High-density area fire early warning method, device and equipment based on multi-source heterogeneous sensor network and storage medium

Abstract

The invention relates to a high-density area fire early warning method based on a multi-source heterogeneous sensor network. The method comprises the steps that a power distribution network line of a target high-density area is monitored based on a target multi-source heterogeneous sensor network to obtain operation state data of the power distribution network line, and the target multi-source heterogeneous sensor network is constructed according to electrical environment characteristics of the target high-density area; carrying out data processing on the current data based on a wavelet packet energy entropy analysis algorithm, carrying out data processing on the voltage data based on a dynamic impedance spectrum correlation analysis algorithm, and carrying out data processing on the cable temperature data based on a space-time correlation algorithm so as to obtain a multi-dimensional feature vector of the power distribution network line; and inputting the multi-dimensional feature vector into a pre-trained early warning index determination model to obtain an early warning index output by the early warning index determination model, and determining an early warning risk level according to the early warning index. By adopting the method, the fire early warning efficiency can be improved.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a high-density area fire early warning method, apparatus, computer equipment, computer-readable storage medium, and computer program product based on a multi-source heterogeneous sensor network. Background Art

[0002] As a special residential form in the process of urbanization, urban villages generally have electrical hazards such as illegal connections, aging lines and three-phase load imbalance. These electrical hazards are one of the common factors causing fires. Therefore, fires caused by electrical hazards can be prevented by early warning.

[0003] In the existing technology, most of the methods are to monitor the threshold of circuit data to achieve early warning of fire.

[0004] However, due to the poor accuracy of this method, the efficiency of the fire early warning method is low. Summary of the Invention

[0005] Based on this, it is necessary to provide a high-density area fire warning method, device, computer equipment, computer-readable storage medium and computer program product based on a multi-source heterogeneous sensor network that can improve the efficiency of fire warning in response to the above technical problems.

[0006] In a first aspect, the present application provides a high-density regional fire early warning method based on a multi-source heterogeneous sensor network, comprising:

[0007] Monitoring the distribution network lines in a target high-density area based on a target multi-source heterogeneous sensor network to obtain operating status data of the distribution network lines. The target multi-source heterogeneous sensor network is constructed based on the electrical environment characteristics of the target high-density area. The operating status data includes current data, voltage data, and cable temperature data.

[0008] The current data is processed based on a wavelet packet energy entropy analysis algorithm, the voltage data is processed based on a dynamic impedance spectrum correlation analysis algorithm, and the cable temperature data is processed based on a spatiotemporal correlation algorithm to obtain a multidimensional feature vector of the distribution network line;

[0009] The multidimensional feature vector is input into a pre-trained early warning index determination model to obtain the early warning index output by the early warning index determination model, and the early warning risk level is determined according to the early warning index. Different early warning risk levels correspond to different early warning strategies.

[0010] In one embodiment, the target multi-source heterogeneous sensing network is based on monitoring the distribution network lines in the target high-density area to obtain the operating status data of the distribution network lines, including: monitoring the distribution network lines in the target high-density area based on the target multi-source heterogeneous sensing network to obtain the initial operating status data of the distribution network line, the initial operating status data including initial current data, initial voltage data and initial cable temperature data; performing noise reduction processing on the initial operating status data through an adaptive wavelet threshold noise reduction algorithm, normalizing the initial operating status data based on the sliding window statistical characteristics, and using a variable time window segmentation strategy to perform time alignment processing on the initial operating status data to obtain the operating status data.

[0011] In one embodiment, the current data is processed based on a wavelet packet energy entropy analysis algorithm, the voltage data is processed based on a dynamic impedance spectrum correlation analysis algorithm, and the cable temperature data is processed based on a time-space correlation algorithm to obtain a multidimensional feature vector of the distribution network line, including: processing the current data based on a wavelet packet energy entropy analysis algorithm to obtain energy entropy of a preset frequency band; processing the voltage data based on a dynamic impedance spectrum correlation analysis algorithm to obtain impedance correlation between low frequency and high frequency; processing the cable temperature data based on a time-space correlation algorithm to obtain an axial temperature gradient and a temperature-current coefficient; and determining the multidimensional feature vector based on the energy entropy of the preset frequency band, the impedance correlation between low frequency and high frequency, the axial temperature gradient, and the temperature-current coefficient.

[0012] In one embodiment, the warning index includes the energy accumulation rate and the temperature rise gradient, and the warning risk level is determined based on the warning index, including: when the energy accumulation rate and the temperature rise gradient meet the first preset condition, determining the warning risk level as the first risk level; wherein the warning strategy executed corresponding to the first risk level is the first warning strategy, and the first warning strategy is to output the first warning information, and the first warning information includes the first predicted warning position and the disposal suggestion for the first predicted warning position.

[0013] In one embodiment, the warning index also includes the cable axial temperature change rate and the contact resistance mutation rate, and the warning risk level is determined based on the warning index, including: when the cable axial temperature change rate and the contact resistance mutation rate meet the second preset condition, determining the warning risk level as the second risk level; wherein the warning strategy executed corresponding to the second risk level is the second warning strategy, and the second warning strategy includes outputting a second warning information, cutting off the power supply to the second predicted warning position, and starting a cooling device for the second predicted warning position, and the second warning information includes the second warning position.

[0014] In one embodiment, the warning index also includes a high-frequency arc characteristic, and determining the warning risk level based on the warning index includes: when the high-frequency arc characteristic meets a third preset condition, determining the warning risk level as a third risk level; wherein the warning strategy executed corresponding to the third risk level is a third warning strategy, and the third warning strategy includes starting fire-fighting equipment for a third warning position, triggering sound and light alarm equipment, and outputting a third warning information, and the third warning information includes the third warning position.

[0015] In a second aspect, the present application also provides a high-density regional fire warning device based on a multi-source heterogeneous sensor network, comprising:

[0016] an execution module, configured to monitor a distribution network line in a target high-density area based on a target multi-source heterogeneous sensing network to obtain operating status data of the distribution network line, wherein the target multi-source heterogeneous sensing network is constructed based on electrical environment characteristics of the target high-density area, and the operating status data includes current data, voltage data, and cable temperature data;

[0017] a processing module, configured to process the current data based on a wavelet packet energy entropy analysis algorithm, process the voltage data based on a dynamic impedance spectrum correlation analysis algorithm, and process the cable temperature data based on a spatiotemporal correlation algorithm, so as to obtain a multidimensional feature vector of the distribution network line;

[0018] The determination module is used to input the multidimensional feature vector into a pre-trained warning index determination model to obtain the warning index output by the warning index determination model, and determine the warning risk level based on the warning index. Different warning risk levels correspond to different warning strategies.

[0019] In a third aspect, the present application further provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the method described in any embodiment of the first aspect are implemented.

[0020] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in any embodiment of the first aspect above.

[0021] In a fifth aspect, the present application further provides a computer program product, comprising a computer program, which, when executed by a processor, implements the steps of the method described in any embodiment of the first aspect above.

[0022] The above-mentioned high-density area fire warning method, device, computer equipment, computer-readable storage medium and computer program product based on a multi-source heterogeneous sensor network first monitors the distribution network line in the target high-density area based on the target multi-source heterogeneous sensor network to obtain the operating status data of the distribution network line. The target multi-source heterogeneous sensor network is constructed according to the electrical environment characteristics of the target high-density area, and the operating status data includes current data, voltage data and cable temperature data; then, the current data is processed based on the wavelet packet energy entropy analysis algorithm, the voltage data is processed based on the dynamic impedance spectrum correlation analysis algorithm, and the cable temperature data is processed based on the spatiotemporal correlation algorithm to obtain a multidimensional feature vector of the distribution network line; then, the multidimensional feature vector is input into a pre-trained warning index determination model to obtain the warning index output by the warning index determination model, and the warning risk level is determined according to the warning index. Different warning risk levels correspond to different warning strategies. The present application provides a high-density area fire warning method based on a multi-source heterogeneous sensor network. The input of the warning index determination model is a multi-dimensional feature vector, and the multi-dimensional feature vector is obtained by monitoring the distribution network lines in the target high-density area based on the target multi-source heterogeneous sensor network, and processing the operating status data of the distribution network lines. On the one hand, due to the high quality of the input data of the warning index determination model, the accuracy of the output of the warning index determination model is higher. On the other hand, the use of the warning index determination model to achieve fire warning has higher reliability and flexibility than the existing technology of threshold monitoring, thereby effectively improving the efficiency of the fire warning method. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following briefly introduces the drawings required for use in the embodiments of the present application or related technical descriptions. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying any creative work.

[0024] Figure 1 1 is a flow chart of a high-density area fire early warning method based on a multi-source heterogeneous sensor network in one embodiment;

[0025] Figure 2 1 is a flow chart of a method for obtaining operating status data of a power distribution network line in one embodiment;

[0026] Figure 3 1 is a flow chart of a method for obtaining a multi-dimensional feature vector of a power distribution network line in one embodiment;

[0027] Figure 4A schematic flow chart of a high-density regional fire warning method based on a multi-source heterogeneous sensor network in another embodiment;

[0028] Figure 5 1. A structural block diagram of a high-density regional fire warning device based on a multi-source heterogeneous sensor network in one embodiment;

[0029] Figure 6 is a diagram of the internal structure of a computer device in one embodiment;

[0030] Figure 7 FIG. 4 is a diagram showing the internal structure of a computer device in another embodiment. DETAILED DESCRIPTION

[0031] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0032] As a special residential form in the process of urbanization, urban villages generally have electrical hazards such as illegal connections, aging lines and three-phase load imbalance. These electrical hazards are one of the common factors causing fires. Therefore, fires caused by electrical hazards can be prevented by early warning.

[0033] In the existing technology, most of the methods are to monitor the threshold of circuit data to achieve early warning of fire.

[0034] However, this method of implementing a threshold monitoring of circuit data to achieve fire early warning has a high false alarm rate, which leads to poor accuracy of fire early warning and low efficiency of the fire early warning method.

[0035] In view of this, the present application provides a high-density regional fire warning method based on a multi-source heterogeneous sensor network, which can effectively improve the accuracy of fire warnings and thereby effectively improve the efficiency of the fire warning method.

[0036] The high-density regional fire warning method based on a multi-source heterogeneous sensor network provided in this application can be executed by a computer device, which can be a terminal or a server.

[0037] In an exemplary embodiment, Figure 1 As shown, a high-density regional fire early warning method based on a multi-source heterogeneous sensor network is provided, which includes the following steps:

[0038] Step 101: Monitor the distribution network lines in the target high-density area based on the target multi-source heterogeneous sensor network to obtain operating status data of the distribution network lines.

[0039] The target multi-source heterogeneous sensing network is constructed based on the electrical environment characteristics of the target high-density area, and the operating status data includes current data, voltage data and cable temperature data.

[0040] For example, the target multi-source heterogeneous sensing network may include current sensors, voltage sensors, and distributed fiber optic temperature sensors. The current sensors and voltage sensors may be installed in distribution boxes and user-side service lines within the distribution network, while the distributed fiber optic temperature sensors may be attached to the surface of cables within the distribution network.

[0041] Furthermore, the current sensor can cover a frequency band of 10Hz-10MHz, with a resolution of 0.5mA and a sampling rate of 200kHz. The voltage sensor can also sample at a 10kHz rate and simultaneously collect transient fluctuation signals from the line. The distributed fiber optic temperature sensor can be deployed at a spacing of 3-5 meters, with a spatial accuracy of 0.1m and a sampling rate of 1Hz.

[0042] In an optional embodiment of the present application, the target multi-source heterogeneous sensor network can use a GPS timing module to align the timestamps of the multi-source signals, that is, to ensure the timing consistency of the data collected by the current sensor, voltage sensor and distributed optical fiber temperature sensor, and try to make the synchronization error less than 10μs.

[0043] For example, a high-density area refers to an area with high residential density, such as an urban village. Electrical environment characteristics of a high-density area include high load density, severely aged lines, widespread illegal wiring, and a complex wiring layout. A target high-density area is a high-density area where fire warnings are to be issued.

[0044] In some exemplary embodiments, a computer device can monitor the distribution network lines in a target high-density area based on current sensors, voltage sensors, and distributed optical fiber temperature sensors in a target multi-source heterogeneous sensing network to obtain current data, voltage data, and cable temperature data of the distribution network lines.

[0045] Step 102: Process the current data based on a wavelet packet energy entropy analysis algorithm, process the voltage data based on a dynamic impedance spectrum correlation analysis algorithm, and process the cable temperature data based on a spatiotemporal correlation algorithm to obtain a multidimensional feature vector of the distribution network line.

[0046] The wavelet packet energy entropy analysis algorithm is a signal analysis method that integrates the wavelet packet transform and information entropy theory. Wavelet packets are an extension of the wavelet transform. While the wavelet transform can only further decompose the low-frequency portion of a signal, the wavelet packet transform can perform multi-level decomposition of both the low-frequency portion and the high-frequency portion, enabling detailed analysis of current data. By performing wavelet packet decomposition on current data, the current data can be expanded across different frequency bands, yielding a series of wavelet packet coefficients for different frequency bands. In current data analysis, the energy of current data reflects the intensity of the current data within a specific frequency band or time period. Within each frequency band after the wavelet packet transform, the energy of the current data can be calculated by calculating the sum of the squares of the wavelet packet coefficients within that frequency band. Entropy can be used to measure the uncertainty or information content of a system. In current data analysis, entropy can be used to describe the complexity or irregularity of the current data.

[0047] The dynamic impedance spectrum correlation analysis algorithm refers to an algorithm that analyzes the intrinsic connection and similarity between different impedance spectrum curves based on dynamic impedance spectrum data.

[0048] The spatiotemporal correlation algorithm refers to an algorithm that comprehensively considers the two dimensions of time and space to explore and analyze the mutual connections and mutual influences between different objects, events or phenomena within the scope of time and space.

[0049] In some exemplary embodiments, a computer device monitors the distribution network lines in a target high-density area using current sensors, voltage sensors, and distributed optical fiber temperature sensors in a target multi-source heterogeneous sensing network. After obtaining current data, voltage data, and cable temperature data of the distribution network lines, the current data can be input into a first model, the voltage data can be input into a second model, and the cable temperature data can be input into a third model, so that the first model processes the current data based on a wavelet packet energy entropy analysis algorithm, the second model processes the voltage data based on a dynamic impedance spectrum correlation analysis algorithm, and the third model processes the cable temperature data based on a spatiotemporal correlation algorithm, so as to obtain the outputs of the first model, the second model, and the third model.

[0050] Furthermore, after obtaining the outputs of the first model, the second model, and the third model, the computer device may perform fusion processing on the outputs of the first model, the second model, and the third model to obtain a multi-dimensional feature vector of the distribution network line.

[0051] Step 103: input the multidimensional feature vector into a pre-trained early warning index determination model to obtain an early warning index output by the early warning index determination model, and determine the early warning risk level according to the early warning index.

[0052] Among them, different warning risk levels correspond to different warning strategies.

[0053] Exemplarily, the warning index determination model can be pre-trained by technicians based on actual needs. Specifically, the warning index determination model can use a ResNet-18 network as its underlying architecture, with the input layer adjusted to a multi-dimensional feature vector, the original classification layer removed, and the fully connected layer retaining the 256-dimensional feature vector output. The warning index determination model can be trained based on training data, which can be a historical dataset of standard distribution rooms that has been annotated to cover potential hazards such as overloads, arcing, and poor contact.

[0054] The loss function of the warning index determination model can adopt the cross entropy loss function, that is, , where y i is the true label, p i is the predicted probability. The maximum mean difference loss of the early warning index determination model can be expressed as ,in, is the Gaussian kernel mapping, and H is the reproducing kernel Hilbert space.

[0055] The early warning index determines the initial learning rate of the optimizer in the model training process is 3×10 -4 , the batch size is 256.

[0056] In some exemplary embodiments, after obtaining the multidimensional feature vector, the computer device may input the multidimensional feature vector into a pre-trained early warning index determination model to obtain the early warning index output by the early warning index determination model.

[0057] Furthermore, after obtaining the warning index output by the warning index determination model, the computer device can input the warning index into a pre-trained warning risk level determination model to obtain the warning risk level output by the warning risk level determination model.

[0058] The above-mentioned high-density area fire warning method based on a multi-source heterogeneous sensor network first monitors the distribution network line in the target high-density area based on the target multi-source heterogeneous sensor network to obtain the operating status data of the distribution network line. The target multi-source heterogeneous sensor network is constructed according to the electrical environment characteristics of the target high-density area, and the operating status data includes current data, voltage data and cable temperature data; then, the current data is processed based on the wavelet packet energy entropy analysis algorithm, the voltage data is processed based on the dynamic impedance spectrum correlation analysis algorithm, and the cable temperature data is processed based on the time-space correlation algorithm to obtain a multi-dimensional feature vector of the distribution network line; then, the multi-dimensional feature vector is input into a pre-trained warning index determination model to obtain the warning index output by the warning index determination model, and the warning risk level is determined according to the warning index. Different warning risk levels correspond to different warning strategies. The present application provides a high-density area fire warning method based on a multi-source heterogeneous sensor network. The input of the warning index determination model is a multi-dimensional feature vector, and the multi-dimensional feature vector is obtained by monitoring the distribution network lines in the target high-density area based on the target multi-source heterogeneous sensor network, and processing the operating status data of the distribution network lines. On the one hand, due to the high quality of the input data of the warning index determination model, the accuracy of the output of the warning index determination model is higher. On the other hand, the use of the warning index determination model to achieve fire warning has higher reliability and flexibility than the existing technology of threshold monitoring, thereby effectively improving the efficiency of the fire warning method.

[0059] In an exemplary embodiment, Figure 2 As shown, the target multi-source heterogeneous sensor network is used to monitor the distribution network lines in the target high-density area to obtain the operating status data of the distribution network lines, including the following steps:

[0060] Step 201: Monitor the distribution network lines in the target high-density area based on the target multi-source heterogeneous sensor network to obtain initial operating status data of the distribution network lines.

[0061] The initial operating state data includes initial current data, initial voltage data and initial cable temperature data.

[0062] In some exemplary embodiments, a computer device can monitor the distribution network lines in a target high-density area based on current sensors, voltage sensors, and distributed optical fiber temperature sensors in a target multi-source heterogeneous sensing network to obtain initial current data, initial voltage data, and initial cable temperature data of the distribution network lines.

[0063] Step 202: Denoise the initial operating status data using an adaptive wavelet threshold denoising algorithm, normalize the initial operating status data based on sliding window statistical characteristics, and perform time series alignment on the initial operating status data using a variable time window segmentation strategy to obtain the operating status data.

[0064] In some exemplary embodiments, after obtaining the initial current data, initial voltage data and initial cable temperature data of the distribution network line, the computer device can first use an adaptive wavelet threshold noise reduction algorithm to reduce the power frequency harmonics, electromagnetic interference and random noise in the initial operating status data.

[0065] Specifically, the computer device can first perform wavelet decomposition, select Db6 wavelet basis for 5 decomposition for the initial current data, and select Sym8 wavelet basis for the initial voltage data. The threshold of each scale can be determined by Determine, among them, is the standard deviation of the wavelet coefficients in the jth layer, N j is the number of coefficients.

[0066] Furthermore, the computer device may also perform normalization processing on the initial operating status data based on the statistical characteristics of the sliding window to eliminate dimensional differences.

[0067] Specifically, for the initial current data and initial voltage data, we can use the window length Tw=1s. Normalize, where , are the mean and standard deviation within the window, =10 -6 For the initial cable temperature data, you can follow Linear normalization is performed to the interval [0, 1], where T min , T max The minimum and maximum ambient temperatures are preset and can be adjusted according to the environment in different regions.

[0068] Furthermore, the computer device may also use a variable time window segmentation strategy to perform time sequence alignment processing on the initial operating status data.

[0069] Specifically, the computer device can perform dynamic time warping by Calculate the timing alignment path for the initial current data, initial voltage data, and initial cable temperature data, where s i and t j is a sequence of different sensor signals, and D(i, j) is a cumulative distance matrix. Then, the computer device can perform variable time window segmentation, and the window length L W According to the volatility Dynamic adjustment can be expressed as ,in, is the current fluctuation rate, and its computer formula is .

[0070] In an optional embodiment of the present application, the computer device performs noise reduction processing on the initial operating status data through an adaptive wavelet threshold noise reduction algorithm, normalizes the initial operating status data based on the statistical characteristics of the sliding window, and uses a variable time window segmentation strategy to perform time alignment processing on the initial operating status data. After obtaining the operating status data, the operating status data can be encrypted based on the AES-256 encryption algorithm and uploaded through the MQTT protocol. The data packet format can include a timestamp, sensor ID and check code.

[0071] In an exemplary embodiment, Figure 3 As shown, the current data is processed based on the wavelet packet energy entropy analysis algorithm, the voltage data is processed based on the dynamic impedance spectrum correlation analysis algorithm, and the cable temperature data is processed based on the time-space correlation algorithm to obtain the multidimensional feature vector of the distribution network line, including the following steps:

[0072] Step 301 : Process the current data based on a wavelet packet energy entropy analysis algorithm to obtain energy entropy of a preset frequency band.

[0073] In some exemplary embodiments, the computer device may process the current data based on a wavelet packet energy entropy analysis algorithm to obtain energy entropy of a preset frequency band.

[0074] Specifically, the computer device can first perform wavelet packet decomposition on the current data, and use the Db6 wavelet basis to perform 5-layer decomposition on the current data to generate 32 sub-bands. The specific process can be expressed as , where W 5，k is the wavelet packet coefficient of the kth sub-band in the 5th layer. Then, the computer equipment performs frequency division, and the bandwidth of each sub-band is , covering the key frequency band of 0-10MHz.

[0075] Furthermore, the computer device can calculate the energy of each sub-band. For each sub-band, the energy value can be calculated based on its corresponding wavelet packet coefficient. Let the wavelet packet coefficient sequence of the ith sub-band be w i (n), where n=1, 2, ..., N, where N is the number of coefficients in the sub-band, then the energy E of the sub-band is i It can be expressed as Based on this, the computer device can obtain the energy value of each of the 32 sub-bands.

[0076] Then, the computer device can perform energy normalization. Calculate the total energy of all sub-bands , and then divide the energy of each sub-band by the total energy to obtain the normalized energy value p i , , where i = 1, 2, 3, ..., 32. The purpose of normalization is to make the energy proportion of each sub-band between 0 and 1, which is convenient for subsequent entropy calculation.

[0077] Finally, the energy entropy H of the preset frequency band is calculated according to the definition of Shannon entropy. This energy entropy value can reflect the uncertainty and complexity of the energy distribution of each sub-band of the current data within the preset frequency band.

[0078] Step 302: Process the voltage data based on a dynamic impedance spectrum correlation analysis algorithm to obtain impedance correlation between low frequency and high frequency.

[0079] In some exemplary embodiments, the computer device may process the voltage data based on a dynamic impedance spectrum correlation analysis algorithm to obtain impedance correlation between low frequency and high frequency.

[0080] Specifically, the computer device can first perform impedance spectrum calculation and perform short-time Fourier transform on the voltage data with a window length of 10ms and an overlap rate of 75%, which can be expressed as ,in, and are the time spectra of voltage and current respectively.

[0081] Furthermore, the computer equipment performs frequency domain coherence determination, that is, calculates the contact point impedance correlation, which can be expressed as , where Z ref is the reference impedance spectrum under normal working conditions, Z test is the measured impedance spectrum. When C(f)<0.85, the frequency point can be marked as an abnormal contact area.

[0082] Step 303: Process the cable temperature data based on a spatiotemporal correlation algorithm to obtain an axial temperature gradient and a temperature-current coefficient.

[0083] In some exemplary embodiments, the computer device may process the cable temperature data based on a spatiotemporal correlation algorithm to obtain an axial temperature gradient and a temperature-current coefficient.

[0084] Specifically, the computer equipment can first calculate the axial temperature gradient and calculate the temperature difference between adjacent measuring points of the cable temperature data, which can be expressed as , where d is the sensor spacing in meters. Then, the correlation coefficient between the temperature gradient and the current fluctuation is calculated within the sliding window (TW=10s), which can be expressed as , to realize the coupled analysis of temperature and current, where cov is the covariance, is the standard deviation. When it is >0.7, it can be determined that the temperature rise is abnormal due to overload.

[0085] Step 304 : Determine the multidimensional feature vector according to the energy entropy of the preset frequency band, the impedance correlation between the low frequency and high frequency, the axial temperature gradient, and the temperature-current coefficient.

[0086] In some exemplary embodiments, the computer device may determine the multidimensional feature vector based on the energy entropy of the preset frequency band, the impedance correlation between the low frequency and high frequency, the axial temperature gradient, and the temperature-current coefficient.

[0087] Specifically, the multidimensional feature vector can be expressed as , among which, H5-H 10 Indicates the energy entropy in the 2-5MHz frequency band, C(1kHz), C(10kHz) indicate the correlation between low-frequency and high-frequency impedance, represents the maximum temperature gradient within the window, represents the temperature-current correlation coefficient.

[0088] In an exemplary embodiment, the warning index includes an energy accumulation rate and a temperature rise gradient, and determining the warning risk level based on the warning index includes: when the energy accumulation rate and the temperature rise gradient meet a first preset condition, determining the warning risk level to be a first risk level.

[0089] Among them, the warning strategy executed corresponding to the first risk level is the first warning strategy, and the first warning strategy is to output first warning information. The first warning information includes a first predicted warning position and a disposal suggestion for the first predicted warning position.

[0090] For example, the first preset condition can be pre-set by technicians according to actual needs. Specifically, the energy integral change rate within the window (TW=10s) , where R is the equivalent resistance of the line, which can be calculated by impedance spectrum inversion. Or the average axial temperature gradient .

[0091] In some exemplary embodiments, the computer device may determine that the warning risk level is the first risk level when the energy accumulation rate and the temperature rise gradient meet a first preset condition.

[0092] Furthermore, after determining that the warning risk level is the first risk level, the computer device may output a first warning message, which includes a first predicted warning location and a recommended action for the first predicted warning location. Specifically, the computer device may push the first predicted warning location via a mobile app, and the recommended action for the first predicted warning location may include checking for loose wiring connections, and storing an event log for subsequent review.

[0093] In an exemplary embodiment, the warning index also includes the cable axial temperature change rate and the contact resistance mutation rate. The warning risk level is determined based on the warning index, including: when the cable axial temperature change rate and the contact resistance mutation rate meet the second preset condition, determining the warning risk level as the second risk level.

[0094] Among them, the warning strategy executed corresponding to the second risk level is the second warning strategy, which includes outputting second warning information, cutting off the power supply of the second predicted warning location, and starting the cooling equipment for the second predicted warning location. The second warning information includes the second warning location.

[0095] For example, the second preset condition can be pre-set by technicians according to actual needs. Specifically, the temperature rise rate of any measuring point in the window , or contact resistance mutation, that is, the contact resistance change rate monitored by dynamic impedance spectroscopy .

[0096] In some exemplary embodiments, the computer device determines that the warning risk level is the second risk level when the cable axial temperature change rate and the contact resistance sudden change rate meet a second preset condition.

[0097] Furthermore, after determining that the warning risk level is the second risk level, the computer device may output second warning information, cut off power supply to the second predicted warning location, and start a cooling device for the second predicted warning location.

[0098] In an exemplary embodiment, the warning index also includes a high-frequency arc feature, and determining the warning risk level based on the warning index includes: when the high-frequency arc feature meets a third preset condition, determining the warning risk level to be a third risk level.

[0099] Among them, the warning strategy corresponding to the third risk level is the third warning strategy, which includes starting fire-fighting equipment for the third warning position, triggering sound and light alarm equipment, and outputting third warning information, and the third warning information includes the third warning position.

[0100] For example, the third preset condition can be pre-set by technicians according to actual needs. Specifically, it is detected that the energy entropy of the wavelet packet in the 2-5MHz frequency band continuously exceeds the threshold. , and the duration is greater than 100ms.

[0101] In some exemplary embodiments, the computer device determines that the warning risk level is a third risk level when the high-frequency arc characteristic meets a third preset condition.

[0102] Furthermore, after determining that the warning risk level is the third risk level, the computer device can start the fire-fighting equipment for the third warning location, trigger the sound and light alarm device, and output the third warning information. Specifically, the fire controller can be triggered via the Modbus protocol.

[0103] In an exemplary embodiment, Figure 4 As shown, another high-density area fire early warning method based on a multi-source heterogeneous sensor network is provided, which includes the following steps:

[0104] Step 401: Monitor the distribution network lines in the target high-density area based on the target multi-source heterogeneous sensor network to obtain initial operating status data of the distribution network lines, where the initial operating status data includes initial current data, initial voltage data, and initial cable temperature data.

[0105] Step 402: Denoise the initial operating status data using an adaptive wavelet threshold noise reduction algorithm, normalize the initial operating status data based on sliding window statistical characteristics, and perform time series alignment on the initial operating status data using a variable time window segmentation strategy to obtain the operating status data. The target multi-source heterogeneous sensor network is constructed based on the electrical environment characteristics of the target high-density area. The operating status data includes current data, voltage data, and cable temperature data.

[0106] Step 403: Process the current data based on a wavelet packet energy entropy analysis algorithm to obtain energy entropy in a preset frequency band; process the voltage data based on a dynamic impedance spectrum correlation analysis algorithm to obtain impedance correlation between low and high frequencies; process the cable temperature data based on a spatiotemporal correlation algorithm to obtain an axial temperature gradient and a temperature-current coefficient; and determine the multidimensional feature vector based on the energy entropy in the preset frequency band, the impedance correlation between low and high frequencies, the axial temperature gradient, and the temperature-current coefficient.

[0107] Step 404: Input the multidimensional feature vector into a pre-trained warning index determination model to obtain a warning index output by the warning index determination model, where the warning index includes energy accumulation rate, temperature rise gradient, cable axial temperature change rate, contact resistance mutation rate, and high-frequency arc characteristics.

[0108] Step 405: If the energy accumulation rate and the temperature rise gradient satisfy a first preset condition, determine that the warning risk level is a first risk level; wherein the warning strategy executed corresponding to the first risk level is a first warning strategy, and the first warning strategy is to output first warning information, wherein the first warning information includes a first predicted warning location and a disposal suggestion for the first predicted warning location;

[0109] Step 406: If the cable axial temperature change rate and the contact resistance mutation rate meet a second preset condition, determine that the warning risk level is a second risk level; wherein the warning strategy executed corresponding to the second risk level is a second warning strategy, and the second warning strategy includes outputting a second warning message, cutting off power to a second predicted warning location, and activating a cooling device for the second predicted warning location, and the second warning message includes the second warning location;

[0110] Step 407: When the high-frequency arc characteristic meets the third preset condition, determine that the warning risk level is the third risk level; wherein the warning strategy executed corresponding to the third risk level is the third warning strategy, and the third warning strategy includes starting the fire-fighting equipment for the third warning position, triggering the sound and light alarm equipment, and outputting the third warning information, and the third warning information includes the third warning position.

[0111] It should be noted that, after verification and testing, the high-density regional fire warning method based on a multi-source heterogeneous sensor network provided by this application automatically triggers an early warning and cuts off the power supply to the potential danger line when a local temperature rise rate ≥3°C / min or a contact resistance mutation ≥0.5Ω / s is detected. When a 2-5MHz high-frequency arc feature lasting 100ms is identified, an emergency response is immediately initiated, the fire protection system is linked, and three-dimensional positioning coordinates (accuracy ±0.5m) are pushed. Moreover, through the cloud platform, multi-region hidden danger data is aggregated and analyzed, and the model weights are updated weekly using an incremental learning module to stably control the false alarm rate within 5%. The high-density regional fire warning method based on a multi-source heterogeneous sensor network provided by the application has been verified by 2000 hours of actual testing and successfully advanced the early hidden danger detection time to 25-40 minutes before the fire.

[0112] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0113] Based on the same inventive concept, embodiments of the present application also provide a high-density regional fire warning device based on a multi-source heterogeneous sensor network, which is used to implement the aforementioned high-density regional fire warning method based on a multi-source heterogeneous sensor network. The implementation solution provided by this device is similar to the implementation solution described in the aforementioned method. Therefore, the specific limitations of one or more embodiments of the high-density regional fire warning device based on a multi-source heterogeneous sensor network provided below can be found in the above-mentioned limitations of the high-density regional fire warning method based on a multi-source heterogeneous sensor network, and will not be repeated here.

[0114] In an exemplary embodiment, Figure 5 As shown, a high-density regional fire warning device 500 based on a multi-source heterogeneous sensor network is provided, comprising: an execution module 501, a processing module 502 and a determination module 503, wherein:

[0115] An execution module 501 is configured to monitor a distribution network line in a target high-density area based on a target multi-source heterogeneous sensor network to obtain operating status data of the distribution network line, wherein the target multi-source heterogeneous sensor network is constructed based on electrical environment characteristics of the target high-density area, and the operating status data includes current data, voltage data, and cable temperature data;

[0116] Processing module 502, configured to process the current data based on a wavelet packet energy entropy analysis algorithm, process the voltage data based on a dynamic impedance spectrum correlation analysis algorithm, and process the cable temperature data based on a spatiotemporal correlation algorithm to obtain a multidimensional feature vector of the distribution network line;

[0117] Determination module 503 is used to input the multidimensional feature vector into a pre-trained warning index determination model to obtain the warning index output by the warning index determination model, and determine the warning risk level according to the warning index. Different warning risk levels correspond to different warning strategies.

[0118] In one embodiment, the execution module 501 is specifically used to monitor the distribution network lines in the target high-density area based on the target multi-source heterogeneous sensor network to obtain the initial operating status data of the distribution network route, and the initial operating status data includes initial current data, initial voltage data and initial cable temperature data; the initial operating status data is denoised by an adaptive wavelet threshold denoising algorithm, the initial operating status data is normalized based on the sliding window statistical characteristics, and the initial operating status data is time-aligned using a variable time window segmentation strategy to obtain the operating status data.

[0119] In one embodiment, the processing module 502 is specifically used to perform data processing on the current data based on a wavelet packet energy entropy analysis algorithm to obtain energy entropy of a preset frequency band; perform data processing on the voltage data based on a dynamic impedance spectrum correlation analysis algorithm to obtain impedance correlation between low frequency and high frequency; perform data processing on the cable temperature data based on a time-space correlation algorithm to obtain an axial temperature gradient and a temperature-current coefficient; and determine the multidimensional feature vector based on the energy entropy of the preset frequency band, the impedance correlation between low frequency and high frequency, the axial temperature gradient, and the temperature-current coefficient.

[0120] In one embodiment, the warning index includes the energy accumulation rate and the temperature rise gradient. The determination module 503 is specifically used to determine that the warning risk level is the first risk level when the energy accumulation rate and the temperature rise gradient meet the first preset condition; wherein, the warning strategy executed corresponding to the first risk level is the first warning strategy, and the first warning strategy is to output the first warning information, and the first warning information includes the first predicted warning position and the disposal suggestions for the first predicted warning position.

[0121] In one embodiment, the warning index also includes the cable axial temperature change rate and the contact resistance mutation rate. The determination module 503 is specifically used to determine that the warning risk level is the second risk level when the cable axial temperature change rate and the contact resistance mutation rate meet the second preset condition; wherein the warning strategy executed corresponding to the second risk level is the second warning strategy, and the second warning strategy includes outputting a second warning information, cutting off the power supply to the second predicted warning position, and starting a cooling device for the second predicted warning position, and the second warning information includes the second warning position.

[0122] In one embodiment, the warning index also includes a high-frequency arc feature, and the determination module 503 is specifically used to determine that the warning risk level is a third risk level when the high-frequency arc feature meets a third preset condition; wherein the warning strategy executed corresponding to the third risk level is a third warning strategy, and the third warning strategy includes starting fire-fighting equipment for a third warning position, triggering sound and light alarm equipment, and outputting a third warning information, and the third warning information includes the third warning position.

[0123] Each module in the aforementioned high-density regional fire warning device based on a multi-source heterogeneous sensor network can be implemented in whole or in part through software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor within a computer device in hardware form, or stored in a computer device memory in software form, allowing the processor to call and execute the corresponding operations of each module.

[0124] In an exemplary embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Figure 6 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O) and a communication interface. The processor, memory and input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device is used to store data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a high-density regional fire warning method based on a multi-source heterogeneous sensor network is implemented.

[0125] In an exemplary embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as shown in FIG. Figure 7As shown. The computer device includes a processor, memory, an input / output interface, a communication interface, a display unit, and an input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals via wired or wireless communication, and the wireless communication can be implemented via Wi-Fi, a mobile cellular network, near field communication (NFC), or other technologies. When executed by the processor, the computer program implements a high-density area fire warning method based on a multi-source heterogeneous sensor network.

[0126] Those skilled in the art will understand that Figure 6 and Figure 7 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0127] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0128] Monitoring the distribution network lines in a target high-density area based on a target multi-source heterogeneous sensor network to obtain operating status data of the distribution network lines. The target multi-source heterogeneous sensor network is constructed based on the electrical environment characteristics of the target high-density area. The operating status data includes current data, voltage data, and cable temperature data.

[0129] The current data is processed based on a wavelet packet energy entropy analysis algorithm, the voltage data is processed based on a dynamic impedance spectrum correlation analysis algorithm, and the cable temperature data is processed based on a spatiotemporal correlation algorithm to obtain a multidimensional feature vector of the distribution network line;

[0130] The multidimensional feature vector is input into a pre-trained early warning index determination model to obtain the early warning index output by the early warning index determination model, and the early warning risk level is determined according to the early warning index. Different early warning risk levels correspond to different early warning strategies.

[0131] In one embodiment, when the processor executes the computer program, the following steps are further implemented: based on the target multi-source heterogeneous sensor network, the distribution network line in the target high-density area is monitored to obtain initial operating status data of the distribution network route, and the initial operating status data includes initial current data, initial voltage data and initial cable temperature data; the initial operating status data is denoised using an adaptive wavelet threshold denoising algorithm, the initial operating status data is normalized based on the sliding window statistical characteristics, and the initial operating status data is time-aligned using a variable time window segmentation strategy to obtain the operating status data.

[0132] In one embodiment, when the processor executes the computer program, the following steps are further implemented: data processing is performed on the current data based on a wavelet packet energy entropy analysis algorithm to obtain energy entropy of a preset frequency band; data processing is performed on the voltage data based on a dynamic impedance spectrum correlation analysis algorithm to obtain impedance correlation between low frequency and high frequency; data processing is performed on the cable temperature data based on a time-space correlation algorithm to obtain an axial temperature gradient and a temperature-current coefficient; and the multidimensional feature vector is determined based on the energy entropy of the preset frequency band, the impedance correlation between low frequency and high frequency, the axial temperature gradient, and the temperature-current coefficient.

[0133] In one embodiment, when the processor executes the computer program, the following steps are also implemented: when the energy accumulation rate and the temperature rise gradient satisfy a first preset condition, the warning risk level is determined to be a first risk level; wherein the warning strategy executed corresponding to the first risk level is a first warning strategy, and the first warning strategy is to output a first warning information, and the first warning information includes a first predicted warning position and a disposal suggestion for the first predicted warning position.

[0134] In one embodiment, when the processor executes the computer program, it also implements the following steps: when the axial temperature change rate of the cable and the contact resistance mutation rate meet the second preset condition, the warning risk level is determined to be the second risk level; wherein the warning strategy executed corresponding to the second risk level is the second warning strategy, and the second warning strategy includes outputting a second warning information, cutting off the power supply to the second predicted warning position, and starting a cooling device for the second predicted warning position, and the second warning information includes the second warning position.

[0135] In one embodiment, when the processor executes the computer program, the following steps are also implemented: when the high-frequency arc characteristic meets the third preset condition, the warning risk level is determined to be the third risk level; wherein the warning strategy executed corresponding to the third risk level is the third warning strategy, and the third warning strategy includes starting the fire-fighting equipment for the third warning position, triggering the sound and light alarm equipment, and outputting the third warning information, and the third warning information includes the third warning position.

[0136] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps described in any of the above embodiments are implemented.

[0137] In one embodiment, a computer program product is provided, including a computer program, which implements the steps described in any of the above embodiments when executed by a processor.

[0138] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. In particular, any reference to memory, database, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the various embodiments provided herein may be, but are not limited to, general-purpose processors, central processing units (CPUs), graphics processing units (GPUs), digital signal processors (DSPs), programmable logic devices (PLDs), quantum computing-based data processing logic devices, artificial intelligence (AI) processors, and the like.

[0139] The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0140] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. A high-density regional fire early warning method based on a multi-source heterogeneous sensor network, characterized in that: The method comprises: Monitoring the distribution network lines in the target high-density area based on a target multi-source heterogeneous sensor network to obtain operating status data of the distribution network lines, wherein the target multi-source heterogeneous sensor network is constructed based on electrical environment characteristics of the target high-density area, and the operating status data includes current data, voltage data, and cable temperature data; Processing the current data based on a wavelet packet energy entropy analysis algorithm, processing the voltage data based on a dynamic impedance spectrum correlation analysis algorithm, and processing the cable temperature data based on a spatiotemporal correlation algorithm to obtain a multidimensional feature vector of the distribution network line; The multidimensional feature vector is input into a pre-trained warning index determination model to obtain a warning index output by the warning index determination model, and the warning risk level is determined according to the warning index. Different warning risk levels correspond to different warning strategies.

2. The method according to claim 1, characterized in that The monitoring of the distribution network lines in the target high-density area based on the target multi-source heterogeneous sensor network to obtain the operating status data of the distribution network lines includes: Monitoring the distribution network lines in the target high-density area based on the target multi-source heterogeneous sensor network to obtain initial operating status data of the distribution network lines, wherein the initial operating status data includes initial current data, initial voltage data, and initial cable temperature data; The initial operating status data is subjected to denoising processing by an adaptive wavelet threshold denoising algorithm, the initial operating status data is normalized based on sliding window statistical characteristics, and the initial operating status data is time-series aligned using a variable time window segmentation strategy to obtain the operating status data.

3. The method according to claim 1, characterized in that The data processing of the current data based on the wavelet packet energy entropy analysis algorithm, the data processing of the voltage data based on the dynamic impedance spectrum correlation analysis algorithm, and the data processing of the cable temperature data based on the spatiotemporal correlation algorithm to obtain the multidimensional feature vector of the distribution network line includes: Processing the current data based on a wavelet packet energy entropy analysis algorithm to obtain energy entropy in a preset frequency band; Processing the voltage data based on a dynamic impedance spectrum correlation analysis algorithm to obtain impedance correlation between low frequency and high frequency; Processing the cable temperature data based on a spatiotemporal correlation algorithm to obtain an axial temperature gradient and a temperature-current coefficient; The multidimensional feature vector is determined according to the energy entropy of the preset frequency band, the impedance correlation between the low frequency and the high frequency, the axial temperature gradient and the temperature-current coefficient.

4. The method according to claim 1, wherein The warning index includes energy accumulation rate and temperature rise gradient, and determining the warning risk level according to the warning index includes: When the energy accumulation rate and the temperature rise gradient meet a first preset condition, determining the warning risk level to be a first risk level; Among them, the warning strategy executed corresponding to the first risk level is the first warning strategy, and the first warning strategy is to output first warning information. The first warning information includes a first predicted warning location and a disposal suggestion for the first predicted warning location.

5. The method according to claim 4, characterized in that The warning index also includes the cable axial temperature change rate and the contact resistance mutation rate. Determining the warning risk level based on the warning index includes: When the cable axial temperature change rate and the contact resistance mutation rate meet a second preset condition, determining the warning risk level to be a second risk level; Among them, the warning strategy executed corresponding to the second risk level is the second warning strategy, which includes outputting second warning information, cutting off the power supply to the second predicted warning location, and starting the cooling equipment for the second predicted warning location. The second warning information includes the second warning location.

6. The method according to claim 5, characterized in that The warning index also includes high-frequency arc characteristics, and determining the warning risk level according to the warning index includes: When the high-frequency arc characteristic meets a third preset condition, determining the warning risk level to be a third risk level; Among them, the warning strategy corresponding to the third risk level is the third warning strategy, and the third warning strategy includes starting the fire-fighting equipment for the third warning position, triggering the sound and light alarm equipment, and outputting the third warning information, and the third warning information includes the third warning position.

7. A high-density regional fire warning device based on a multi-source heterogeneous sensor network, characterized in that: The device comprises: an execution module, configured to monitor a distribution network line in a target high-density area based on a target multi-source heterogeneous sensing network to obtain operating status data of the distribution network line, wherein the target multi-source heterogeneous sensing network is constructed based on electrical environment characteristics of the target high-density area, and the operating status data includes current data, voltage data, and cable temperature data; a processing module, configured to process the current data based on a wavelet packet energy entropy analysis algorithm, process the voltage data based on a dynamic impedance spectrum correlation analysis algorithm, and process the cable temperature data based on a spatiotemporal correlation algorithm, so as to obtain a multidimensional feature vector of the distribution network line; The determination module is used to input the multidimensional feature vector into a pre-trained warning index determination model to obtain the warning index output by the warning index determination model, and determine the warning risk level according to the warning index. Different warning risk levels correspond to different warning strategies.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images