according to the electrical usage characteristics; for dangerous electrical loads that enter the household for charging, generate corresponding warning information.

[0006] Optionally, determining the regional attribution of the dangerous electrical load includes: Obtain the recognition result of the dangerous electrical load by the entrance access control system of this area, and determine the electronic tag information of the dangerous electrical load according to the recognition result; Match the electronic tag information with a pre-set automated management ledger; For dangerous electrical loads whose electronic tag information does not match the automated management ledger, determine that the dangerous electrical load does not belong to this area; for dangerous electrical loads whose electronic tag information matches the automated management ledger, determine that the dangerous electrical load belongs to this area.

[0007] Optionally, determining the charging status of the dangerous electrical load includes: Obtain the charging data and environmental data of the dangerous electrical load; input the charging data and environmental data into a pre-trained long short-term memory network health assessment model, and the long short-term memory network health assessment model outputs the health status assessment result of the dangerous electrical load; Determine whether the health condition of the dangerous electrical load is normal according to the health status assessment result; For dangerous electrical loads with normal health status, determine that the charging status of the dangerous electrical load is normal; for dangerous electrical loads with abnormal health status, determine that the charging status of the dangerous electrical load is abnormal.

[0008] Optionally, the long short-term memory network health assessment model is trained in the following manner: Obtain the historical charging data, historical environmental data and simulation data of the dangerous electrical load, as well as the corresponding health labels; Fuse and preprocess the historical charging data, the environmental data and the simulation data of the dangerous electrical load to construct a time series feature set; Use the time series feature set as the input of the long short-term memory network, and use the corresponding historical health labels as the output of the long short-term memory network to construct a training set; Train the long short-term memory network based on the training set. After training is completed, obtain the long short-term memory network health assessment model.

[0009] Optionally, the construction of the long short-term memory network health assessment model further includes: Obtain the charging data, environmental data, and simulation data of the updated dangerous power consumption load, as well as the corresponding health labels, and construct an updated training set; Optimize and train the long short-term memory network health assessment model based on the updated training set to obtain an optimized long short-term memory network health assessment model.

[0010] Optionally, it further includes: Obtain the alarm information corresponding to the dangerous power consumption load; Use the alarm information corresponding to the dangerous power consumption load as the input of a pre-set dangerous behavior classification and grading model, and the dangerous behavior classification and grading model outputs a dangerous behavior classification and grading result; Determine the dangerous power consumption behavior level of the dangerous power consumption load according to the dangerous behavior classification and grading result; Match the dangerous power consumption behavior level with a pre-constructed solution library, and determine the corresponding solution according to the matching result.

[0011] In the second aspect of the present invention, a dangerous power consumption behavior recognition system is provided. Based on the dangerous power consumption behavior recognition method provided in any of the above embodiments, it includes: An entrance AI camera system for monitoring whether there is a dangerous power consumption load entering this area; An electronic access control system for determining the area attribution of the dangerous power consumption load when there is a dangerous power consumption load entering this area; for dangerous power consumption loads that do not belong to this area, generate an alarm message for refusing entry; A charging shed monitoring system for monitoring whether the dangerous power consumption load belonging to this area enters the shed for charging; A charging health detection system for determining the charging status of the dangerous power consumption load that enters the shed for charging, and generating a corresponding alarm message when the charging status is abnormal; A building electronic fence system for determining whether the dangerous power consumption load that does not enter the shed for charging enters the building electronic fence; for dangerous power consumption loads that enter the building electronic fence, generate a corresponding alarm message; An elevator AI camera system for obtaining the monitoring result of the elevator AI camera for the dangerous power consumption load that does not enter the building electronic fence, and determining whether the dangerous power consumption load enters the elevator according to the monitoring result; for dangerous power consumption loads that enter the elevator, generate a corresponding alarm message; An intelligent electricity meter system for obtaining the electricity consumption characteristics of the user identified and analyzed by the intelligent electricity meter for the dangerous power consumption load that does not enter the elevator, and determining whether the dangerous power consumption load charges at home according to the electricity consumption characteristics; for dangerous power consumption loads that charge at home, generate a corresponding alarm message; The background main station system is used to receive and record alarm information and notify relevant management personnel.

[0012] In the third aspect of the present invention, an identification device for dangerous electricity consumption behaviors is provided. Based on the method for identifying dangerous electricity consumption behaviors provided in any one of the above embodiments, it includes: The first identification module is used to monitor whether there is a dangerous electricity consumption load entering this area. When there is a dangerous electricity consumption load entering this area, determine the area attribution of the dangerous electricity consumption load; for the dangerous electricity consumption load that does not belong to this area, generate an alarm message for refusing entry. The second identification module is used to, for the dangerous electricity consumption load that belongs to this area, monitor whether the dangerous electricity consumption load enters the shed for charging; for the dangerous electricity consumption load that enters the shed for charging, determine the charging status of the dangerous electricity consumption load. When the charging status is abnormal, generate a corresponding alarm message. The third identification module is used to, for the dangerous electricity consumption load that does not enter the shed for charging, determine whether the dangerous electricity consumption load enters the building electronic fence; for the dangerous electricity consumption load that enters the building electronic fence, generate a corresponding alarm message. The fourth identification module is used to, for the dangerous electricity consumption load that does not enter the building electronic fence, obtain the monitoring result of the elevator AI camera for the dangerous electricity consumption load, and determine whether the dangerous electricity consumption load enters the elevator according to the monitoring result; for the dangerous electricity consumption load that enters the elevator, generate a corresponding alarm message. The fifth identification module is used to, for the dangerous electricity consumption load that does not enter the elevator, obtain the electricity consumption characteristics of the user obtained by the intelligent electricity meter identification and analysis, and determine whether the dangerous electricity consumption load enters the household for charging according to the electricity consumption characteristics; for the dangerous electricity consumption load that enters the household for charging, generate a corresponding alarm message.

[0013] In the fourth aspect of the present invention, an electronic device is provided, including a memory and a processor; The memory is used to store a computer program; The processor is used to, when executing the computer program, implement the method for identifying dangerous electricity consumption behaviors provided in any one of the above embodiments.

[0014] In the fifth aspect of the present invention, a computer-readable storage medium is provided. A computer program is stored on the computer-readable storage medium. When the computer program is executed by the processor, the method for identifying dangerous electricity consumption behaviors provided in any one of the above embodiments is implemented.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: A method for identifying dangerous behaviors provided by the present invention solves the technical problems of high missed detection rate and response lag in the prior art, and achieves beneficial effects: through the hierarchical logic of areas, bike sheds, buildings, elevators, and users, combined with multiple devices (AI cameras, electronic fences, smart electricity meters), multi-link hierarchical monitoring and intelligent identification are realized, significantly reducing the missed detection rate. At the same time, through the linkage of different scenarios (bike sheds, buildings, elevators, and households) and multiple devices (AI cameras, electronic fences, smart electricity meters), corresponding warning information is generated, constituting a dynamic warning mechanism, realizing the real-time identification and response of dangerous electricity-using behaviors, and improving the processing efficiency.

[0016] Furthermore, by using electronic tags and automated management ledgers, the entry of external dangerous electricity-using loads is prevented, reducing potential safety hazards.

[0017] Furthermore, through charging data, environmental data, and the LSTM model, the health status of dangerous electricity-using loads is predicted. The LSTM model captures temporal features to improve the accuracy of anomaly detection.

[0018] Furthermore, a long short-term memory network model is trained through charging data and environmental data. At the same time, the model is optimized through digital twin simulation and incremental learning, enabling a more accurate assessment of the health status of dangerous electricity-using loads.

[0019] Furthermore, by continuously updating the training set, the long short-term memory network health assessment model is optimized, further improving the accuracy of health status detection.

[0020] Furthermore, through a classification and grading model, the warning information is graded and matched with the processing measures in the solution library to improve the efficiency of resource allocation.

[0021] An identification system, device, electronic device, and computer-readable storage medium for dangerous electricity-using behaviors provided by the present invention also solve the technical problems of high missed detection rate and response lag proposed in the background art. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The accompanying drawings forming a part of this application are used to provide a further understanding of the present invention. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings: Figure 1 is a flowchart of the method for identifying dangerous electricity-using behaviors provided by an embodiment of the present invention; Figure 2 is a structural block diagram of the device for identifying dangerous electricity-using behaviors provided by an embodiment of the present invention; Figure 3 is a structural block diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0023] The present invention will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments. It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments may be combined with each other.

[0024] The following detailed descriptions are all exemplary descriptions, aiming to provide further detailed descriptions of the present invention. Unless otherwise specified, all technical terms adopted in the present invention have the same meaning as commonly understood by those of ordinary skill in the art to which the present application belongs. The terms used in the present invention are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present invention.

[0025] Embodiment 1 As Figure 1 shown, in the first aspect of the present invention, a method for identifying dangerous electricity - using behaviors is provided, including: S1: Monitor whether there is a dangerous electricity - using load entering this area. When a dangerous electricity - using load enters this area, determine the area attribution of the dangerous electricity - using load; for a dangerous electricity - using load that does not belong to this area, generate an alarm message for rejecting entry. S2: For a dangerous electricity - using load that belongs to this area, monitor whether the dangerous electricity - using load enters the shed for charging; for a dangerous electricity - using load that enters the shed for charging, determine the charging state of the dangerous electricity - using load. When the charging state is abnormal, generate a corresponding alarm message. S3: For a dangerous electricity - using load that does not enter the shed for charging, determine whether the dangerous electricity - using load enters the building electronic fence; for a dangerous electricity - using load that enters the building electronic fence, generate a corresponding alarm message. S4: For a dangerous electricity - using load that does not enter the building electronic fence, obtain the monitoring result of the elevator AI camera for the dangerous electricity - using load, and determine whether the dangerous electricity - using load enters the elevator according to the monitoring result; for a dangerous electricity - using load that enters the elevator, generate a corresponding alarm message. S5: For a dangerous electricity - using load that does not enter the elevator, obtain the electricity - using characteristics of the user obtained by the intelligent electricity meter identification and analysis, and determine whether the dangerous electricity - using load enters the household for charging according to the electricity - using characteristics; for a dangerous electricity - using load that enters the household for charging, generate a corresponding alarm message.

[0026] It should be noted that the dangerous power consumption load can be a rechargeable battery; the location for monitoring whether the dangerous power consumption load enters this area and determining its belonging can be set at the entrance of this area; the location for detecting the charging status of the load entering the shed can be set on the charging pile; the location for judging the illegal entry of the building for the load that has not entered the shed can be set at the building entrance; when monitoring the load that has not entered the building in the elevator, the corresponding floor can be recorded through an AI camera; when identifying the power consumption characteristics of the load that has not entered the elevator and then judging whether it is charging at home, it can be identified through a smart electricity meter, and the identified power consumption characteristics can be linked with the recognition results of the AI camera.

[0027] Thus, through the hierarchical logic of the area, shed, building, elevator, and user, combined with multiple devices (AI camera, electronic fence, smart electricity meter), multi-link hierarchical monitoring and intelligent identification are achieved, significantly reducing the missed detection rate. At the same time, corresponding alarm information is generated through the linkage of different scenarios (shed, building, elevator, household) and multiple devices (AI camera, electronic fence, smart electricity meter), constituting a dynamic alarm mechanism, realizing the real-time identification and response of dangerous power consumption behaviors, and improving the processing efficiency.

[0028] In step S1: Monitor whether there is a dangerous power consumption load entering this area. When there is a dangerous power consumption load entering this area, determine the area belonging of the dangerous power consumption load; for the dangerous power consumption load that does not belong to this area, generate an alarm message for refusing entry.

[0029] Here, an AI camera for identifying dangerous power consumption loads can be set at the area entrance. An identification model for dangerous power consumption loads can be set in the AI camera. The AI camera can integrate a variety of advanced technologies such as artificial intelligence, network connection, and high resolution, and can also have functions such as face recognition and object detection. It should be noted that in this step, it is possible to repeatedly cycle through the detection to judge whether dangerous power consumption loads such as electric bicycle rechargeable batteries enter the area. Thus, by monitoring the entry of dangerous power consumption loads in real time, potential safety threats can be quickly identified and responded to, effectively preventing safety accidents caused by the improper use or misoperation of dangerous power consumption loads. At the same time, determining the area belonging of dangerous power consumption loads helps to clarify management responsibilities and ensure that managers in each area can timely understand and handle potential safety hazards within their jurisdiction.

[0030] In one embodiment, the determining the area belonging of the dangerous power consumption load includes: Obtain the recognition result of the dangerous power consumption load by the access control system at the entrance of this area, and determine the electronic tag information of the dangerous power consumption load according to the recognition result; Match the electronic tag information with a pre-set automated management ledger; For the dangerous power consumption loads with mismatched electronic tag information and the automated management ledger, determine that the dangerous power consumption loads do not belong to this area; for the dangerous power consumption loads with matched electronic tag information and the automated management ledger, determine that the dangerous power consumption loads belong to this area.

[0031] Here, the access control system at the entrance can be configured with an electronic tag (RFID / Bluetooth) identification module and a communication module; it should be noted that the electronic tag information includes, but is not limited to, the brand, power, user, and area code of the dangerous power consumption load, etc.; the automated management ledger is determined based on the electronic tag information of all dangerous power consumption loads belonging to this area, including the electronic tag information of all dangerous power consumption loads belonging to this area; the access of the dangerous power consumption load to this area can be refused through management means; when the dangerous power consumption load belongs to this area, the access control system can record the entry time of the dangerous power consumption load; thus, by using the electronic tag and the automated management ledger, it prevents the entry of external dangerous power consumption loads and reduces potential safety hazards.

[0032] In step S2: For the dangerous power consumption loads belonging to this area, monitor whether the dangerous power consumption loads enter the shed for charging; for the dangerous power consumption loads that enter the shed for charging, determine the charging status of the dangerous power consumption load, and when the charging status is abnormal, generate corresponding warning information.

[0033] Here, an AI camera for detecting whether a dangerous power consumption load enters the shed and a charging health status detection module for detecting the charging status of the dangerous power consumption load can be configured in the shed. It should be noted that the AI camera can be pre-configured with an identification model; the charging status health detection module can be pre-configured with a health assessment model, and after obtaining the assessment result, generate warning information; thus, it ensures that the dangerous power consumption load charges within the designated area and complies with relevant charging safety regulations and management requirements, which helps to promote compliance management and reduce legal risks caused by illegal operations. At the same time, by real-time monitoring the charging status of the dangerous power consumption load in the shed, potential charging safety hazards such as overheating, overcharging, and short-circuiting of the battery can be detected in a timely manner, thereby effectively preventing the occurrence of safety accidents such as fires and explosions.

[0034] In one embodiment, the determining the charging status of the dangerous power consumption load includes: Obtain the charging data and environmental data of the dangerous power consumption load; input the charging data and environmental data into a pre-trained long short-term memory network health assessment model, and the long short-term memory network health assessment model outputs the health status assessment result of the dangerous power consumption load; Determine whether the health condition of the dangerous power consumption load is normal according to the health status assessment result; For a dangerous electrical load with normal health status, determine that the charging status of the dangerous electrical load is normal; for a dangerous electrical load with abnormal health status, determine that the charging status of the dangerous electrical load is abnormal.

[0035] Here, a charging health status detection module can be configured in the shed. The charging health status detection module can be provided with a charging process information collection module and a health prediction module. It should be noted that the charging process information collection module is responsible for collecting data during the charging process, including but not limited to charging input AC voltage, current, power factor, output voltage, output current, battery pack temperature, single cell voltage, battery pack temperature, battery pack position information, battery pack protection mechanism status information, etc. The health prediction module can utilize the long short-term memory network (LSTM) deep learning architecture to capture the complex patterns in the data during the charging process, improve the prediction accuracy, and construct a long short-term memory network health assessment model. Thus, the health status of the dangerous electrical load is predicted through charging data, environmental data, and the LSTM model. The LSTM model captures the temporal characteristics to improve the accuracy of anomaly detection.

[0036] In one embodiment, the long short-term memory network health assessment model is trained in the following manner: Obtain the historical charging data, historical environmental data, and simulation data of the dangerous electrical load, as well as the corresponding health labels. Fuse and preprocess the historical charging data, environmental data, and simulation data of the dangerous electrical load to construct a temporal feature set. Use the temporal feature set as the input of the long short-term memory network, and use the corresponding historical health labels as the output of the long short-term memory network to construct a training set. Train the long short-term memory network based on the training set. After training is completed, a long short-term memory network health assessment model is obtained.

[0037] Here, a digital twin simulation module can be set up in the shed. The digital twin simulation module can accurately model based on the characteristics of dangerous power consumption loads of different brands, including electrochemical reactions, thermal effects, etc. In addition, it can also combine the performance characteristics of actual dangerous power consumption loads under different working conditions, including temperature changes, voltage and current fluctuations, etc.; it can also update the state of the digital twin in real time through charging data and environmental data; it should be noted that the fusion preprocessing can first align the actual charging data and simulation data according to the time stamp, construct a joint time series dataset, then standardize the voltage and current data, normalize the temperature data, and finally extract statistical features (mean, variance, peak value) and time series features (sliding window difference, trend term decomposition); thus, a long short-term memory network model is trained through charging data and environmental data. At the same time, the model is optimized through digital twin simulation and incremental learning, and the health state of the dangerous power consumption load can be evaluated more accurately.

[0038] In one embodiment, the construction of the long short-term memory network health assessment model further includes: Obtain the updated charging data, environmental data and simulation data of the dangerous power consumption load and the corresponding health labels, and construct an updated training set; Based on the updated training set, optimize and train the long short-term memory network health assessment model to obtain an optimized long short-term memory network health assessment model.

[0039] Here, the model is retrained regularly using newly collected data to update the model; it should be noted that based on multiple elements such as the number of charge and discharge cycles and environmental temperature, the comprehensiveness and reliability of the prediction can be improved; based on the safety detection model, the health state of the dangerous power consumption load can be detected by combining charging data and environmental data, and abnormal behaviors such as voltage and current mutations can be identified; thus, by continuously updating the training set, the long short-term memory network health assessment model is optimized, and the accuracy of the health state detection is further improved.

[0040] In step S3: For the dangerous power consumption load that does not enter the shed for charging, determine whether the dangerous power consumption load enters the building electronic fence; for the dangerous power consumption load that enters the building electronic fence, generate corresponding warning information.

[0041] Here, a communication module can be set on the building electronic fence. If there is an alarm message on the building electronic fence, the management personnel will be immediately notified to handle the matter; thus, potential safety threats can be discovered and blocked in time to avoid the occurrence of safety accidents. When an alarm occurs, a warning is provided to the user, which can enhance everyone's safety awareness and jointly maintain the safety environment of the building.

[0042] In step S4: For the dangerous power consumption loads that have not entered the building electronic fence, obtain the monitoring results of the elevator AI camera for the dangerous power consumption loads, and determine whether the dangerous power consumption loads enter the elevator according to the monitoring results; for the dangerous power consumption loads that enter the elevator, generate corresponding warning information.

[0043] Here, a pre-trained model for identifying dangerous power consumption loads can be set in the AI camera. The image data obtained by the AI camera is used as the input of the model for identifying dangerous power consumption loads, and the item labels are used as the output of the model for identifying dangerous power consumption loads. It should be noted that the absence of alarm information in the building electronic fence may also be due to reasons such as weak signal and poor recognition effect, resulting in the failure to trigger the electronic fence alarm. Therefore, it cannot be excluded that the dangerous power consumption loads are still sent upstairs. The intelligent camera in the elevator can further identify the dangerous power consumption loads. Thus, by real-time monitoring and identifying the dangerous power consumption loads in the elevator, potential safety hazards can be discovered and handled in a timely manner, thereby reducing the risks of accidents such as fires and electric shocks, reducing labor costs and time costs, improving the management efficiency of power facilities, and reducing the missed inspection rate.

[0044] In one embodiment, the training of the model for the AI camera to identify dangerous power consumption loads can be divided into the following steps: The first step: Collect data, collect a large amount of sample data as the basic content of machine learning. The second step: Feature extraction, extract features from the large number of collected pictures. Features refer to numbers or attributes such as colors and shapes that can depict certain aspects of things. The third step: Label data, map the relevant data to the real categories according to the designed features, and tell the machine the corresponding relationship between features and objects. The fourth step: Train the model. After marking the input data, the machine can learn and train based on these data, and then find a suitable classification model. The data set used for training the model is called the training set.

[0045] The fifth step: Model verification. After training is completed, provide the pictures of the test set to the machine to verify the classification accuracy rate to test the performance of the model, and then continuously optimize the model to obtain the optimal model.

[0046] The sixth step: Model application, distinguish the types of items according to the actual pictures.

[0047] Here, pictures containing different electricity loads can be collected from surveillance cameras, power safety databases, or public datasets; the data can be manually labeled or labeled using tools; when training the model, a pre-trained convolutional neural network (CNN) model such as ResNet, VGG, or MobileNet can be used and fine-tuned on this basis; it should be noted that in the verification stage, an independent test set from the training set should be used to evaluate the performance of the model; thus, by identifying dangerous electricity loads through the model, manual intervention can be reduced and the recognition accuracy can be improved.

[0048] In step S5: for dangerous electricity loads that have not entered the elevator, obtain the electricity consumption characteristics of the user identified and analyzed by the intelligent electricity meter, and determine whether the dangerous electricity load is charging at home according to the electricity consumption characteristics; for dangerous electricity loads charging at home, generate corresponding warning information.

[0049] Here, a load identification module for identifying and analyzing the electricity consumption characteristics of users can be configured for the smart electricity meter. The load identification module can effectively identify the electricity consumption characteristics (turn-on time period, power magnitude, electricity consumption) of common household appliances (such as air conditioners, water heaters, etc.) according to different types of load characteristics, and can also identify and alarm dangerous electricity-consuming loads (such as electric vehicle charging) in the line to prevent certain fire accidents. A communication module can be configured for the smart electricity meter. The communication module can be wired Modbus, Ethernet or other wireless methods. A metering module can be configured for the smart electricity meter. The metering module can measure and record the electricity consumption, including parameters such as voltage, current, active power, reactive power, and electric energy. A power-off-free meter replacement module can also be configured for the smart electricity meter to ensure the continuity of electricity consumption. It should be noted that the failure of the intelligent camera in the elevator to identify dangerous electricity-consuming loads does not rule out the event of dangerous electricity-consuming loads charging at home. Through the load identification function of the smart electricity meter, the charging status of dangerous electricity-consuming loads entering the home is analyzed and judged. An identification model can be preset in the load identification module to quickly identify the electricity consumption characteristics of common household appliances and dangerous electricity-consuming loads. In different application scenarios, considering factors such as cost, safety, and scope, a relatively suitable communication method is selected to provide a strong data basis for advanced application analysis. The power-off-free meter replacement module can be a plug-in base. When replacing the meter, a "plug-and-play" short-circuit plug-in module can be used to bypass the smart electricity meter, thus realizing power-off-free meter replacement on the user side. Thus, by analyzing parameters such as voltage, current, and power of the household circuit, the smart electricity meter can identify the characteristics of dangerous electricity-consuming behaviors such as electric bicycle charging. Even if the dangerous load does not trigger the previous monitoring link (such as not entering the elevator), it can still accurately locate the charging behavior entering the home, achieving full-scenario coverage from the "public area" to the "user indoor", significantly reducing the missed detection rate, ensuring no dead-angle monitoring, improving the missed detection rate, providing sufficient time for management personnel to intervene, avoiding accidents such as fires caused by abnormal charging, and improving the response ability.

[0050] In one embodiment, it further includes: Obtain the alarm information corresponding to the dangerous electricity-consuming load; Use the alarm information corresponding to the dangerous electricity-consuming load as the input of a preset dangerous behavior classification and grading model, and the dangerous behavior classification and grading model outputs a dangerous behavior classification and grading result; Determine the dangerous electricity-consuming behavior level of the dangerous electricity-consuming load according to the dangerous behavior classification and grading result; Match the dangerous electricity-consuming behavior level with a pre-constructed solution library, and determine the corresponding solution according to the matching result.

[0051] Here, the background main station can be configured. The background main station system can adopt an advanced architecture design, including a data acquisition layer, a data processing layer, a data analysis layer, a data storage layer, and an application layer. Interaction can be carried out between layers through efficient data interfaces and communication protocols. The background main station system can also obtain historical data of all dangerous electricity consumption loads in this area, as well as identification data and results of all devices. It should be noted that the collected data can be preprocessed, including removing noise, filling in missing values, handling outliers, etc. The preprocessed data can be further cleaned and formatted. Key features can be extracted from the electricity consumption behavior dataset, such as peak electricity consumption periods, electricity consumption change trends, charge and discharge characteristics of common battery types, etc. Machine learning algorithms can be used to classify and grade electricity consumption behaviors based on the extracted features to identify dangerous electricity consumption behaviors. Statistical analysis can be carried out on various electricity consumption behaviors, including year-on-year and month-on-month analysis of electricity consumption, electricity consumption trend prediction, etc., to reveal the laws and anomalies of electricity consumption behaviors. Alarm information can be classified and graded according to the severity level, such as: unauthorized entry into the area is level one, abnormal charging is level two, unauthorized entry into the building is level three, unauthorized entry into the elevator is level four, and charging into the household is level five. Thus, the background main station can continuously collect, judge, and statistically analyze relevant data on dangerous electricity consumption behaviors of various dangerous electricity consumption loads, and continuously accumulate and improve the relevant database models (such as charge and discharge characteristics of common battery types, charging behavior preferences of users, etc.), thereby supporting the scientific classification and grading of dangerous electricity consumption behaviors and the preparation of solution plans.

[0052] Thus, a method for identifying dangerous electricity consumption behaviors provided by the present invention solves technical problems such as high missed detection rate and response lag in the prior art, and achieves beneficial effects: through the hierarchical logic of area, shed, building, elevator, and user, combined with multiple devices (AI cameras, electronic fences, smart electricity meters), multi-link hierarchical monitoring and intelligent identification are realized, significantly reducing the missed detection rate. At the same time, corresponding alarm information is generated through the linkage of different scenarios (shed, building, elevator, household) and multiple devices (AI cameras, electronic fences, smart electricity meters), constituting a dynamic alarm mechanism, realizing real-time identification and response to dangerous electricity consumption behaviors, and improving the processing efficiency.

[0053] Furthermore, the entry of external dangerous electricity consumption loads is prevented through electronic tags and automated management ledgers, reducing potential safety hazards.

[0054] Furthermore, the health status of dangerous electricity consumption loads is predicted through charging data, environmental data, and the LSTM model. The LSTM model captures temporal features to improve the accuracy of anomaly detection.

[0055] Furthermore, a long short-term memory network model is trained through charging data and environmental data. At the same time, the model is optimized through digital twin simulation and incremental learning, enabling more accurate assessment of the health status of dangerous electricity consumption loads.

[0056] Furthermore, by continuously updating the training set, the long short-term memory network health assessment model is optimized, further improving the accuracy of health status detection.

[0057] Furthermore, the alarm information is classified by the classification and grading model, and the processing measures in the solution library are matched to improve the resource allocation efficiency.

[0058] Embodiment 2 In the second aspect of the present invention, a recognition system for dangerous electricity consumption behaviors is provided. Based on the dangerous electricity consumption behavior recognition method provided in any one of the above embodiments, it includes: An entrance AI camera system for monitoring whether there is a dangerous electricity consumption load entering this area; An electronic access control system for determining the area attribution of the dangerous electricity consumption load when a dangerous electricity consumption load enters this area; for a dangerous electricity consumption load that does not belong to this area, an alarm message for refusing entry is generated; A charging shed monitoring system for monitoring whether the dangerous electricity consumption load belonging to this area enters the shed for charging; A charging health detection system for determining the charging status of the dangerous electricity consumption load that enters the shed for charging, and generating a corresponding alarm message when the charging status is abnormal; A building electronic fence system for determining whether the dangerous electricity consumption load that does not enter the shed for charging enters the building electronic fence; for a dangerous electricity consumption load that enters the building electronic fence, a corresponding alarm message is generated; An elevator AI camera system for obtaining the monitoring result of the elevator AI camera for the dangerous electricity consumption load when the dangerous electricity consumption load does not enter the building electronic fence, and determining whether the dangerous electricity consumption load enters the elevator according to the monitoring result; for a dangerous electricity consumption load that enters the elevator, a corresponding alarm message is generated; An intelligent electricity meter system for obtaining the electricity consumption characteristics of the user obtained by the intelligent electricity meter identification and analysis for the dangerous electricity consumption load that does not enter the elevator, and determining whether the dangerous electricity consumption load enters the household for charging according to the electricity consumption characteristics; for a dangerous electricity consumption load that enters the household for charging, a corresponding alarm message is generated; A background main station system for receiving and recording alarm messages and notifying relevant management personnel.

[0059] It should be noted that each system can record relevant events of the dangerous electricity consumption load. The background main station can continuously collect, judge, statistically analyze relevant data of the dangerous electricity consumption behaviors of various dangerous electricity consumption loads, and continuously accumulate and improve the relevant database model, thereby supporting the scientific classification and grading of dangerous electricity consumption behaviors and the compilation of solution plans.

[0060] Embodiment 3 As Figure 2 shown, based on the same inventive concept as the above embodiments, in the third aspect of the present invention, there is also provided an identification device for dangerous electricity - using behaviors. Based on the dangerous electricity - using behavior identification method provided in any one of the above embodiments, it includes: A first identification module, configured to monitor whether there is a dangerous electricity - using load entering this area. When there is a dangerous electricity - using load entering this area, determine the area attribution of the dangerous electricity - using load; for a dangerous electricity - using load that does not belong to this area, generate an alarm message for refusal to enter; A second identification module, configured to, for a dangerous electricity - using load that belongs to this area, monitor whether the dangerous electricity - using load enters the shed for charging; for a dangerous electricity - using load that enters the shed for charging, determine the charging state of the dangerous electricity - using load. When the charging state is abnormal, generate a corresponding alarm message; A third identification module, configured to, for a dangerous electricity - using load that does not enter the shed for charging, determine whether the dangerous electricity - using load enters the building electronic fence; for a dangerous electricity - using load that enters the building electronic fence, generate a corresponding alarm message; A fourth identification module, configured to, for a dangerous electricity - using load that does not enter the building electronic fence, obtain the monitoring result of the elevator AI camera for the dangerous electricity - using load, and determine whether the dangerous electricity - using load enters the elevator according to the monitoring result; for a dangerous electricity - using load that enters the elevator, generate a corresponding alarm message; A fifth identification module, configured to, for a dangerous electricity - using load that does not enter the elevator, obtain the electricity - using characteristics of the user obtained by the intelligent electricity meter identification and analysis, and determine whether the dangerous electricity - using load enters the household for charging according to the electricity - using characteristics; for a dangerous electricity - using load that enters the household for charging, generate a corresponding alarm message.

[0061] Embodiment 4 As Figure 3 shown, in the fourth aspect of the present invention, there is also provided an electronic device 100 for implementing the dangerous electricity - using behavior identification method provided in any one of the above embodiments; The electronic device 100 includes a memory 101, at least one processor 102, a computer program 103 stored in the memory 101 and executable on at least one processor 102, and at least one communication bus 104.

[0062] The memory 101 can be used to store the computer program 103. The processor 102 realizes the steps of a dangerous electricity - using behavior identification method by running or executing the computer program stored in the memory 101 and calling the data stored in the memory 101.

[0063] The memory 101 may mainly include a program storage area and a data storage area. Among them, the program storage area may store an operating system, application programs required for at least one function (such as a sound playback function, an image playback function, etc.); the data storage area may store data created according to the use of the electronic device 100 (such as audio data, etc.). In addition, the memory 101 may include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, at least one magnetic disk storage device, a flash memory device, or other non-volatile solid-state storage devices.

[0064] At least one processor 102 may be a Central Processing Unit (CPU), or may also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor 102 may be a microprocessor or the processor 102 may also be any conventional processor, etc. The processor 102 is the control center of the electronic device 100, and connects various parts of the entire electronic device 100 through various interfaces and lines.

[0065] The memory 101 in the electronic device 100 stores multiple instructions to implement a method for identifying dangerous electricity consumption behaviors. The processor 102 can execute the multiple instructions to implement: Monitor whether there is a dangerous electricity consumption load entering this area. When there is a dangerous electricity consumption load entering this area, determine the area attribution of the dangerous electricity consumption load; for the dangerous electricity consumption load that does not belong to this area, generate an alarm message for refusing entry; For the dangerous electricity consumption load that belongs to this area, monitor whether the dangerous electricity consumption load enters the shed for charging; for the dangerous electricity consumption load that enters the shed for charging, determine the charging state of the dangerous electricity consumption load. When the charging state is abnormal, generate a corresponding alarm message; For the dangerous electricity consumption load that does not enter the shed for charging, determine whether the dangerous electricity consumption load enters the building electronic fence; for the dangerous electricity consumption load that enters the building electronic fence, generate a corresponding alarm message; For dangerous power consumption loads that have not entered the building's electronic fence, obtain the monitoring results of the dangerous power consumption loads by the elevator AI camera, and determine whether the dangerous power consumption loads have entered the elevator according to the monitoring results; for dangerous power consumption loads that have entered the elevator, generate corresponding warning information. For dangerous power consumption loads that have not entered the elevator, obtain the power consumption characteristics of the user identified and analyzed by the intelligent electricity meter, and determine whether the dangerous power consumption load is charging at home according to the power consumption characteristics; for dangerous power consumption loads that are charging at home, generate corresponding warning information.

[0066] Embodiment 5 In the fifth aspect of the present invention, there is also provided a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the dangerous power consumption behavior recognition method provided in any of the above embodiments.

[0067] It should be noted that if the modules / units integrated in the electronic device 100 are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, to implement all or part of the processes in the above embodiment methods of the present invention, it can also be completed by a computer program instructing relevant hardware. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, and read-only memory (ROM, Read-Only Memory).

[0068] Thus, the identification system, device, electronic device, and computer-readable storage medium for dangerous power consumption behavior provided by the present invention also solve the technical problems such as high missed detection rate and response lag proposed in the background art.

[0069] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, system, or computer program product. Therefore, the present invention can be implemented in the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can be implemented in the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0070] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, as well as the combination of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate a means for realizing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or a means for realizing the functions specified in multiple blocks.

[0071] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing devices to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means, and the instruction means realizes the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or a means for realizing the functions specified in multiple blocks.

[0072] These computer program instructions can also be loaded onto a computer or other programmable data processing devices, such that a series of operation steps are executed on the computer or other programmable devices to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable devices provide steps for realizing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or a means for realizing the functions specified in multiple blocks.

[0073] In the description of this specification, the descriptions referring to terms such as "one embodiment", "example", "specific example", etc. mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples.

[0074] 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 above embodiments, those of ordinary skill in the art should understand that: modifications or equivalent replacements can still be made to the specific embodiments of the present invention, and any modifications or equivalent replacements that do not depart from the spirit and scope of the present invention should be covered by the protection scope of the claims of the present invention.

Claims

1. A method for identifying dangerous electricity use behavior, characterized in that: include: Monitor whether there is any dangerous power load entering the area. When a dangerous power load enters the area, determine the area to which the dangerous power load belongs; for dangerous power loads that do not belong to the area, generate an alarm message of denying entry; For dangerous power loads belonging to this area, monitor whether the dangerous power loads enter the carport for charging; for dangerous power loads entering the carport for charging, determine the charging status of the dangerous power loads, and generate corresponding alarm information when the charging status is abnormal; For the dangerous power load that has not entered the carport for charging, determine whether the dangerous power load has entered the building electronic fence; Generate corresponding alarm information for dangerous power loads entering the building's electronic fence; For dangerous power loads that have not entered the building's electronic fence, obtain the monitoring results of the elevator AI camera on the dangerous power loads, and determine whether the dangerous power loads have entered the elevator based on the monitoring results; Generate corresponding alarm information for dangerous electrical loads entering the elevator; For dangerous power loads that have not entered the elevator, the power consumption characteristics of the user obtained by the smart power meter identification and analysis are obtained, and it is determined whether the dangerous power load is charged at home based on the power consumption characteristics; for dangerous power loads charged at home, corresponding alarm information is generated.

2. A method for identifying dangerous electricity use behavior according to claim 1, characterized in that: The step of determining the regional affiliation of the dangerous power load comprises: Obtaining the identification result of the entrance access control system of the area for the dangerous power load, and determining the electronic tag information of the dangerous power load according to the identification result; Matching the electronic tag information with a pre-set automated management ledger; For the dangerous power load whose electronic tag information does not match the automated management ledger, it is determined that the dangerous power load does not belong to this area; for the dangerous power load whose electronic tag information matches the automated management ledger, it is determined that the dangerous power load belongs to this area.

3. A method for identifying dangerous electricity use behavior according to claim 1, characterized in that: The determining the charging state of the dangerous electrical load includes: Acquire charging data and environmental data of the dangerous power load; input the charging data and environmental data into a pre-trained long short-term memory network health assessment model, and the long short-term memory network health assessment model outputs a health status assessment result of the dangerous power load; Determining whether the health status of the dangerous power load is normal according to the health status assessment result; For a dangerous power load in a normal health state, it is determined that the charging state of the dangerous power load is normal; for a dangerous power load in an abnormal health state, it is determined that the charging state of the dangerous power load is abnormal.

4. A method for identifying dangerous electricity use behavior according to claim 3, characterized in that: The long short-term memory network health assessment model is trained as follows: Obtain historical charging data, historical environmental data, and simulation data of hazardous power loads, as well as corresponding health labels; Performing fusion preprocessing on the charging data of the historical dangerous power load, the environmental data and the simulation data to construct a time series feature set; Using the time series feature set as the input of the long short-term memory network, using the corresponding historical health label as the output of the long short-term memory network, and constructing a training set; The long short-term memory network is trained based on the training set. After the training is completed, a long short-term memory network health assessment model is obtained.

5. A method for identifying dangerous electricity use behavior according to claim 4, characterized in that: The construction of the long short-term memory network health assessment model also includes: Obtain updated charging data, environmental data, simulation data and corresponding health labels of dangerous power loads to build an updated training set; The long short-term memory network health assessment model is optimized and trained based on the updated training set to obtain an optimized long short-term memory network health assessment model.

6. A method for identifying dangerous electricity use behavior according to claim 1, characterized in that: Also includes: Obtaining warning information corresponding to dangerous electrical loads; The warning information corresponding to the dangerous power load is used as an input of a pre-set dangerous behavior classification and grading model, and the dangerous behavior classification and grading model outputs a dangerous behavior classification and grading result; Determine the dangerous electricity consumption behavior level of the dangerous electricity load according to the dangerous behavior classification and grading results; The dangerous electricity usage behavior level is matched with a pre-built solution library, and a corresponding solution is determined based on the matching result.

7. A system for identifying dangerous electricity use behavior, characterized in that: The method for identifying dangerous electricity use behavior according to any one of claims 1 to 6 comprises: Entrance AI camera system to monitor whether dangerous electrical loads enter the area; The electronic access control system is used to determine the area to which a dangerous power load belongs when it enters the area; for dangerous power loads that do not belong to the area, an alarm message of denying entry is generated; The charging carport monitoring system is used to monitor whether the dangerous power loads belonging to this area enter the carport for charging; A charging health detection system is used to determine the charging status of dangerous power loads entering the carport for charging, and generate corresponding alarm information when the charging status is abnormal; The building electronic fence system is used to determine whether the dangerous power load that has not entered the carport for charging has entered the building electronic fence; for the dangerous power load that has entered the building electronic fence, generate corresponding alarm information; The elevator AI camera system is used to obtain the monitoring results of the elevator AI camera on the dangerous power load that has not entered the building electronic fence, and determine whether the dangerous power load has entered the elevator according to the monitoring results; for the dangerous power load that has entered the elevator, generate corresponding alarm information; The smart energy meter system is used to obtain the power consumption characteristics of the user obtained by the smart energy meter identification and analysis for the dangerous power load that has not entered the elevator, and determine whether the dangerous power load is charged at home according to the power consumption characteristics; for the dangerous power load charged at home, generate corresponding alarm information; The backend master station system is used to receive and record alarm information and notify relevant managers.

8. A device for identifying dangerous electricity use behavior, characterized in that: include: The first identification module is used to monitor whether there is a dangerous power load entering the area. When a dangerous power load enters the area, the area to which the dangerous power load belongs is determined; for dangerous power loads that do not belong to the area, an entry denial alarm message is generated; The second identification module is used to monitor whether the dangerous power load belonging to the local area enters the carport for charging; for the dangerous power load entering the carport for charging, determine the charging state of the dangerous power load, and generate corresponding alarm information when the charging state is abnormal; The third identification module is used to determine whether the dangerous power load that has not entered the carport for charging has entered the building electronic fence; Generate corresponding alarm information for dangerous power loads entering the building's electronic fence; The fourth identification module is used to obtain the monitoring results of the elevator AI camera on the dangerous power load that has not entered the building electronic fence, and determine whether the dangerous power load has entered the elevator according to the monitoring results; Generate corresponding alarm information for dangerous electrical loads entering the elevator; The fifth identification module is used to obtain the power consumption characteristics of the user obtained by the smart power meter identification and analysis for the dangerous power load that has not entered the elevator, and determine whether the dangerous power load is charged at home based on the power consumption characteristics; for the dangerous power load charged at home, generate corresponding alarm information.

9. An electronic device, characterized in that: including memory and processor; The memory is used to store computer programs; The processor is used to implement the method for identifying dangerous electricity usage behavior as described in any one of claims 1 to 6 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for identifying dangerous electricity usage behavior as described in any one of claims 1 to 6 is implemented.

Images