A home security control method and system based on big data analysis

Abstract

This application relates to a home security control method and system based on big data analysis, belonging to the technical field of home security control. The control method includes: obtaining the original sensor data stream in the home environment; preprocessing the original sensor data stream to obtain a synchronous calibration matrix; calculating the temperature change rate, smoke concentration accumulation, current harmonic components, and temperature-current correlation coefficient to obtain a feature matrix; performing anomaly detection on the temperature change rate and outputting a temperature isolation score; calculating a smoke risk coefficient based on the smoke concentration accumulation; reconstructing the current harmonic components and calculating the current reconstruction error; determining an environmental risk coefficient according to the door and window status; performing weighted fusion on the above data to calculate a dynamic risk index; and sending corresponding security control instructions according to the dynamic risk index. This application can efficiently and accurately identify potential security risks, adjust control strategies according to real-time data, and provide a more efficient and secure home environment guarantee.

Description

Technical Field

[0001] This application relates to the technical field of home security control, and in particular, to a home security control method and system based on big data analysis. Background Art

[0002] With the continuous development of society and the improvement of people's living standards, the issue of home environment security has become increasingly important. Home security not only includes traditional security issues such as anti-theft and fire prevention, but also covers various factors such as electrical faults, gas leaks, and air quality monitoring. Due to the complexity and diversity of the living environment, the design of home security control systems increasingly tends to be intelligent and automated to improve the response speed and accuracy, reduce manual intervention, and enhance the reliability and security of the system.

[0003] Currently, common home security control systems have limitations in data processing. Most rely on single-sensor data for judgment and cannot comprehensively monitor potential security risks. Especially when facing a complex dynamic environment (such as environmental changes or the interweaving of multiple risk factors), they cannot achieve multi-dimensional collaborative monitoring and lack intelligent decision support, resulting in misjudgment or missed judgment in case of some emergencies and poor security. Summary of the Invention

[0004] To provide a more efficient and secure home environment guarantee, this application provides a home security control method and system based on big data analysis.

[0005] In the first aspect, this application provides a home security control method based on big data analysis, adopting the following technical solution:

[0006] A home security control method based on big data analysis, the control method includes:

[0007] Obtain the original sensor data stream in the home environment; wherein, the original sensor data stream includes real-time temperature data, smoke concentration data, real-time current data, and door / window status;

[0008] Preprocess the original sensor data stream to obtain a synchronous calibration matrix;

[0009] Calculate the temperature change rate, smoke concentration accumulation, current harmonic components, and temperature-current correlation coefficient according to the synchronous calibration matrix to obtain a feature matrix;

[0010] Perform anomaly detection on the temperature change rate through the Isolation Forest algorithm and output the temperature isolation score;

[0011] Based on a preset smoke risk mapping relationship, calculate the smoke risk coefficient according to the smoke concentration accumulation;

[0012] Reconstruct the current harmonic components through an autoencoder model and calculate the current reconstruction error;

[0013] Based on a preset environmental risk mapping relationship, determine the environmental risk coefficient according to the door and window status;

[0014] Perform weighted fusion on the temperature isolation fraction, smoke risk coefficient, current reconstruction error, and environmental risk coefficient to calculate the dynamic risk index;

[0015] Send corresponding safety control instructions according to the dynamic risk index.

[0016] By adopting the above technical solutions, through the real-time collection, preprocessing, and analysis of multi-source data, combined with methods such as multi-dimensional feature extraction, dynamic risk calculation, and intelligent decision-making control, the automatic monitoring and real-time response of home security are realized. Through the application of big data analysis and intelligent algorithms, potential safety risks can be efficiently and accurately identified, and the control strategy can be adjusted according to real-time data to ensure that the home security system can make reasonable responses according to real-time risks and protect users from risks such as fires or electrical failures.

[0017] Optionally, the steps of preprocessing the original sensor data stream to obtain a synchronous calibration matrix include:

[0018] Align the timestamps of the original sensor data stream to obtain synchronized and calibrated real-time temperature data, smoke concentration data, and real-time current data;

[0019] Establish a temperature compensation function based on the thermodynamic diffusion model and perform environmental compensation on the synchronized and calibrated real-time temperature data;

[0020] Calculate the dynamic baseline based on the moving average method of a sliding window and correct the baseline drift of the synchronized and calibrated smoke concentration data;

[0021] Perform adaptive filtering on the synchronized and calibrated real-time current data based on wavelet transform;

[0022] Combine the ordered discrete coding corresponding to the door and window status to construct a synchronous calibration matrix.

[0023] By adopting the above technical solutions, the efficient preprocessing of multi-source sensor data is realized. By using wavelet transform and Fourier transform to extract key signal features and combining advanced compensation and filtering technologies, problems such as data noise interference, environmental change influence, and baseline drift are solved. Finally, through multi-modal data fusion and a circular buffer storage structure, an efficient, stable, and continuous data matrix is provided, laying a solid foundation for subsequent intelligent analysis and decision-making control.

[0024] Optionally, the steps of performing anomaly detection on the temperature change rate through the isolation forest algorithm and outputting the temperature isolation score include:

[0025] Perform normalization processing on the temperature change rate to obtain a normalized temperature change rate sequence;

[0026] Cache the data within a preset historical period in the normalized temperature change rate sequence as a training set;

[0027] Dynamically configure the number of trees according to the training set, construct a set of isolation trees with the maximum depth, randomly select features and split values during each split, recursively split the data until a single sample is isolated or the depth limit is reached, and obtain an isolation forest model;

[0028] Based on the isolation forest model, calculate the average path length of the real-time data in the normalized temperature change rate sequence in each tree, determine a preset weighting coefficient according to the real-time temperature data, and calculate the temperature isolation score.

[0029] By adopting the above technical solution, anomaly detection is performed on the temperature change rate based on the isolation forest algorithm, the anomaly of the data is evaluated by using the path lengths of multiple trees, and the detection accuracy is improved by adjusting the dynamic weighting coefficient. Through steps such as normalizing the temperature change rate, dynamically training the isolation forest model, and calculating the anomaly score in real time, this technical solution can effectively identify temperature changes caused by emergencies such as fires and equipment failures, and provide accurate anomaly warning capabilities for the home security system.

[0030] Optionally, it further includes the training steps of the autoencoder model, and the training steps include:

[0031] Collect historical sample data; among them, the historical sample data includes current harmonic component data under historical normal working conditions;

[0032] Preprocess the historical sample data and divide it into training samples and reserved samples;

[0033] Train the pre-constructed stacked convolutional autoencoder based on the training samples, optimize the model parameters, and obtain the autoencoder model;

[0034] Test the autoencoder model based on the reserved samples and correct the model parameters until the loss function of the model meets the preset conditions or the number of model iterations reaches the preset number of times, and obtain the trained autoencoder model.

[0035] By adopting the above technical solution, the high - efficient learning and reconstruction of current harmonic components are carried out based on the stacked convolutional auto - encoder (SCAE), which can accurately capture the normal mode in the current signal. Through standardization processing, the division of training and reserved samples, and multiple rounds of training and calibration of the model, an auto - encoder model with excellent performance is finally obtained. This model can accurately calculate the reconstruction error, which serves as the basis for electrical fault detection, has efficient and stable fault recognition capabilities, and can adapt to the changes of different devices.

[0036] Optionally, the steps of weighted - fusing the temperature isolation fraction, smoke risk coefficient, current reconstruction error, and environmental risk coefficient to calculate the dynamic risk index include:

[0037] Receiving the basic weight configuration input by the user;

[0038] Determining the corresponding current risk mode according to the temperature - current correlation coefficient in the feature matrix;

[0039] Optimally adjusting the basic weight configuration according to the current risk mode to obtain an optimized weight configuration;

[0040] According to the optimized weight configuration, combining the temperature isolation fraction, smoke risk coefficient, current reconstruction error, and environmental risk coefficient for weighted fusion to calculate the corresponding dynamic risk index.

[0041] By adopting the above technical solution, combined with the dynamic correlation analysis of temperature and current, the weights of each feature can be adjusted in real - time, and then the dynamic risk index can be accurately calculated. This technical solution has strong adaptability, can cope with different risk modes (such as electrical overload and environmental fire), improves the accuracy and response speed of risk criteria, and can adapt to complex and changeable home security scenarios.

[0042] Optionally, the steps of sending the corresponding safety control instruction according to the dynamic risk index include:

[0043] Determining the corresponding dynamic risk level according to the dynamic risk index;

[0044] Based on the preset policy mapping table, determining the corresponding safety control strategy according to the dynamic risk level;

[0045] Generating the corresponding safety control instruction according to the safety control strategy and sending it to the home control terminal.

[0046] By adopting the above technical solution, the dynamic risk index is combined with the risk level, and corresponding safety control strategies are selected according to different risk levels, and then safety control instructions are generated and automatically executed. This technical solution has a high degree of automation and real-time response capabilities, and can flexibly respond to different security risk scenarios, such as emergencies like electrical overload or fire. Through the preset policy mapping table, the system can accurately judge and take corresponding measures to ensure the safety of the home environment under different risk conditions, and optimize the response speed and accuracy of safety management.

[0047] In a second aspect, the present application provides a home security control system based on big data analysis, adopting the following technical solution:

[0048] A home security control system based on big data analysis, the control method includes:

[0049] A data acquisition module, configured to acquire the original sensor data stream in the home environment; wherein, the original sensor data stream includes real-time temperature data, smoke concentration data, real-time current data, and the state of doors and windows;

[0050] A data preprocessing module, configured to preprocess the original sensor data stream to obtain a synchronous calibration matrix;

[0051] A feature matrix generation module, configured to calculate the temperature change rate, smoke concentration accumulation, current harmonic components, and temperature-current correlation coefficient according to the synchronous calibration matrix to obtain a feature matrix;

[0052] A temperature isolation module, configured to perform anomaly detection on the temperature change rate through the isolation forest algorithm and output a temperature isolation score;

[0053] A smoke risk calculation module, configured to calculate a smoke risk coefficient based on a preset smoke risk mapping relationship according to the smoke concentration accumulation;

[0054] A current reconstruction module, configured to reconstruct the current harmonic components through an autoencoder model and calculate the current reconstruction error;

[0055] An environmental risk determination module, configured to determine an environmental risk coefficient based on a preset environmental risk mapping relationship according to the state of doors and windows;

[0056] A dynamic risk assessment module, configured to perform weighted fusion on the temperature isolation score, smoke risk coefficient, current reconstruction error, and environmental risk coefficient to calculate a dynamic risk index;

[0057] A safety control module, configured to send corresponding safety control instructions according to the dynamic risk index.

[0058] Optionally, the data preprocessing module includes:

[0059] A synchronization calibration module, configured to perform timestamp alignment on the original sensor data stream to obtain real-time temperature data, smoke concentration data, and real-time current data after synchronization calibration;

[0060] An environmental compensation module, configured to establish a temperature compensation function based on a thermodynamic diffusion model and perform environmental compensation on the real-time temperature data after synchronization calibration;

[0061] A baseline drift correction module, configured to calculate a dynamic baseline based on a moving average method with a sliding window and perform baseline drift correction on the smoke concentration data after synchronization calibration;

[0062] A filtering module, configured to perform adaptive filtering on the real-time current data after synchronization calibration based on wavelet transform;

[0063] A calibration matrix construction module, configured to construct a synchronization calibration matrix by combining the ordered discrete encoding corresponding to the door and window states.

[0064] In a third aspect, the present application provides a computer device, adopting the following technical solution:

[0065] A computer device includes a memory, a processor, and a computer program stored on the memory. The processor executes the computer program to implement the steps of the method as described in the first aspect.

[0066] In a fourth aspect, the present application provides a computer-readable storage medium, adopting the following technical solution:

[0067] A computer-readable storage medium stores a computer program that can be loaded and executed by a processor to implement any one of the methods in the first aspect.

[0068] In summary, the present application includes at least one of the following beneficial technical effects: By accurately identifying potential risks, such as fires, electrical overloads, etc., and taking timely safety measures based on real-time risk assessment, it realizes automated and intelligent home safety control, greatly improves the intelligent level of home safety management, has significant practical significance and application value, especially in improving home safety protection and emergency response capabilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 is the first process schematic diagram of a home safety control method based on big data analysis in one embodiment of the present application.

[0070] Figure 2 is the second process schematic diagram of a home safety control method based on big data analysis in one embodiment of the present application.

[0071] Figure 3It is the third process schematic diagram of a home security control method based on big data analysis in one embodiment of the present application.

[0072] Figure 4 It is the fourth process schematic diagram of a home security control method based on big data analysis in one embodiment of the present application.

[0073] Figure 5 It is the fifth process schematic diagram of a home security control method based on big data analysis in one embodiment of the present application.

[0074] Figure 6 It is the sixth process schematic diagram of a home security control method based on big data analysis in one embodiment of the present application. Detailed implementation manners

[0075] In order to make the purpose, technical solutions and advantages of the present application clearer and more understandable, the following will further elaborate on the present application in conjunction with the attached Figures 1-6 drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0076] An embodiment of the present application discloses a home security control method based on big data analysis.

[0077] Referring to Figure 1 , a home security control method based on big data analysis, the control method includes:

[0078] Step S101, obtaining the original sensor data stream in the home environment;

[0079] Among them, the original sensor data stream includes real-time temperature data, smoke concentration data, real-time current data and door / window status;

[0080] Specifically, a variety of sensor devices are pre-arranged in the home environment. These sensors form the perception basis of the home security system. Mainly through the sensors installed in high-risk areas such as the kitchen and living room, environmental data is monitored, ensuring the real-time acquisition of multi-dimensional data in the home environment and providing complete and comprehensive original data support for subsequent analysis.

[0081] Exemplarily, the core sensors include: temperature sensors (deployed in fire-prone areas such as the kitchen and living room), smoke concentration sensors (for detecting PM2.5 and CO concentrations), current monitors (for monitoring the real-time current of high-power electrical appliances), door / window opening / closing state sensors (for detecting the door / window status), and so on.

[0082] Step S102, preprocessing the original sensor data stream to obtain a synchronous calibration matrix;

[0083] Among them, since there may be timing deviations in the data of different sensors, the clocks of each sensor can be calibrated through the Network Time Protocol (NTP) to ensure the time synchronization of all sensor data. At the same time, environmental compensation is performed on the temperature sensor, baseline drift correction is performed on the smoke sensor data, and the sliding window moving average method is used to filter out instantaneous noise. Through the timing calibration and noise processing of the data, high-quality synchronized data is obtained, providing accurate basic data for subsequent analysis.

[0084] Step S103, calculate the temperature change rate, smoke concentration accumulation, current harmonic components, and temperature-current correlation coefficient according to the synchronization calibration matrix to obtain a feature matrix;

[0085] Among them, multi-dimensional features are extracted from the synchronized calibration data matrix, such as the temperature change rate, the accumulation of smoke concentration, the harmonic components of the current, and the Pearson correlation coefficient between temperature and current. These features help to identify potential fires or other dangerous situations.

[0086] In some embodiments, the temperature change rate can be obtained by calculating the temperature difference between the current time and the previous 60 seconds; the accumulation of smoke concentration within 5 minutes can be used to judge the persistence of smoke concentration; the harmonic components of the current can be extracted through Fourier transform to identify whether there is an overload phenomenon; the correlation between temperature and current helps to judge whether there is a risk of equipment failure or electrical fire and can be used for subsequent weight adjustment.

[0087] Step S104, perform anomaly detection on the temperature change rate through the Isolation Forest algorithm and output the temperature isolation score;

[0088] Among them, the temperature isolation score is generated by the anomaly detection of the temperature change rate by the Isolation Forest algorithm, and its essence is an anomaly quantification index of real-time temperature data. The Isolation Forest algorithm is a tree-based outlier detection method that can identify abnormal data in the temperature change rate by constructing multiple trees. An abnormal temperature change rate may be an indication of a fire or equipment failure.

[0089] It should be noted that if the temperature change rate suddenly exceeds the set threshold, the temperature isolation score output by the algorithm is relatively high, indicating that the data may be abnormal and further inspection or response is required.

[0090] Step S105, based on the preset smoke risk mapping relationship, calculate the smoke risk coefficient according to the smoke concentration accumulation;

[0091] Among them, the smoke concentration accumulation is calculated by detecting the change in smoke concentration per unit time. When the cumulative value of smoke concentration exceeds a specific threshold, the system determines a relatively high smoke risk, providing an accurate decision-making basis for the system by effectively distinguishing the risk levels of different smoke concentrations.

[0092] Exemplarily, the preset smoke risk mapping relationship can be configured and adjusted in advance. For example, if the smoke concentration exceeds 200 mg·min / m³, the smoke risk coefficient can be set to 1.0, indicating a high risk; if it is between 50 and 200 mg·min / m³, the risk coefficient is set to 0.5; if it is below 50, it is a low risk.

[0093] Step S106: Reconstruct the current harmonic components through the autoencoder model and calculate the current reconstruction error.

[0094] Among them, the reconstruction error is used to reflect the abnormal degree of the current waveform. Essentially, it is the deviation of the characteristics of the current data from quantization. The autoencoder model is used to learn the normal mode of the current harmonic components and evaluate the abnormal degree of the current waveform by calculating the reconstruction error between the actual input and the model output. If the reconstruction error is large, it indicates that the current waveform is abnormal, which may involve potential risks of electrical appliance failures or electrical fires. For example, during the operation of an electrical appliance, if the proportion of harmonic components suddenly exceeds 15% and the reconstruction error increases, the system will mark it as an overloaded state.

[0095] Step S107: Based on the preset environmental risk mapping relationship, determine the environmental risk coefficient according to the door and window status.

[0096] Among them, the door and window status directly affects the ventilation and smoke diffusion of the home environment. If the doors and windows are closed, the risk of fire is relatively low; if the doors and windows are open, the smoke may spread quickly and the risk is high. Therefore, the environmental risk coefficient needs to be adjusted dynamically.

[0097] Exemplarily, the preset environmental risk mapping relationship can be pre-configured and adjusted according to the actual situation. For example, if the doors and windows are in a closed state, the environmental risk coefficient can be set to 1.0; if the doors and windows are in a partially open state, the risk coefficient can be set to 0.6; if the doors and windows are in a fully open state, the risk coefficient can be set to 0.3.

[0098] Step S108: Perform weighted fusion on the temperature isolation fraction, smoke risk coefficient, current reconstruction error, and environmental risk coefficient to calculate the dynamic risk index.

[0099] Among them, by weighted fusion of multiple features, the safety risk of the home environment is comprehensively evaluated. Different features are assigned different weights according to their importance in specific scenarios, and finally a dynamic risk index is generated to reflect the real-time home safety risk, which helps to make control decisions in a timely manner.

[0100] Step S109: Send corresponding safety control instructions according to the dynamic risk index.

[0101] Among them, according to the calculated dynamic risk index, the system sends corresponding security control instructions through preset policy configurations. For example, if the risk index reaches the high-risk value, the system may take measures such as cutting off the circuit, starting the exhaust fan, and giving an audible and visual alarm; if the risk is relatively low, the system makes minor adjustments, such as simply reducing the electrical power.

[0102] In the above implementation, through the real-time collection, preprocessing, and analysis of multi-source data, combined with methods such as multi-dimensional feature extraction, dynamic risk calculation, and intelligent decision-making control, the automatic monitoring and real-time response of home security are realized. Through the application of big data analysis and intelligent algorithms, potential security risks can be efficiently and accurately identified, and the control strategy can be adjusted according to real-time data to ensure that the home security system can make reasonable responses based on real-time risks, protecting users from risks such as fires or electrical failures.

[0103] Referring to Figure 2 , as an implementation of step S102, the steps of preprocessing the original sensor data stream to obtain a synchronous calibration matrix include:

[0104] Step S201, perform timestamp alignment on the original sensor data stream to obtain synchronized and calibrated real-time temperature data, smoke concentration data, and real-time current data;

[0105] Among them, the original sensor data stream may come from different sensors, and the clocks of these sensors may be asynchronous. To ensure the temporal consistency of the data, first, the clocks of each sensor are calibrated through the Network Time Protocol (NTP) to ensure that the timestamps of all data are consistent. Second, for data with a low sampling rate, cubic spline interpolation is used to supplement the data to make its sampling frequency meet the unified requirements; for data with a high sampling rate, a method of downsampling while retaining the peak-valley value characteristics is used to reduce the data volume while maintaining the main characteristics of the current waveform.

[0106] Exemplarily, if the temperature sensor data is sampled once per minute and the current sensor data is sampled once per second, the cubic spline interpolation method will supplement the real-time temperature data to once per second to ensure that the timestamps of the two data are aligned.

[0107] Step S202, establish a temperature compensation function based on the thermodynamic diffusion model and perform environmental compensation on the synchronized and calibrated real-time temperature data;

[0108] Among them, the temperature sensor is affected by the local environment (such as heat sources, airflows, etc.), which may cause data deviation. By establishing a temperature compensation function based on the thermodynamic diffusion model and combining the data of adjacent sensors, environmental compensation is performed on the real-time temperature data. The compensated real-time temperature data is further smoothed by a Kalman filter to avoid inaccurate results caused by sudden data fluctuations.

[0109] Exemplarily, assume that the temperature sensor is affected by a heat source and causes a high reading. The compensation value of the adjacent sensor can be calculated through a thermodynamic diffusion model, and then the Kalman filter is used to smooth the compensated data to ensure that the real-time temperature data is more stable and reliable.

[0110] Step S203, calculate the dynamic baseline based on the sliding window moving average method, and correct the baseline drift of the smoke concentration data after synchronous calibration;

[0111] Among them, the measurement of the smoke sensor may have baseline drift due to factors such as long-term use and environmental changes, resulting in data deviation. Calculate the dynamic baseline through the sliding window moving average method to correct the drift in the smoke concentration data. This method can adjust the baseline value in real time to ensure the stability and accuracy of the measurement results. For example, assume that the reading of a certain smoke sensor gradually increases due to environmental changes. The moving average value can be calculated through a sliding window (for example, every 2 hours), and the concentration value can be corrected after adjusting the baseline.

[0112] Step S204, perform adaptive filtering on the synchronized and calibrated real-time current data based on wavelet transform;

[0113] Among them, the current data is often affected by high-frequency noise, which may lead to inaccurate data. Through wavelet transform, the Daubechies 4 wavelet is used to separate the high-frequency noise and the fundamental wave component, and adaptive filtering is performed on the noise in the current waveform to further improve the data quality. In addition, the fast Fourier transform (FFT) is used to extract the fundamental wave and harmonic components of the current signal for monitoring the working state of electrical equipment. If the proportion of the harmonic component exceeds 15%, it is marked as an overloaded state. For example, for the current waveform of a certain electrical appliance, after removing the high-frequency noise through wavelet transform, the 50Hz fundamental wave and the third harmonic of 150Hz are calculated using FFT. If the amplitude of the 150Hz harmonic exceeds 15% of the 50Hz fundamental wave, the current signal at this time will be marked as potentially overloaded.

[0114] Step S205, combine the ordered discrete coding corresponding to the door and window states to construct a synchronized calibration matrix.

[0115] Among them, the door and window state data is encoded as 0 (corresponding to the door and window closed state), 0.5 (corresponding to the door and window partially open state), 1 (corresponding to the door and window fully open state), and aligned with other sensor data according to the time stamp to generate a data matrix.

[0116] Specifically, different types of sensors (such as temperature, smoke, current, door / window status) provide data in different dimensions. These data are fused to facilitate better comprehensive analysis. By encoding the door / window status as an ordered discrete code, all the data are aligned according to the timestamp and encapsulated into a multi-dimensional data matrix, which is stored in a circular buffer structure to ensure data continuity and real-time performance.

[0117] In the above embodiments, efficient preprocessing of multi-source sensor data is achieved. By using wavelet transform and Fourier transform to extract key signal features and combining advanced compensation and filtering techniques, problems such as data noise interference, environmental change impact, and baseline drift are solved. Finally, through multi-modal data fusion and the circular buffer storage structure, an efficient, stable, and continuous data matrix is provided, laying a solid foundation for subsequent intelligent analysis and decision control.

[0118] Referring to Figure 3 , as an embodiment of step S104, the steps of performing anomaly detection on the temperature change rate through the Isolation Forest algorithm and outputting the temperature isolation score include:

[0119] Step S301, standardize the temperature change rate to obtain a standardized temperature change rate sequence;

[0120] Among them, the temperature change rate reflects how fast the temperature changes over time. Since temperature data are usually affected by factors such as different devices, environments, and sensor accuracies, there may be significant differences in the scales of the data. Therefore, the Z-score standardization method is used to standardize the temperature change rate so that its mean is 0 and the standard deviation is 1, thereby eliminating the scale differences in the data and enabling various data to be compared under the same standard.

[0121] Step S302, cache the data within a preset historical period in the standardized temperature change rate sequence as the training set;

[0122] Among them, to establish an accurate Isolation Forest model, a large amount of data is required for training. This step caches the standardized temperature change rate data within a certain recent time period (such as the data in the most recent 1 hour), providing a reliable training set for training the Isolation Forest model. This historical data set can capture the normal range of the temperature change rate, thereby helping to detect abnormal behaviors. For example, if the standardized data of the temperature change rate sequence is [0.8, 1.2, -0.4,..., 1.0], then by caching the most recent 3600 data points (1 hour of data), a training set is formed.

[0123] Step S303: Dynamically configure the number of trees according to the training set, construct a set of isolation trees with the maximum depth, randomly select features and splitting values during each split, and recursively split the data until a single sample is isolated or the depth limit is reached to obtain an isolation forest model;

[0124] Among them, the isolation forest algorithm is an anomaly detection method based on a tree structure. Each tree recursively splits the data until each data point is isolated. In this step, according to the size of the training set, the number of trees and the maximum depth of each tree are dynamically configured. The number of trees and the depth are the key parameters of the isolation forest. Adjust the number of trees according to the data density to improve the effect of anomaly detection. The split of each tree is random, that is, randomly select features (in this technical solution, the rate of temperature change) and split points during each split.

[0125] Exemplarily, if the training set D_train has 3600 data points, according to the formula n_trees = ⌈100 ⋅ (1 - e^(-0.002 ⋅ |D_train|))⌉, 100 trees may be obtained. In the construction of each tree, randomly select the rate of temperature change for splitting, and recursively divide the data into smaller subsets until all samples are isolated.

[0126] It can be understood that by dynamically configuring the number of trees and the depth, the isolation forest can automatically optimize parameters according to the data density, improving the accuracy of the model for temperature anomaly detection.

[0127] Step S304: Based on the isolation forest model, calculate the average path length of the real-time data in the standardized rate-of-temperature-change sequence in each tree, determine a preset weighting coefficient according to the real-time temperature data, and calculate the temperature isolation score.

[0128] Among them, the path length of each data point in the isolation forest is an important indicator to measure its abnormality. The shorter the path length, the faster the data point is isolated, and the greater the possibility of being an abnormal point. By calculating the path length for each tree and taking the average, the degree of abnormality of the data point in the entire forest can be obtained. This process helps to capture the mutation of the rate of temperature change and further determine whether it is abnormal.

[0129] Exemplarily, assume that the real-time rate of temperature change is 1 °C / s. After passing through all the trees of the isolation forest, calculate the path length of each tree. For example, the path lengths are [3, 4, 2, 5, 3], then the average path length of this data point is 3.4. By calculating the path lengths of multiple trees, the reliability of anomaly detection can be improved, enabling the system to more accurately identify the temperature mutations caused by fires or equipment failures.

[0130] To further enhance the accuracy of the model, the weighting coefficient of the anomaly score is dynamically adjusted according to the absolute value of the temperature in the real-time temperature data. When the temperature value is high, there may be stronger anomaly signals. Therefore, the isolation score is adjusted through the weighting coefficient (for example, when the temperature exceeds 50 °C, the coefficient is 1.2, otherwise it is 1.0) to ensure that important temperature changes can be detected more accurately.

[0131] For example, if the real-time temperature is 60 °C, then according to the preset rule, the weighting coefficient of the temperature isolation score will be set to 1.2. If the calculated average path length of the real-time data is 0.4, then multiplying it by 1.2 can obtain the final temperature isolation score.

[0132] In the above embodiment, the anomaly detection of the temperature change rate is based on the Isolation Forest algorithm. The path lengths of multiple trees are used to evaluate the anomaly of the data, and the detection accuracy is improved by adjusting the dynamic weighting coefficient. Through steps such as standardizing the temperature change rate, dynamically training the Isolation Forest model, and calculating the anomaly score in real time, this technical solution can effectively identify the temperature changes caused by emergencies such as fires and equipment failures, providing accurate anomaly warning capabilities for the home security system.

[0133] Referring to Figure 4 , as an embodiment of the autoencoder model, the training steps of the autoencoder model include:

[0134] Step S401, collecting historical sample data;

[0135] Among them, the historical sample data includes the current harmonic component data under historical normal operating conditions;

[0136] Specifically, the current harmonic component data under normal operating conditions reflects the current harmonic characteristics of the equipment during normal operation and serves as the reference data for subsequent training of the stacked convolutional autoencoder model. The quality and diversity of the historical sample data directly affect the effectiveness of model training. Exemplarily, the current harmonic component data of common household appliances such as water heaters and air conditioners during normal operation can be collected. These data may include the proportions of the fundamental wave, 3rd harmonic, and 5th harmonic, as well as other relevant information.

[0137] Step S402, preprocessing the historical sample data and dividing it into training samples and reserved samples;

[0138] Among them, to ensure the effectiveness of model training, the historical sample data is first preprocessed, and the processing content includes operations such as denoising and standardization. Then the data is divided into training samples and reserved samples. The training samples are used for model training, and the reserved samples are used for model verification and calibration to ensure the model's generalization ability and avoid overfitting.

[0139] Exemplarily, Z-score normalization is performed on the current harmonic data to ensure that different harmonic components have the same scale. At the same time, the data is cut into time series using a 10-second window and divided into multiple training samples (e.g., divided by device type) and reserved samples. The preprocessing and data division ensure the stability and accuracy during model training and testing, and eliminate possible noise and bias in the data through the normalization operation, guaranteeing the effectiveness of the training process.

[0140] Step S403: Train the pre-constructed stacked convolutional autoencoder based on the training samples, optimize the model parameters, and obtain the autoencoder model.

[0141] Specifically, utilize the pre-constructed stacked convolutional autoencoder (SCAE) to optimize the model parameters according to the training samples. The autoencoder automatically extracts features from the input data through convolutional operations for data reconstruction. During the training process, the reconstruction error is used as the loss function, and the model parameters are optimized through the backpropagation algorithm to enhance the model's learning and reconstruction capabilities for current harmonic components. For example, the normal operating condition data of air conditioning equipment can be used to train the model. The model learns the features of different order harmonics through the convolutional layer and gradually optimizes the model parameters to reduce the reconstruction error.

[0142] It can be understood that through the training of the convolutional autoencoder, the model can effectively learn the normal mode of current harmonic components, providing high-precision reconstruction error calculation for subsequent electrical fault detection.

[0143] Step S404: Test the autoencoder model and correct the model parameters based on the reserved samples until the loss function of the model meets the preset conditions or the number of model iterations reaches the preset number, obtaining the trained autoencoder model.

[0144] Specifically, during the training process, the reserved samples are used to verify and test the model. By calculating the reconstruction error of the validation set, the generalization ability of the model is evaluated, and the model parameters are adjusted according to the performance of the loss function until the preset conditions are met or the set maximum number of iterations is reached. This process can effectively prevent overfitting and optimize the performance of the model.

[0145] Exemplarily, in the later stage of training, the reserved samples are used to test the reconstruction error of the model, and parameters such as the learning rate and the number of training epochs of the model are adjusted according to the error trend to ensure that the finally obtained model can accurately process the current harmonic data. Through testing and correction, the model can be adjusted and optimized according to the actual data, ensuring that it has strong generalization ability and low error, thereby improving the accuracy and robustness of electrical fault detection.

[0146] In the above embodiments, based on the stacked convolutional autoencoder (SCAE), efficient learning and reconstruction of current harmonic components are carried out, which can accurately capture the normal mode in the current signal. Through standardization processing, division of training and reserved samples, and multiple rounds of training and correction of the model, an autoencoder model with excellent performance is finally obtained. This model can accurately calculate the reconstruction error, which is used as the basis for electrical fault detection, has efficient and stable fault identification capabilities, and can adapt to the changes of different devices.

[0147] Referring to Figure 5 , as an embodiment of step 108, the steps of weighted fusion of temperature isolation fraction, smoke risk coefficient, current reconstruction error, and environmental risk coefficient to calculate the dynamic risk index include:

[0148] Step S501, receiving the basic weight configuration input by the user;

[0149] Among them, the user inputs the initially configured feature weights according to the system requirements. This basic weight configuration reflects the initial importance of each input feature in the dynamic risk assessment, and the basic weight configuration provides a starting point for subsequent dynamic adjustment.

[0150] Exemplarily, the user may set the initial weights according to the actual situation. For example, the weight of the current reconstruction error is 0.3, the weight of the smoke risk coefficient is 0.2, etc. These weights will reflect the importance of each feature in the initial calculation of the dynamic risk index.

[0151] Step S502, determining the corresponding current risk mode according to the temperature-current correlation coefficient in the feature matrix;

[0152] Among them, according to the correlation coefficient between temperature and current, the system judges the current dominant risk mode. By calculating the correlation coefficient between temperature and current signals, the system can real-time capture whether there is an electrical overload mode or an environmental fire mode. This judgment is based on the dynamically changing temperature and current data to adjust the risk mode in real time.

[0153] Exemplarily, during the operation of the system, the correlation coefficient of temperature time-series data and current harmonic data will be calculated. For example, if the calculated correlation coefficient is higher than 0.7, the system will determine it as the electrical overload mode; if the correlation coefficient is lower than 0.7 and lasts for 5 minutes, it will switch to the environmental fire mode.

[0154] Step S503, optimizing and adjusting the basic weight configuration according to the current risk mode to obtain the optimized weight configuration;

[0155] Once the current risk mode (electrical overload or environmental fire) is determined, the system will optimize and adjust the basic weight configuration according to this mode. Different risk modes have different focuses on features, so it is necessary to adjust the weight values of each feature. For example, in the electrical overload mode, the current reconstruction error may require a higher weight, while in the environmental fire mode, the smoke risk coefficient may occupy a greater weight.

[0156] Exemplarily, when the system detects the electrical overload mode, the weight of the current reconstruction error may increase from 0.3 in the basic configuration to 0.5, and the weight of the smoke risk coefficient may be reduced to 0.1; while in the environmental fire mode, the weight of the smoke risk coefficient may increase to 0.4, and the weight of the current reconstruction error decreases.

[0157] Step S504, according to the optimized weight configuration, combine the temperature isolation fraction, smoke risk coefficient, current reconstruction error, and environmental risk coefficient for weighted fusion, and calculate the corresponding dynamic risk index.

[0158] Among them, using the optimized weight configuration, the system performs weighted fusion on each feature to obtain the final dynamic risk index. Each feature (such as temperature isolation fraction, smoke risk coefficient, etc.) is weighted according to its weight in the current risk mode, and then a comprehensive risk index is generated.

[0159] Exemplarily, assume that the current risk mode is electrical overload, and the optimized weight configuration is: [0.2, 0.15, 0.5, 0.15]. This weight configuration corresponds to the temperature isolation fraction, smoke risk coefficient, current reconstruction error, and environmental risk coefficient respectively. The system will calculate the dynamic risk index according to these weights.

[0160] In the above embodiments, combined with the dynamic correlation analysis of temperature and current, the weights of each feature can be adjusted in real time, and then the dynamic risk index can be accurately calculated. This technical solution has strong adaptability, can cope with different risk modes (such as electrical overload and environmental fire), improves the accuracy and response speed of the risk criterion, and can adapt to complex and changeable home security scenarios.

[0161] Refer to Figure 6 , as an embodiment of step S109, the steps of sending corresponding safety control instructions according to the dynamic risk index include:

[0162] Step S601, determine the corresponding dynamic risk level according to the dynamic risk index;

[0163] Among them, the calculated dynamic risk index is mapped by the system into a preset risk level system. This risk level system usually has multiple levels, such as low risk, medium risk, and high risk. Each risk level corresponds to different risk response strategies. The dynamic risk index reflects the real-time risk situation, and based on this index, the safety of the home environment can be evaluated in real time.

[0164] Example: Suppose the calculated dynamic risk index is 0.8, which will fall into the high-risk level, indicating that there are significant potential safety threats in the home environment, such as possible fires or electrical overloads. If the dynamic risk index is 0.3, it corresponds to the low-risk level, indicating that the home environment is safe with relatively small risks.

[0165] Step S602: Based on the preset policy mapping table, determine the corresponding safety control policy according to the dynamic risk level.

[0166] Among them, once the dynamic risk level is determined, the system will select the safety control policy corresponding to this risk level according to the preset policy mapping table. The policy mapping table is a predefined rule that determines the safety control measures to be taken under different risk levels. For example, the high-risk level may require immediately cutting off the power supply and starting the fire-fighting equipment, while the low-risk level may only need to reduce the electrical power.

[0167] For example, in a high-risk situation (such as the dynamic risk index ≥ 0.7), the system may select the following safety control policies: cut off the power supply of non-essential electrical appliances; start the exhaust system to remove smoke; emit audible and visual alarms to alert household members. In a low-risk situation (such as the dynamic risk index < 0.4), the system may only take: reduce the power output of electrical appliances; make minor adjustments to the operation of electrical appliances to reduce the load.

[0168] Step S603: Generate corresponding safety control instructions according to the safety control policy and send them to the home control terminal.

[0169] Among them, once the corresponding safety control policy is determined, the system will generate specific safety control instructions according to these policies and send them to the home control terminal (such as smart sockets, home central control systems, fire protection systems, etc.) through the home automation platform. These instructions can trigger actual safety measures, such as cutting off the power supply, starting the exhaust system, starting the alarm system, etc.

[0170] In the above embodiments, the dynamic risk index is combined with the risk level, and the corresponding security control strategy is selected according to different risk levels, and then the security control instruction is generated and automatically executed. This technical solution has a high degree of automation and real-time response capabilities, and can flexibly respond to different security risk scenarios, such as emergencies like electrical overload or fire. Through the preset policy mapping table, the system can accurately judge and take corresponding measures to ensure the safety of the home environment under different risk situations, and optimize the response speed and accuracy of security management.

[0171] The embodiment of the present application also discloses a home security control system based on big data analysis.

[0172] A home security control system based on big data analysis, and the control method includes:

[0173] A data acquisition module, configured to acquire the original sensor data stream in the home environment; wherein, the original sensor data stream includes real-time temperature data, smoke concentration data, real-time current data, and door and window states;

[0174] A data preprocessing module, configured to preprocess the original sensor data stream to obtain a synchronous calibration matrix;

[0175] A feature matrix generation module, configured to calculate the temperature change rate, smoke concentration accumulation, current harmonic components, and temperature-current correlation coefficient according to the synchronous calibration matrix to obtain a feature matrix;

[0176] A temperature isolation module, configured to perform anomaly detection on the temperature change rate through the isolation forest algorithm and output a temperature isolation score;

[0177] A smoke risk calculation module, configured to calculate a smoke risk coefficient based on a preset smoke risk mapping relationship according to the smoke concentration accumulation;

[0178] A current reconstruction module, configured to reconstruct the current harmonic components through an autoencoder model and calculate the current reconstruction error;

[0179] An environmental risk determination module, configured to determine an environmental risk coefficient based on a preset environmental risk mapping relationship according to the door and window states;

[0180] A dynamic risk assessment module, configured to perform weighted fusion on the temperature isolation score, smoke risk coefficient, current reconstruction error, and environmental risk coefficient to calculate a dynamic risk index;

[0181] A security control module, configured to send a corresponding security control instruction according to the dynamic risk index.

[0182] As an embodiment of the data preprocessing module, the data preprocessing module includes:

[0183] A synchronization calibration module, which is used to align the timestamps of the original sensor data stream to obtain synchronized and calibrated real-time temperature data, smoke concentration data, and real-time current data;

[0184] An environmental compensation module, which is used to establish a temperature compensation function based on the thermodynamic diffusion model and perform environmental compensation on the synchronized and calibrated real-time temperature data;

[0185] A baseline drift correction module, which is used to calculate the dynamic baseline based on the moving average method of a sliding window and perform baseline drift correction on the synchronized and calibrated smoke concentration data;

[0186] A filtering module, which is used to perform adaptive filtering on the synchronized and calibrated real-time current data based on wavelet transform;

[0187] A calibration matrix construction module, which is used to construct a synchronized calibration matrix by combining the ordered discrete coding corresponding to the door and window states.

[0188] An in-home safety control system based on big data analysis according to an embodiment of the present application can implement any of the above in-home safety control methods, and the specific working processes of the various modules in the in-home safety control system can refer to the corresponding processes in the above method embodiments.

[0189] In several embodiments provided by the present application, it should be understood that the provided methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for example, the division of a certain module is only a logical function division, and there may be other division methods in actual implementation. For example, multiple modules can be combined or integrated into another system, or some features can be ignored or not executed.

[0190] An embodiment of the present application also discloses a computer device.

[0191] A computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements an in-home safety control method based on big data analysis as described above.

[0192] An embodiment of the present application also discloses a computer-readable storage medium.

[0193] A computer-readable storage medium stores a computer program that can be loaded and executed by a processor to implement any of the in-home safety control methods based on big data analysis as described above.

[0194] Among them, the computer-readable storage medium can be any tangible medium that contains or stores a program, which can be used by or in conjunction with an instruction execution system, apparatus, or device; the program code contained on the computer-readable medium can be transmitted by any appropriate medium, including but not limited to wireless, wire, optical fiber, RF, etc., or any suitable combination of the above.

[0195] It should be noted that in the above embodiments, the descriptions of the various embodiments have their own emphases. For the parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0196] The above are all preferred embodiments of the present application. The protection scope of the present application is not limited thereby. Any feature disclosed in this specification (including the abstract and drawings), unless specifically described, can be replaced by other equivalent or similar-purpose alternative features. That is, unless specifically described, each feature is only an example of a series of equivalent or similar features.

Claims

1. A home security control method based on big data analysis, characterized in that, The control method includes: Obtain the original sensor data stream in the home environment; wherein, the original sensor data stream includes real-time temperature data, smoke concentration data, real-time current data, and the states of doors and windows; Preprocess the original sensor data stream to obtain a synchronous calibration matrix; Calculate the temperature change rate, smoke concentration accumulation, current harmonic components, and Pearson correlation coefficient between temperature and current according to the synchronous calibration matrix to obtain a feature matrix; Perform anomaly detection on the temperature change rate through the Isolation Forest algorithm and output the temperature isolation score; Calculate the smoke risk coefficient based on the preset smoke risk mapping relationship according to the smoke concentration accumulation; Reconstruct the current harmonic components through an autoencoder model and calculate the current reconstruction error; Determine the environmental risk coefficient based on the preset environmental risk mapping relationship according to the states of doors and windows; Perform weighted fusion on the temperature isolation score, smoke risk coefficient, current reconstruction error, and environmental risk coefficient, and calculate the dynamic risk index; Send corresponding safety control instructions according to the dynamic risk index; The steps of calculating the dynamic risk index include: Receive the basic weight configuration input by the user; Determine the corresponding current risk mode according to the Pearson correlation coefficient between temperature and current in the feature matrix; wherein, the current risk mode includes the electrical overload mode or the environmental fire mode; Optimize and adjust the basic weight configuration according to the current risk mode to obtain an optimized weight configuration; According to the optimized weight configuration, perform weighted fusion in combination with the temperature isolation score, smoke risk coefficient, current reconstruction error, and environmental risk coefficient, and calculate the corresponding dynamic risk index.

2. The home security control method based on big data analysis according to claim 1, wherein The steps of preprocessing the original sensor data stream to obtain a synchronous calibration matrix include: Align the timestamps of the original sensor data stream to obtain the synchronized and calibrated real-time temperature data, smoke concentration data, and real-time current data; Establish a temperature compensation function based on the thermodynamic diffusion model and perform environmental compensation on the synchronized and calibrated real-time temperature data; Calculate the dynamic baseline based on the moving average method of a sliding window and correct the baseline drift of the synchronized and calibrated smoke concentration data; Perform adaptive filtering on the synchronized and calibrated real-time current data based on wavelet transform; Construct a synchronous calibration matrix in combination with the ordered discrete coding corresponding to the states of doors and windows.

3. A home security control method based on big data analysis according to claim 1, characterized in that, The steps of performing anomaly detection on the temperature change rate through the Isolation Forest algorithm and outputting the temperature isolation score include: Standardize the temperature change rate to obtain a standardized temperature change rate sequence; Cache the data within a preset historical period in the standardized temperature change rate sequence as a training set; Dynamically configure the number of trees according to the training set, construct a set of isolation trees with the maximum depth, randomly select the temperature change rate feature and the splitting value during each split, and recursively split the data until a single sample is isolated or the depth limit is reached to obtain the Isolation Forest model; Based on the isolation forest model, calculate the average path length of the real-time data in the standardized temperature change rate sequence in each tree, determine a preset weighting coefficient according to the real-time temperature data, and calculate the temperature isolation score.

4. The home security control method based on big data analysis according to claim 1, characterized in that, It further includes the training step of the autoencoder model, and the training step includes: Collect historical sample data; wherein, the historical sample data includes current harmonic component data under historical normal working conditions. Preprocess the historical sample data and divide it into training samples and reserved samples. Train a pre-constructed stacked convolutional autoencoder based on the training samples, optimize the model parameters, and obtain the autoencoder model. Test the autoencoder model based on the reserved samples and correct the model parameters until the loss function of the model meets the preset conditions or the number of model iterations reaches the preset number, and obtain the trained autoencoder model.

5. A home security control method based on big data analysis according to any one of claims 1 to 4, characterized in that, The step of sending a corresponding safety control instruction according to the dynamic risk index includes: Determine the corresponding dynamic risk level according to the dynamic risk index. Based on a preset policy mapping table, determine the corresponding safety control policy according to the dynamic risk level. Generate a corresponding safety control instruction according to the safety control policy and send it to the home control terminal.

6. A home security control system based on big data analysis, characterized in that For implementing a home safety control method according to any one of claims 1-5, the control system includes: A data acquisition module for acquiring the original sensor data stream in the home environment; wherein, the original sensor data stream includes real-time temperature data, smoke concentration data, real-time current data, and door and window states. A data preprocessing module for preprocessing the original sensor data stream to obtain a synchronous calibration matrix. A feature matrix generation module for calculating the temperature change rate, smoke concentration accumulation, current harmonic components, and Pearson correlation coefficient between temperature and current according to the synchronous calibration matrix to obtain a feature matrix. A temperature isolation module for performing anomaly detection on the temperature change rate through the isolation forest algorithm and outputting a temperature isolation score. A smoke risk calculation module for calculating a smoke risk coefficient based on a preset smoke risk mapping relationship according to the smoke concentration accumulation. A current reconstruction module for reconstructing the current harmonic components through the autoencoder model and calculating the current reconstruction error. An environmental risk determination module for determining an environmental risk coefficient based on a preset environmental risk mapping relationship according to the door and window states. A dynamic risk assessment module for performing weighted fusion on the temperature isolation score, smoke risk coefficient, current reconstruction error, and environmental risk coefficient to calculate a dynamic risk index. A safety control module for sending a corresponding safety control instruction according to the dynamic risk index.

7. The home security control system based on big data analysis according to claim 6, wherein The data preprocessing module includes: A synchronous calibration module for aligning the timestamps of the original sensor data stream to obtain synchronously calibrated real-time temperature data, smoke concentration data, and real-time current data. An environmental compensation module for establishing a temperature compensation function based on the thermodynamic diffusion model and performing environmental compensation on the synchronously calibrated real-time temperature data. A baseline drift correction module, which is used to calculate a dynamic baseline based on the sliding window moving average method and correct the baseline drift of the smoke concentration data after synchronous calibration; A filtering module, which is used to perform adaptive filtering on the synchronized and calibrated real-time current data based on wavelet transform; A calibration matrix construction module, which is used to construct a synchronous calibration matrix in combination with the ordered discrete coding corresponding to the door and window states.

8. A computer device, characterized in that: It includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, it implements the method according to any one of claims 1 to 5.

9. A computer-readable storage medium, characterized in that: A computer program is stored that can be loaded and executed by a processor to implement the method according to any one of claims 1 to 5.

Images