Safety monitoring method and device for home-entry charging of electric vehicle

Abstract

The invention discloses a safety monitoring method and device for electric vehicle home-entry charging, and relates to the field of electric vehicle charging safety, and the method comprises the steps: obtaining a current value of a target resident according to a preset collection frequency, and obtaining a power utilization data sequence; extracting a plurality of time domain characteristic values of the power consumption data sequence; converting the power utilization data sequence into a frequency domain, and extracting frequency composition and energy distribution of the current signal to obtain a plurality of frequency domain characteristic values; splicing and integrating the time domain characteristic value and the frequency domain characteristic value according to a preset format to obtain a power utilization characteristic vector; performing encoding processing on the power utilization characteristic vector to obtain a power utilization encoding file, and sending the power utilization encoding file to a back-end server; and receiving a power utilization identification result returned by the back-end server, and sending early warning information to the property management terminal when determining that the target resident has an electric vehicle charging behavior according to the power utilization identification result. According to the application, the electric vehicle home-entry charging behavior can be monitored and recognized in real time, and the fire safety of residential buildings of a community is guaranteed.

Description

Technical Field

[0001] The present application relates to the field of electric vehicle charging safety, and in particular to a safety monitoring method and device for home charging of electric vehicles. Background Art

[0002] In recent years, electric bicycles have gained widespread adoption due to their convenience, affordability, and environmental friendliness. However, with the surge in their use, fires caused by battery quality issues and improper charging have become increasingly common. In particular, fires caused by electric bicycle charging in residential areas have resulted in significant casualties and property damage, raising the issue of electric bicycle charging safety.

[0003] In related technologies, property management departments mainly prevent illegal charging of electric bicycles through manual management methods such as daily inspections and posting notices prohibiting charging; some communities have installed video surveillance systems, which use feature recognition of video images to monitor electric vehicles and their batteries entering elevators or homes, and provide timely warnings.

[0004] However, patrols at night or during periods when there are few people are often insufficient, and manual processing can easily lead to a regulatory vacuum; and although video surveillance works around the clock, it is difficult to identify illegal charging behaviors that are deliberately obscured. Summary of the Invention

[0005] The present application provides a safety monitoring method and device for charging electric vehicles at home, which is used to monitor and identify the behavior of charging electric vehicles at home in real time, and ensure the fire safety of residential buildings in the community.

[0006] In the first aspect, the present application provides a safety monitoring method for charging electric vehicles at home, which is applied to a safety monitoring device, and the method includes: obtaining the current value of the target household according to a preset collection frequency to obtain a power consumption data sequence; when it is detected that the current value in the power consumption data sequence exceeds a preset normal threshold and the duration exceeds a preset duration, extracting multiple time domain eigenvalues ​​of the power consumption data sequence; converting the power consumption data sequence to the frequency domain, extracting the frequency composition and energy distribution of the current signal, and obtaining multiple frequency domain eigenvalues; splicing and integrating the time domain eigenvalues ​​and the frequency domain eigenvalues ​​according to a preset format to obtain a power consumption feature vector; encoding the power consumption feature vector to obtain a power consumption coding file, and sending the power consumption coding file to a back-end server; receiving the power consumption identification result returned by the back-end server, and when it is determined that the target household has an electric vehicle charging behavior based on the power consumption identification result, sending an early warning information to the property management terminal.

[0007] In the above embodiment, the safety monitoring device collects current values ​​at a preset frequency and analyzes time domain and frequency domain characteristics. Combined with the preset threshold and duration judgment, it can accurately identify the charging behavior of electric vehicles, encode the feature vector and send it to the back-end server for identification, and send an early warning to the property in a timely manner when the charging behavior is confirmed. It realizes real-time monitoring and rapid response to the charging behavior of electric vehicles entering homes, and ensures the fire safety of residential buildings in the community.

[0008] In combination with some embodiments of the first aspect, in some embodiments, the step of obtaining the current value of the target household according to a preset collection frequency to obtain the electricity consumption data sequence specifically includes: collecting the electricity load characteristics of the target household, and determining the initial sampling frequency of the current value based on the electricity load characteristics; collecting the current value at the initial sampling frequency to obtain the initial electricity consumption sequence, and calculating the signal-to-noise ratio of the initial electricity consumption sequence in real time; when the signal-to-noise ratio is lower than the stable threshold, increasing the initial sampling frequency to a corrected sampling frequency so that the signal-to-noise ratio reaches the stable threshold; collecting the current value at the corrected sampling frequency to obtain the electricity consumption data sequence.

[0009] In the above embodiment, the safety monitoring device can adaptively select a suitable sampling frequency according to the power load characteristics by dynamically adjusting the sampling frequency, and optimize the sampling effect through signal-to-noise ratio evaluation. When the signal-to-noise ratio is insufficient, the sampling frequency is automatically increased to ensure the quality and reliability of the collected data.

[0010] In combination with some embodiments of the first aspect, in some embodiments, the time domain eigenvalues ​​and frequency domain eigenvalues ​​are spliced ​​and integrated according to a preset format to obtain a power consumption characteristic vector, which specifically includes: performing data standardization processing on the time domain eigenvalues, converting eigenvalues ​​of different dimensions into the same numerical range, and obtaining time domain standard values; using the principal component analysis method to extract the main frequency domain characteristic components of the frequency domain eigenvalues, reducing the characteristic space dimension of the frequency domain eigenvalues, and obtaining frequency domain standard values; splicing and integrating the time domain standard values ​​and the frequency domain standard values ​​according to a preset splicing order to obtain the power consumption characteristic vector.

[0011] In the above embodiment, the safety monitoring device effectively solves the problem of unifying features of different dimensions through data standardization processing and principal component analysis, and reduces the dimension of the feature space, which not only improves the expression efficiency of the feature vector, but also enhances the generalization ability of the model.

[0012] In combination with some embodiments of the first aspect, in some embodiments, before the step of obtaining the current value of the target household according to the preset collection frequency to obtain the electricity consumption data sequence, the method also includes: obtaining the target household's electrical equipment information including the type of electrical equipment, the rated power of the equipment, and the equipment usage period; determining the target household's electricity load baseline and electricity consumption fluctuation range for each time period based on the electrical equipment information, and constructing a basic electricity consumption model; determining the preset normal threshold and preset duration for triggering electricity consumption identification based on the electrical equipment information and the basic electricity consumption model.

[0013] In the above embodiment, the safety monitoring device can accurately grasp the normal electricity usage patterns of residents by pre-acquiring information on electrical equipment and establishing a basic electricity usage model. This model building method based on historical data improves the rationality of threshold setting and the accuracy of identification.

[0014] In combination with some embodiments of the first aspect, in some embodiments, the steps of determining the preset normal threshold and preset duration for triggering electricity consumption identification based on the electricity consumption equipment information and the basic electricity consumption model specifically include: based on the electricity consumption equipment information, using the sliding time window method to perform statistical analysis on the electricity load in each time period to obtain the electricity consumption fluctuation pattern in different time periods; performing cluster analysis on the electricity consumption fluctuation pattern to obtain multiple typical electricity consumption scenarios; calculating the current mean and standard deviation of typical electricity consumption scenarios based on the basic electricity consumption model; determining the preset normal threshold based on the current mean and standard deviation; determining the shortest duration for distinguishing normal electricity consumption from electric vehicle charging behavior based on the electric vehicle charging characteristic curve and the electricity consumption fluctuation pattern to obtain the preset duration.

[0015] In the above embodiment, the safety monitoring device can accurately characterize the electricity consumption characteristics of different time periods through sliding time window statistical analysis and cluster analysis, and establish scientific judgment standards in combination with the electric vehicle charging characteristic curve, effectively reducing the false alarm rate.

[0016] In combination with some embodiments of the first aspect, in some embodiments, after receiving the electricity usage identification result returned by the back-end server, and determining that the target resident has engaged in electric vehicle charging behavior based on the electricity usage identification result, after the step of sending an early warning message to the property management terminal, the method also includes: obtaining the historical violation records of the target resident and the fire safety level of the building unit; determining the processing priority of the early warning processing based on the historical violation records and the fire safety level; and generating a task processing list based on multiple early warning information and corresponding processing priorities.

[0017] In the above embodiment, the security monitoring device establishes a scientific early warning processing priority mechanism by combining historical violation records and fire safety levels, enabling the property to allocate management resources more effectively and improving the efficiency and pertinence of early warning processing.

[0018] In combination with some embodiments of the first aspect, in some embodiments, after the step of generating a task processing list based on multiple warning information and corresponding processing priorities, the method also includes: collecting resident information and warning information of the target residents to generate warning details; constructing a warning tracking file based on the warning details; and updating the warning tracking file based on the processing time and processing results uploaded by the property personnel.

[0019] In the above embodiment, the safety monitoring device realizes the whole-process tracking management of the warning events by establishing the warning tracking file and updating the processing results in real time, which not only improves the processing efficiency but also provides data support for subsequent management optimization.

[0020] In second aspect, an embodiment of the present application provides a security monitoring device, which includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the security monitoring device to execute the method described in the first aspect and any possible implementation method of the first aspect.

[0021] In a third aspect, an embodiment of the present application provides a computer program product comprising instructions, which, when the computer program product is run on a security monitoring device, enables the security monitoring device to execute the method described in the first aspect and any possible implementation of the first aspect.

[0022] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium comprising instructions. When the instructions are executed on a security monitoring device, the security monitoring device executes the method described in the first aspect and any possible implementation of the first aspect.

[0023] It is understood that the safety monitoring device provided in the second aspect, the computer program product provided in the third aspect, and the computer storage medium provided in the fourth aspect are all used to execute the methods provided in the embodiments of the present application. Therefore, the beneficial effects that can be achieved can be referred to the beneficial effects of the corresponding methods and will not be repeated here.

[0024] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages: 1. By using a multi-dimensional analysis method based on current characteristics, it can automatically collect and analyze electricity usage data. By extracting and integrating time-domain and frequency-domain features, it can accurately identify electric vehicle charging behavior. Therefore, charging behavior detection can be completed without relying on manual inspections or video surveillance. This effectively solves the problems of manual inspections and video surveillance that are easily circumvented in existing technologies. It then realizes all-weather and automated electric vehicle charging safety monitoring, improving the efficiency and reliability of residential electricity safety management.

[0025] 2. By using a model-building method based on electrical device information, we can accurately capture and analyze household information such as device type, power, and usage habits, establishing a personalized baseline model for electricity usage. This allows us to accurately identify each household's normal electricity usage patterns and fluctuation ranges, effectively resolving the issues of irrational early warning threshold settings and high false alarm rates in existing technologies. This enables precise identification based on actual electricity usage characteristics, improving the accuracy of identifying abnormal electricity usage behavior and reducing false alarms and missed alerts.

[0026] 3. Utilizing a multi-dimensional risk assessment mechanism based on historical violation records and fire safety levels, the system intelligently analyzes and assesses the risk level of each warning event, determining its handling priority accordingly. This allows for scientific, tiered handling of warning events, effectively resolving the existing issues of a single warning handling mechanism and irrational resource allocation. This enables differentiated management of warning handling, improves property management efficiency, and ensures that high-risk events are promptly addressed. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 This is a flow chart of a safety monitoring method for home charging of electric vehicles in an embodiment of the present application; Figure 2 This is another flow chart of the safety monitoring method for home charging of electric vehicles in an embodiment of the present application; Figure 3 This is a schematic diagram of the physical device structure of the safety monitoring device in the embodiment of the present application. DETAILED DESCRIPTION

[0028] The terms used in the following examples of the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in the specification of the present application, the singular expressions "a", "an", "above", "the", and "this" are intended to include plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used in the present application refers to any or all possible combinations of one or more of the listed items.

[0029] In the following, the terms "first" and "second" are used for descriptive purposes only and should not be understood to imply or suggest relative importance or implicitly indicate the number of the technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of this application, unless otherwise specified, "plurality" means two or more.

[0030] For ease of understanding, the application scenarios of the embodiments of the present application are introduced below.

[0031] In a large residential complex, with the recent rise in popularity of electric bicycles, residents frequently charge their vehicles at home. In January 2023, a fire broke out in Building A of the complex, sparked by a battery failure while charging an electric vehicle. This necessitated the evacuation of the entire floor and posed a serious safety hazard. A property management investigation revealed that approximately 10% of residents in the complex were charging their vehicles indoors without authorization, but traditional inspections struggled to detect these activities. Furthermore, the complex is plagued by aging wiring, and unauthorized charging can easily overload the wiring, increasing the risk of fire.

[0032] In related technologies, charging behavior can be monitored through manual inspections. However, manual inspections rely primarily on regular checks by property management personnel, which can be inefficient. The following describes a scenario using a related technology method for monitoring the safety of electric vehicle charging at home.

[0033] The residential property management initially implemented a plan of scheduled inspections and smoke alarms. Security guards were assigned to conduct inspections at fixed times daily, focusing on checking for charging activity in the hallways. Smoke alarms were also installed in every household to alert the public when smoke was detected. However, due to the high labor costs of inspections and the lack of 24-hour monitoring, residents often charged their devices at night, avoiding inspection hours. As a result, smoke alarms could only activate after a fire had occurred, failing to prevent it. This made it difficult for the property management to accurately locate offending residents, leading to conflicts between the property management and residents.

[0034] The safety monitoring method for electric vehicle charging at home, as described in the embodiments of this application, achieves accurate identification of charging behavior by establishing a basic electricity usage model and a multi-dimensional feature analysis method. This not only accurately detects illegal charging but also reduces the false alarm rate. The following describes a scenario in which the safety monitoring method for electric vehicle charging at home, as described in this application, is used.

[0035] After implementing this solution's safety monitoring device, the property management company conducted a pilot program in Building B. The device collected and analyzed each household's electricity usage data to build a power usage model. When a resident charged their electric vehicle at night, the device detected a discrepancy between their electricity usage and their daily pattern, immediately alerting the property management team. Property management personnel promptly visited the resident's home and discovered that they had engaged in illegal charging. The detailed electricity usage data provided by the safety monitoring device provided concrete evidence of the charging behavior, which the resident approved of and ceased, eliminating the safety hazard.

[0036] It can be seen that the safety monitoring method for home charging of electric vehicles in the embodiment of the present application can not only realize charging behavior monitoring, but also effectively solve the problems of frequent false alarms, insufficient evidence and low efficiency in traditional solutions, thereby realizing intelligent management of electricity safety in residential areas.

[0037] For ease of understanding, the following describes the process of the method provided by this implementation in combination with the above scenario. Figure 1 , which is a flow chart of a safety monitoring method for home charging of electric vehicles in an embodiment of the present application.

[0038] S101. Obtain the current value of the target household according to a preset collection frequency to obtain a power consumption data sequence.

[0039] The preset sampling frequency refers to the time interval for current sampling, typically expressed in Hertz (Hz), set by the safety monitoring device. The target household represents the specific household unit for which electricity usage monitoring is required. The current value refers to the instantaneous value of the current measured at the sampling moment, expressed in amperes (A). The electricity usage data sequence represents a data set consisting of multiple current sampling values ​​arranged in chronological order, and is used to characterize the household's electricity usage behavior.

[0040] After completing initial configuration, the safety monitoring device needs to continuously collect electricity usage data from the target household. Specifically, the safety monitoring device first obtains the preset sampling frequency parameters based on the system configuration, and then periodically reads the current value at this frequency at the meter terminal or a dedicated data collection device. For each sampling, the safety monitoring device associates and stores the acquired current value with the corresponding timestamp and organizes this data into a data sequence in chronological order. At the same time, the safety monitoring device performs a preliminary quality check on the collected data, eliminating obviously abnormal data points to ensure the continuity and reliability of the data sequence.

[0041] In some embodiments, the collection and sequence construction of current data can be achieved in a variety of ways: Optionally, a dedicated data acquisition module is installed at the meter end, the initial sampling frequency is set to 1Hz, and the data quality is monitored in real time; when a sharp fluctuation in the data is detected, the sampling frequency is automatically increased to 10Hz, and the collected data is pre-processed, including denoising and outlier processing; the processed data is sorted and stored by timestamp. Optionally, an intelligent current detection device is installed at the distribution box, and an adaptive sampling strategy is adopted to dynamically adjust the sampling frequency according to changes in power load; a data caching mechanism is established to ensure data continuity, calculate the statistical characteristics of the data in real time, and dynamically optimize the sampling parameters according to the statistical characteristics. It is understandable that other methods can also be used to achieve the collection and processing of current data, such as using distributed sensor networks or smart power meters, which are not limited here.

[0042] S102 : When it is detected that the current value in the power consumption data sequence exceeds a preset normal threshold and the duration exceeds a preset duration, extract multiple time domain feature values ​​of the power consumption data sequence.

[0043] The preset normal threshold refers to the current peak value judgment standard preset by the safety monitoring device, which is used to distinguish normal from abnormal power usage. The preset duration indicates the minimum time the current value must exceed the threshold. The time domain characteristic value refers to the statistical quantity that describes the current variation characteristics in the time dimension, including indicators such as mean, variance, and crest factor.

[0044] During the continuous data collection process, the safety monitoring device needs to monitor the changing trends of the current value in real time. Specifically, the safety monitoring device first compares the real-time current value with a preset normal threshold and simultaneously starts a timer to record the duration of the threshold exceeding the threshold. When the dual trigger conditions are met, the safety monitoring device traces back a certain period of data from the current time and extracts the time domain characteristics of this data. This includes calculating statistics and analyzing waveform characteristics, providing a basis for subsequent identification.

[0045] The time domain characteristic values ​​include maximum value, minimum value, range, mean, median, mode, standard deviation, root mean square value, mean square value (second-order origin moment), k-order central moment, k-order origin moment, skewness (third-order standard moment), kurtosis (fourth-order standard moment), shape factor, peak factor, impulse factor, kurtosis factor, margin factor, etc. The calculation formula is: maximum value ma = max(y); minimum value mi = min(y); average value me = mean(y); peak value pk = ma-mi; mean value av = mean(abs(y)); variance va = var(y); standard deviation st = std(y); kurtosis ku = kurtosis(y); root mean square rm = rms(y); waveform factor S = rm / av; peak factor C = pk / rm; impulse factor I = pk / av; kurtosis factor Kr = sum(y.^4) / sqrt(sum(y.^2)); margin factor L = pk / mean(sqrt(abs(y)))^2.

[0046] In some embodiments, anomaly detection and feature extraction can be achieved through a variety of methods: optionally, a multi-level threshold judgment mechanism can be established; a sliding window method can be used to calculate feature values; basic statistics such as mean, standard deviation, and peak can be extracted; slope and curvature characteristics of the waveform can be calculated; and trend and periodic characteristics of the data can be analyzed. Optionally, an adaptive threshold update mechanism can be established; time domain analysis can be performed using wavelet transforms; time series correlation indicators can be calculated; morphological features of the waveform can be extracted; and mutation characteristics of the data can be analyzed. It is understood that other methods for anomaly detection and feature extraction can also be used, such as deep learning methods or expert systems, which are not limited here.

[0047] S103 : Convert the electricity consumption data sequence into the frequency domain, extract the frequency composition and energy distribution of the current signal, and obtain a plurality of frequency domain eigenvalues.

[0048] The frequency domain refers to the representation space of a signal in the frequency dimension. Frequency composition describes the composition of different frequency components in a signal. Energy distribution refers to the distribution of signal energy across various frequency components. Frequency domain eigenvalues ​​represent various indicators that describe the frequency characteristics of a signal.

[0049] After acquiring time-domain data, the safety monitoring device needs to perform frequency-domain analysis to obtain more characteristic information. Specifically, the safety monitoring device first performs a fast Fourier transform (FFT) on the time-domain data to obtain the signal's spectrum. It then analyzes the energy distribution characteristics of the spectrum, extracting frequency-domain eigenvalues ​​such as characteristic frequencies and harmonic ratios, and calculating statistics such as power spectral density.

[0050] Frequency domain eigenvalues ​​include spectrum, average frequency, center of gravity frequency, root mean square frequency, frequency standard deviation, harmonic component, power spectrum density, zero crossing frequency and total energy.

[0051] The calculation formula is: 1. Average frequency: Wherein, u(n) is the signal sequence, N is the sequence length, and the average frequency is used to represent the arithmetic mean of the signal sequence.

[0052] 2. Center of gravity frequency: Among them, u(n) is the signal sequence, is the conjugate value of the signal sequence, N is the sequence length, and 2π is the angular frequency conversion coefficient.

[0053] 3. Root mean square frequency: Among them, u(n) is the signal sequence, N is the sequence length, 4π 2 is the square of the angular frequency conversion coefficient; the RMS frequency is used to represent the RMS value of the signal frequency component.

[0054] 4. Frequency standard deviation: Among them, u(n) is the signal sequence, N is the sequence length, F FC is the center frequency; the frequency standard deviation indicates the degree of dispersion of the frequency component relative to the center frequency.

[0055] 5. Harmonic components are analyzed based on Fourier transform: Where X(k) is the frequency domain sequence, x(n) is the time domain sequence, N is the sequence length, j is the imaginary unit, k is the frequency index (0 to N-1), and n is the time index.

[0056] 6. Total Energy: Where X(k) is the frequency domain sequence after Fourier transform, and |X(k)|2 is the square of the modulus of the complex number.

[0057] 7. Power Spectral Density: Where T is the signal period, X(f) is the continuous Fourier transform, |X(f)|² is the square of the modulus of the spectrum amplitude, and the power spectral density represents the signal power distribution within a unit frequency interval.

[0058] S104: splicing and integrating the time domain eigenvalues ​​and the frequency domain eigenvalues ​​according to a preset format to obtain a power consumption eigenvector.

[0059] The preset format refers to the predefined way of organizing feature values ​​in the safety monitoring device. The feature vector represents the vector form of multi-dimensional features organized in a specific order.

[0060] After obtaining time-domain and frequency-domain features, the safety monitoring device needs to integrate them into a unified feature representation. Specifically, the safety monitoring device first standardizes the eigenvalues ​​to make them comparable. Then, it combines the eigenvalues ​​into a feature vector in a preset order and format, ensuring that the dimension and format of the feature vector meet the requirements of subsequent processing.

[0061] S105: Encode the electricity usage feature vector to obtain an electricity usage encoding file, and send the electricity usage encoding file to a backend server.

[0062] Encoding refers to converting feature vectors into a data format suitable for transmission and storage. The encoded feature data file is represented by an electronically encoded file. The backend server is a remote computing server used to process and analyze data.

[0063] After obtaining the feature vector, the security monitoring device needs to transmit the data to the backend for processing. Specifically, the security monitoring device first compresses and encodes the feature vector to generate a data file that is easy to transmit. Then, it establishes a communication connection with the backend server to complete the secure transmission of the data and ensure its integrity and reliability.

[0064] In some embodiments, data encoding and transmission can be achieved through a variety of methods: optionally, using a data compression algorithm, adding checksum information for encryption, establishing a communication connection, and executing data transmission. Optionally, a distributed storage strategy can be used to resume breakpoint transmission, establish a data cache and perform flow control, and monitor transmission quality in real time. It is understood that other methods for data processing and transmission can also be used, such as blockchain or cloud storage technologies, which are not limited here.

[0065] S106. Receive the electricity usage identification result returned by the backend server, and when it is determined based on the electricity usage identification result that the target resident has engaged in electric vehicle charging, send an early warning message to the property management terminal.

[0066] The power usage identification result refers to the backend server's analysis and judgment of power usage behavior. Warning information indicates abnormal event information that requires notification to the property management system for handling. The property management terminal refers to the management system terminal used by property management personnel.

[0067] After sending data, the safety monitoring device needs to process the identification results returned by the server. Specifically, the safety monitoring device continuously monitors the identification results returned by the server. When it confirms that an electric vehicle is charging, it generates an early warning message containing resident information, location information, and time information. This is then sent to the property management terminal via a pre-set communication channel, ensuring timely delivery of the warning information.

[0068] It should be noted that the backend server utilizes a large model for electricity usage identification. This model employs a deep learning architecture and comprises three core modules: feature extraction, time series analysis, and behavior recognition. The feature extraction module uses a multi-layer convolutional neural network to extract local features of the current signal. Each convolution kernel has a size of 3×3 and a stride of 1. The time series analysis module employs a bidirectional LSTM network with a hidden layer dimension of 128, capturing the temporal dependencies of electricity usage through forward and backward propagation. The behavior recognition module uses a fully connected layer for feature fusion and classification, employing a softmax function to output a probability score for charging behavior. The model is trained using a cross-entropy loss function and the Adam optimizer for parameter updates. The learning rate is initially set to 0.001 and dynamically adjusted using a cosine annealing strategy. To improve model generalization, dropout (rate = 0.5) and L2 regularization (λ = 0.0001) are applied during training. The model input is a normalized feature vector with a dimension of 256, consisting of 128 time-domain features and 128 frequency-domain features. The output is a binary classification result, indicating whether charging activity is present. When the probability of identifying charging activity exceeds 0.85, the model determines that charging activity is present.

[0069] In the above embodiment, the safety monitoring device primarily collects electricity usage data using a fixed sampling frequency. In practical applications, the sampling frequency can be dynamically adjusted based on electricity usage characteristics during different time periods, optimizing system resource utilization while ensuring effective monitoring. The following supplements the scenarios of this embodiment.

[0070] Following the successful pilot, the property management company upgraded and optimized the safety monitoring device. The new device not only monitors charging behavior but also analyzes residents' electricity usage habits, providing personalized recommendations. For example, the safety monitoring device discovered that Mr. Li's family frequently used high-power appliances during peak hours and, via a mobile app, recommended adjusting usage times. Furthermore, the safety monitoring device sets differentiated warning thresholds for different buildings based on factors such as building age and line conditions, and automatically adjusts the sampling frequency to improve monitoring accuracy.

[0071] After combining the above scenarios, the following is a more detailed description of the process of the method provided by this implementation. Figure 2 , which is another flow chart of the safety monitoring method for home charging of electric vehicles in an embodiment of the present application.

[0072] S201. Obtain electrical equipment information of a target household, including electrical equipment type, equipment rated power, and equipment usage period.

[0073] The electrical equipment type refers to the classification of various electrical appliances used by households, such as air conditioners and washing machines. The equipment rated power represents the nominal power of the electrical equipment under normal operating conditions, measured in watts (W). The equipment usage period represents the typical usage time period for each type of equipment. Electrical equipment information refers to the basic data set that describes household electricity usage characteristics.

[0074] Before monitoring begins, the safety monitoring device must first establish a basic record of residents' electricity usage. Specifically, through user declarations, on-site surveys, or intelligent identification, the safety monitoring device collects basic information about all electrical devices uploaded by property management personnel or residents, including device type, model specifications, and installation location. It also records usage habits and time patterns for each type of device, forming a complete database of electrical equipment information.

[0075] In some embodiments, the collection of electrical device information can be achieved through a variety of methods: Optionally, a form for collecting electrical device information can be designed, household surveys and registrations can be conducted, device nameplate information can be collected, historical electricity usage data can be used to record electricity usage habits, and a device information database can be established and regularly updated. Optionally, a smart meter collection system can be deployed to analyze historical electricity usage data, identify typical electricity usage characteristics, infer device usage patterns, construct electricity usage behavior models, and verify the accuracy of the information. It is understood that other methods can also be used to obtain electrical device information, such as using IoT sensing or artificial intelligence recognition technologies, which are not limited here.

[0076] S202: Determine the target household's electricity load baseline and electricity consumption fluctuation range for each time period based on the electricity consumption equipment information, and construct a basic electricity consumption model.

[0077] The electricity load baseline refers to the typical power consumption curve of a household under normal electricity usage conditions. The electricity consumption fluctuation range represents the reasonable range of variation in actual electricity load relative to the baseline. The basic electricity consumption model is a mathematical model that describes the characteristics of a household's normal electricity consumption behavior.

[0078] After acquiring information about electrical devices, the safety monitoring device needs to build a model of the household's electricity usage characteristics. Specifically, the safety monitoring device first calculates the base load level for different time periods based on the rated power and usage period of each device. It then analyzes the usage characteristics and mutual influence of each device to determine a reasonable load fluctuation range, ultimately constructing a basic model that accurately describes the household's normal electricity usage behavior.

[0079] It should be noted that the basic electricity consumption model inputs historical electricity consumption data sequences (including current values, timestamps) and electricity equipment information (equipment type, rated power) during training. The sliding time window method is used to segment the data, and the clustering algorithm is used to identify and determine typical electricity consumption patterns. The training standards include clustering effect evaluation indicators (such as silhouette coefficient) and pattern discrimination. The basic electricity consumption model contains statistical features such as the mean, standard deviation, and fluctuation range of electricity load in different time periods, as well as a feature template library of typical electricity consumption scenarios. The model structure adopts a multi-level design, including a basic load layer, an equipment feature layer, and a scenario feature layer. By inputting real-time electricity consumption data, the basic electricity consumption model can output the matching score and abnormality judgment of the current electricity consumption status, and make abnormality judgments based on preset thresholds and duration.

[0080] S203: Determine a preset normal threshold and a preset duration for triggering power consumption identification based on the power consumption device information and the basic power consumption model.

[0081] The preset normal threshold refers to the current or power threshold used to identify abnormal power usage. The preset duration indicates the minimum duration of the abnormal state. The trigger for power usage identification refers to the conditions that initiate the identification of abnormal power usage behavior.

[0082] After establishing a basic power usage model, the safety monitoring device needs to set abnormality judgment criteria. Specifically, the safety monitoring device first analyzes the startup and operating characteristics of various electrical appliances. Based on the fluctuation range of the basic power usage model, it determines the threshold level that can effectively distinguish between normal power usage and charging behavior. It also sets reasonable duration requirements based on the charging characteristic curve, forming a complete trigger judgment mechanism.

[0083] In some embodiments, the safety monitoring device will use the sliding time window method to perform statistical analysis on the power load in each time period based on the information of the power-consuming equipment, and obtain the power consumption fluctuation pattern in different time periods; perform cluster analysis on the power consumption fluctuation pattern to obtain multiple typical power consumption scenarios; calculate the current mean and standard deviation of the typical power consumption scenario based on the basic power consumption model; determine the preset normal threshold value based on the current mean and standard deviation; determine the shortest duration to distinguish between normal power consumption and electric vehicle charging behavior based on the electric vehicle charging characteristic curve and the power consumption fluctuation pattern, and obtain the preset duration.

[0084] The sliding time window method is a data analysis method that uses a fixed-length observation window to move across a time series. The power consumption fluctuation pattern represents the regular characteristics of power load changes over time. Typical power usage scenarios refer to types of power usage behaviors with similar power usage characteristics. The current mean is the average current level in a specific scenario. The standard deviation indicates the degree of dispersion of current values. The preset duration represents the minimum time required to determine charging behavior.

[0085] After establishing a basic power usage model, the safety monitoring device needs to determine criteria for abnormal power usage. Specifically, the safety monitoring device first analyzes historical data in segments using a sliding time window, extracting power usage characteristics for each time period. Then, using a clustering algorithm, it groups similar power usage patterns into several typical scenarios. For each typical scenario, the safety monitoring device calculates the current statistical characteristics and, based on the typical curve characteristics of electric vehicle charging, sets appropriate thresholds and duration requirements to establish a comprehensive judgment mechanism.

[0086] S204: Obtain the current value of the target household according to the preset collection frequency to obtain a power consumption data sequence.

[0087] Referring to step S101 , the safety monitoring device collects a sequence of electricity consumption data.

[0088] In some embodiments, the security monitoring device collects the electricity load characteristics of the target household, determines the initial sampling frequency of the current value based on the electricity load characteristics; collects the current value at the initial sampling frequency to obtain the initial electricity sequence, and calculates the signal-to-noise ratio of the initial electricity sequence in real time; when the signal-to-noise ratio is lower than the stable threshold, increases the initial sampling frequency to the corrected sampling frequency so that the signal-to-noise ratio reaches the stable threshold; collects the current value at the corrected sampling frequency to obtain the electricity data sequence.

[0089] The power load characteristic refers to the current variation characteristics exhibited by electrical equipment during operation. The initial sampling frequency represents the data sampling rate set when the safety monitoring device begins data collection. The initial power sequence refers to the raw data sequence collected at the initial frequency. The signal-to-noise ratio represents the ratio of effective signal to noise. The stability threshold represents the minimum signal-to-noise ratio requirement to ensure data quality. The adjusted sampling frequency represents the final adjusted sampling rate.

[0090] Before starting data collection, the safety monitoring device must determine an appropriate sampling strategy. Specifically, the device sets an initial sampling frequency based on the load characteristics of the electrical equipment. After data collection begins, the device monitors data quality in real time and evaluates the sampling effectiveness by calculating the signal-to-noise ratio. If data quality falls short of expectations, the device automatically increases the sampling frequency until the expected data quality standard is met, ensuring that the acquired data sequence accurately reflects the characteristics of electricity consumption.

[0091] The calculation of the signal-to-noise ratio uses the wavelet decomposition method to decompose the signal into 5 scales and calculate the signal energy and noise energy at each scale. Let the wavelet coefficient of the j-th scale be wj,k, then the signal energy is: The noise energy En is estimated by the median absolute deviation of the finest scale coefficients: The signal-to-noise ratio is:

[0092] When SNR < 15dB, the sampling frequency is doubled until SNR ≥ 15dB or the maximum sampling frequency of 2kHz is reached; the sampling frequency is adjusted using a segmented strategy, with each data segment length being 10s to ensure the continuity of data quality.

[0093] S205 : When it is detected that the current value in the power consumption data sequence exceeds a preset normal threshold and the duration exceeds a preset duration, extract multiple time domain feature values ​​of the power consumption data sequence.

[0094] Referring to step S102 , the security monitoring device extracts time domain feature values.

[0095] S206 , converting the electricity consumption data sequence into the frequency domain, extracting the frequency composition and energy distribution of the current signal, and obtaining a plurality of frequency domain eigenvalues.

[0096] Referring to step S103 , the safety monitoring device extracts frequency domain eigenvalues.

[0097] It should be noted that the extraction of time domain eigenvalues ​​and frequency domain eigenvalues ​​can be used in the electricity consumption feature extraction model. The electricity consumption feature extraction model inputs labeled normal electricity consumption and charging behavior data samples during training. Time domain features (mean, variance, peak, etc.) and frequency domain features (main frequency components, energy distribution, etc.) are extracted respectively. Principal component analysis is used to determine the optimal feature combination, and the training criteria are the feature discrimination and information content. The electricity consumption feature extraction model includes feature extraction operators, data standardization parameters, principal component transformation matrix and feature importance weights. The electricity consumption feature extraction model adopts a cascade structure to realize the conversion from raw data to standardized feature vectors. By inputting the real-time collected current sequence, after feature extraction and dimensionality reduction processing, the electricity consumption feature extraction model can output a standardized feature vector for subsequent behavior recognition.

[0098] S207 : Concatenate and integrate the time domain eigenvalues ​​and the frequency domain eigenvalues ​​according to a preset format to obtain a power consumption eigenvector.

[0099] Referring to step S104 , the safety monitoring device constructs a power consumption feature vector.

[0100] In some embodiments, the safety monitoring device will perform data standardization on the time domain eigenvalues, convert the eigenvalues ​​of different dimensions into the same numerical range to obtain the time domain standard value; use the principal component analysis method to extract the main frequency domain characteristic components of the frequency domain eigenvalues, reduce the characteristic space dimension of the frequency domain eigenvalues, and obtain the frequency domain standard value; perform splicing and integration of the time domain standard value and the frequency domain standard value according to a preset splicing order to obtain the electricity consumption characteristic vector.

[0101] Data normalization refers to the process of converting data of different dimensions to a unified scale. Time domain standard values ​​represent standardized time domain feature data. Principal component analysis is a statistical method for extracting key features through dimensionality reduction. Feature space dimension refers to the number of dimensions of a feature vector. Predefined concatenation order refers to a predefined feature combination method.

[0102] After obtaining time-domain and frequency-domain features, the safety monitoring device needs to integrate these features. Specifically, the device first normalizes the time-domain features to make them comparable. It then applies principal component analysis to the frequency-domain features to extract the most representative feature components and reduce data redundancy. Finally, the processed features are combined into a unified feature vector according to a predefined order, providing standardized input for subsequent analysis.

[0103] S208: Encode the electricity usage feature vector to obtain an electricity usage encoding file, and send the electricity usage encoding file to the backend server.

[0104] Referring to step S105 , the security monitoring device sends the power usage code file to the backend server.

[0105] S209: Receive the electricity usage identification result returned by the backend server, and when it is determined based on the electricity usage identification result that the target resident has engaged in electric vehicle charging, send an early warning message to the property management terminal.

[0106] Referring to step S106 , the safety monitoring device will issue an early warning when it identifies that the electric vehicle is charging.

[0107] S210. Obtain the historical violation records of the target resident and the fire safety level of the building unit where the resident resides.

[0108] The "history violation record" refers to a resident's past violations related to charging safety. The "building unit" refers to the specific building unit where the resident resides. The "fire safety level" refers to the completeness of the building's fire protection facilities and the safety risk assessment level.

[0109] After detecting a violation, the security monitoring device assesses the risk level of the incident. Specifically, it first retrieves the resident's historical violation information from the management database, including the number of violations, violation type, and handling results. It also obtains safety level information such as the fire protection facilities and evacuation routes of the building unit where the resident resides, providing a risk assessment basis for subsequent early warning processing.

[0110] S211. Determine the priority level for early warning processing based on historical violation records and fire safety levels.

[0111] The handling priority refers to the urgency level of the warning event. Warning handling refers to the intervention and disposal measures for illegal charging behavior.

[0112] After obtaining risk information, the safety monitoring device needs to determine the processing order of early warnings. Specifically, the safety monitoring device comprehensively evaluates the risk level of the current early warning event based on historical data such as the violation frequency and degree of the household, combined with the fire safety status of the building, and determines the priority of processing accordingly to ensure that high-risk events can be processed in a timely manner.

[0113] Among them, the calculation of the risk level can adopt a weighted scoring method. The weight of historical violation records is 0.4, and the scoring standard is: 0 points for no violation records, 2 points for 1 violation, 4 points for 2 violations, and 6 points for 3 or more violations. The weight of the fire safety level is 0.6, and the scoring standard is: 0 points for Class A, 2 points for Class B, 4 points for Class C, and 6 points for Class D.

[0114] The total score S = 0.4 × violation score + 0.6 × safety level score. The safety monitoring device divides the priority according to the total score: S ≤ 2 is the low priority, 2 < S ≤ 4 is the medium priority, and S > 4 is the high priority. For early warnings of the same priority, they are processed in chronological order. The safety monitoring device automatically re-evaluates the risk level every 24 hours to ensure the timeliness of risk assessment.

[0115] It should be noted that the evaluation of the risk level can also use the early warning risk assessment model. This early warning risk assessment model inputs historical early warning records, processing results, and building safety level data during training. Through statistical analysis, a risk level划分标准 is established, and the training criteria include early warning accuracy and processing timeliness. This early warning risk assessment model includes risk scoring rules, priority judgment logic, and a dynamic adjustment mechanism. The model structure includes a basic scoring module and a dynamic adjustment module. By inputting the current early warning information, household historical records, and building information, this early warning risk assessment model can output the risk level and processing priority of the early warning event to guide the early warning processing process.

[0116] S212. Generate a task processing list based on multiple early warning information and corresponding processing priorities.

[0117] Among them, the early warning information refers to the detailed information of the illegal charging event discovered by the safety monitoring device. The task processing list represents a list of early warning processing tasks sorted by priority.

[0118] After determining the priority of each early warning event, the safety monitoring device needs to generate a standardized task list for processing. Specifically, the safety monitoring device sorts all the early warning events to be processed according to the priority, organizes a detailed task list including household information, location information, violation type, processing suggestions, etc., and dynamically updates the task sorting according to the real-time situation to provide clear work guidance for property management personnel.

[0119] In some embodiments, the security monitoring device collects the target resident's household information and warning information to generate warning details; builds a warning tracking file based on the warning details; and updates the warning tracking file based on the processing time and processing results uploaded by the property staff.

[0120] Resident information refers to basic user identity and contact information. Warning details refer to data records containing complete warning-related information. Warning tracking files are used to record the handling process and results of warning events.

[0121] After issuing an alert, the security monitoring device needs to track the handling of the alert. Specifically, the device first collects various information related to the alert, including basic information about the resident and characteristics of the violation, to create an alert event file. As property management personnel proceed with the handling process, the security monitoring device continuously records the time points and handling measures during the handling process, and updates the file information based on the final handling results, forming a complete record of the alert handling process.

[0122] In the embodiment of the present application, due to the adoption of an intelligent monitoring method based on electricity consumption data analysis, combined with multi-dimensional analysis of time domain characteristics and frequency domain characteristics, and the establishment of a dynamic electricity consumption behavior model, it is possible to accurately identify and warn of illegal charging behaviors. At the same time, priority management and early warning tracking mechanisms are used to ensure timely and effective processing, effectively solving the problems of low efficiency and inability to detect violations in a timely manner in traditional manual inspection methods, thereby achieving precise and standardized management of charging safety in residential areas, improving the level of community safety management, reducing fire hazards, and providing reliable decision-making basis and management tools for property management.

[0123] The following describes the safety monitoring device in the embodiment of the present invention from the perspective of hardware processing. Figure 3 , is a schematic diagram of the physical device structure of the safety monitoring device in an embodiment of the present application.

[0124] It should be noted that Figure 3 The structure of the safety monitoring device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0125] like Figure 3As shown, the safety monitoring device includes a central processing unit (CPU) 301, which can perform various appropriate actions and processes based on programs stored in a read-only memory (ROM) 302 or programs loaded from a storage unit 308 into a random access memory (RAM) 303, such as executing the methods described in the above embodiments. Various programs and data required for system operation are also stored in RAM 303. CPU 301, ROM 302, and RAM 303 are connected to each other via a bus 304. An input / output (I / O) interface 305 is also connected to bus 304.

[0126] The following components are connected to the I / O interface 305: an input section 306 including an audio input device, a push button switch, and the like; an output section 307 including a liquid crystal display (LCD), an audio output device, an indicator light, and the like; a storage section 308 including a hard disk and the like; and a communication section 309 including a network interface card such as a LAN (Local Area Network) card or a modem. The communication section 309 performs communication processing via a network such as the Internet. A drive 310 is also connected to the I / O interface 305 as needed. Removable media 311, such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory, is installed in the drive 310 as needed, so that computer programs read therefrom can be installed into the storage section 308 as needed.

[0127] In particular, according to an embodiment of the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present invention includes a computer program product that includes a computer program carried on a computer-readable medium, the computer program including a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network via the communication section 309 and / or installed from a removable medium 311. When the computer program is executed by the central processing unit (CPU) 301, the various functions defined in the present invention are performed.

[0128] It should be noted that specific examples of computer-readable storage media may include, but are not limited to, an electrical connection having one or more conductors, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0129] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present invention. Each box in the flowchart or block diagram can represent a module, program segment, or part of the code, and the above-mentioned module, program segment, or part of the code contains one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings.

[0130] Specifically, the safety monitoring device of this embodiment includes a processor and a memory. The memory stores a computer program. When the computer program is executed by the processor, the safety monitoring method for home charging of electric vehicles provided in the above embodiment is implemented.

[0131] As another aspect, the present invention further provides a computer-readable storage medium, which may be included in the safety monitoring device described in the above embodiments, or may exist independently and not incorporated into the safety monitoring device. The storage medium carries one or more computer programs, which, when executed by a processor of the safety monitoring device, enable the safety monitoring device to implement the safety monitoring method for home charging of electric vehicles provided in the above embodiments.

[0132] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

[0133] As used in the above embodiments, the term “when…” may be interpreted to mean “if…” or “after…” or “in response to determining…” or “in response to detecting…”, depending on the context. Similarly, the phrases “upon determining…” or “if (stated condition or event) is detected” may be interpreted to mean “if determining…” or “in response to determining…” or “upon detecting (stated condition or event)” or “in response to detecting (stated condition or event)”, depending on the context.

[0134] Those skilled in the art will appreciate that all or part of the process steps in the above-described method embodiments can be implemented by a computer program instructing the relevant hardware. The program can be stored in a computer-readable storage medium, and when executed, the program can include the process steps in the above-described method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM or random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A safety monitoring method for charging an electric vehicle at home, characterized in that: Applied to a safety monitoring device, the method comprises: The current value of the target household is obtained according to the preset collection frequency to obtain a power consumption data sequence; When it is detected that the current value in the power consumption data sequence exceeds a preset normal threshold value and the duration exceeds a preset duration, extracting a plurality of time domain feature values ​​of the power consumption data sequence; Converting the power consumption data sequence into the frequency domain, extracting the frequency composition and energy distribution of the current signal, and obtaining a plurality of frequency domain eigenvalues; The time domain eigenvalues ​​and the frequency domain eigenvalues ​​are spliced ​​and integrated according to a preset format to obtain an electricity consumption eigenvector; Encoding the power consumption feature vector to obtain a power consumption encoding file, and sending the power consumption encoding file to a back-end server; The power usage identification result returned by the back-end server is received, and when it is determined according to the power usage identification result that the target resident has an electric vehicle charging behavior, an early warning message is sent to the property management terminal.

2. The method according to claim 1, characterized in that The step of acquiring the current value of the target household according to the preset collection frequency to obtain the electricity consumption data sequence specifically includes: Collecting the power load characteristics of the target household, and determining the initial sampling frequency of the current value according to the power load characteristics; Collecting current values ​​at the initial sampling frequency to obtain an initial power usage sequence, and calculating a signal-to-noise ratio of the initial power usage sequence in real time; When the signal-to-noise ratio is lower than a stable threshold, increasing the initial sampling frequency to a modified sampling frequency so that the signal-to-noise ratio reaches the stable threshold; The current value is collected at the modified sampling frequency to obtain a power consumption data sequence.

3. The method according to claim 1, characterized in that The step of splicing and integrating the time domain eigenvalues ​​and the frequency domain eigenvalues ​​according to a preset format to obtain the power consumption eigenvector specifically includes: Performing data standardization processing on the time domain characteristic values, converting characteristic values ​​of different dimensions into the same numerical range, and obtaining time domain standard values; A principal component analysis method is used to extract the main frequency domain characteristic components of the frequency domain characteristic values, reduce the characteristic space dimension of the frequency domain characteristic values, and obtain a frequency domain standard value; The time domain standard value and the frequency domain standard value are spliced ​​and integrated according to a preset splicing order to obtain a power consumption feature vector.

4. The method according to claim 1, characterized in that Before the step of acquiring the current value of the target household according to the preset collection frequency to obtain the electricity consumption data sequence, the method further includes: Obtaining the target household's electrical equipment information including the type of electrical equipment, the rated power of the equipment, and the equipment usage period; Determine the power load baseline and power consumption fluctuation range of the target household in each time period according to the power consumption equipment information, and construct a basic power consumption model; According to the power consumption equipment information and the basic power consumption model, a preset normal threshold and a preset duration for triggering power consumption identification are determined.

5. The method according to claim 4, characterized in that The step of determining a preset normal threshold and a preset duration for triggering power consumption identification according to the power consumption device information and the basic power consumption model specifically includes: Based on the power consumption equipment information, a sliding time window method is used to perform statistical analysis on the power consumption load in each time period to obtain power consumption fluctuation patterns in different time periods; Performing cluster analysis on the power consumption fluctuation pattern to obtain multiple typical power consumption scenarios; Calculate the current mean and standard deviation of the typical power usage scenario based on the basic power usage model; Determining a preset normal threshold value according to the current mean value and the standard deviation; According to the electric vehicle charging characteristic curve and the power consumption fluctuation pattern, the shortest duration for distinguishing normal power consumption from electric vehicle charging behavior is determined to obtain a preset duration.

6. The method according to claim 1, characterized in that After the step of receiving the power usage identification result returned by the back-end server and, when it is determined according to the power usage identification result that the target resident has an electric vehicle charging behavior, sending an early warning message to the property management terminal, the method further includes: Obtain the historical violation records of the target resident and the fire safety rating of the building unit; Determining a processing priority for early warning processing based on the historical violation records and the fire safety level; Based on multiple warning information and corresponding processing priorities, a task processing list is generated.

7. The method according to claim 6, characterized in that After the step of generating a task processing list based on the plurality of warning information and the corresponding processing priorities, the method further comprises: Collecting household information and warning information of the target household and generating warning details; Building a warning tracking file based on the warning details; Based on the processing time and processing results uploaded by the property personnel, the early warning tracking file is updated.

8. A safety monitoring device, characterized in that: The safety monitoring device includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the safety monitoring device to execute the method described in any one of claims 1-7.

9. A computer-readable storage medium comprising instructions, characterized in that: When the instruction is executed on a safety monitoring device, the safety monitoring device is caused to execute the method according to any one of claims 1 to 7.

10. A computer program product, characterized in that When the computer program product runs on a safety monitoring device, the safety monitoring device is enabled to execute the method according to any one of claims 1 to 7.

Images