Electrical fire early-stage symptom monitoring and identifying device based on multi-fault linkage simulation

Abstract

The invention, which relates to the technical field of electrical monitoring, discloses a multi-fault linkage simulation-based electrical fire early-stage symptom monitoring and identification apparatus comprising the steps of performing monitoring to obtain thermal parameter distribution, gas parameter distribution and optical parameter distribution, performing distinction analysis to obtain thermal distinction distribution, gas distinction distribution and optical distinction distribution, and then performing combination to obtain an electrical fire early-stage symptom identification result; obtaining a plurality of distinction degree distribution combinations, carrying out distinction degree alignment processing to obtain a plurality of alignment position distributions, and calculating a thermal alignment deviation degree, a gas alignment deviation degree and a light alignment deviation degree; and performing arc prediction to obtain a thermal prediction arc parameter, a gas prediction arc parameter and a light prediction arc parameter, and performing weighted calculation to obtain an arc parameter as a monitoring identification result. According to the invention, the technical problems of low electrical fire fault early-stage identification accuracy and delayed early-warning response in the prior art are solved, and the technical effect of improving the electrical fire fault identification accuracy and reliability is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of electrical monitoring, and particularly to a device for monitoring and identifying early signs of electrical fire with multi-fault linkage simulation. Background Art

[0002] In modern power systems, arc fault is a common and dangerous form of electrical fault. Its early characteristics are complex and easily overlooked. Arc faults can not only cause equipment damage, but also easily lead to electrical fires. Therefore, its monitoring and identification are of great significance. However, there are obvious deficiencies in the existing technology for fault monitoring and identification. Traditional methods usually rely on the change of a single parameter for fault detection, but a single parameter is easily affected by environmental noise or other factors, resulting in insufficient identification accuracy. In addition, it is difficult to capture the early characteristics of arc faults in a timely manner, and the existing early warning responses are usually lagged, making it difficult to take effective protective measures in time. Especially in complex arc faults, the correlation of multi-dimensional parameters such as heat, gas, and light cannot be fully utilized, and the traditional system lacks the joint analysis of these data, resulting in unreliable identification results. On the other hand, the existing technology relies on fixed thresholds or simple algorithms, lacking intelligence and adaptability, and unable to dynamically adjust the identification strategy to cope with the complex characteristics of arc faults. This processing method not only affects the timely identification of arc faults, but also results in low monitoring accuracy of the early signs of electrical fires. Summary of the Invention

[0003] The present application provides a device for monitoring and identifying early signs of electrical fire with multi-fault linkage simulation, which is used to solve the technical problems of low accuracy in early identification of electrical fire faults and lagged early warning response in the existing technology.

[0004] In view of the above problems, the present application provides a device for monitoring and identifying early signs of electrical fire with multi-fault linkage simulation.

[0005] The present application provides a device for monitoring and identifying early signs of electrical fire with multi-fault linkage simulation, and the device includes:

[0006] The parameter monitoring and analysis module is used to monitor the multi-dimensional characteristics of the early stage of arc on the target line, obtain the thermal parameter distribution, gas parameter distribution, and optical parameter distribution, respectively perform discrimination analysis to obtain the thermal discrimination distribution, gas discrimination distribution, and optical discrimination distribution; the combination processing module combines the thermal discrimination distribution, gas discrimination distribution, and optical discrimination distribution to obtain multiple discrimination distribution combinations, respectively perform discrimination alignment processing to obtain multiple alignment position distributions, and calculate and obtain the thermal alignment deviation degree, gas alignment deviation degree, and optical alignment deviation degree; the arc prediction module is used to perform arc prediction respectively according to the thermal parameter distribution, gas parameter distribution, and optical parameter distribution to obtain the thermal predicted arc parameters, gas predicted arc parameters, and optical predicted arc parameters; the monitoring and recognition result acquisition module is used to perform weighted calculation on the predicted arc parameter set according to the thermal alignment deviation degree, gas alignment deviation degree, and optical alignment deviation degree to obtain the arc parameters as the monitoring and recognition result.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] This application monitors the multi-dimensional characteristics of the early stage of arc on the target line, obtains the thermal parameter distribution, gas parameter distribution, and optical parameter distribution, respectively performs discrimination analysis to obtain the thermal discrimination distribution, gas discrimination distribution, and optical discrimination distribution; combines the thermal discrimination distribution, gas discrimination distribution, and optical discrimination distribution to obtain multiple discrimination distribution combinations, respectively performs discrimination alignment processing to obtain multiple alignment position distributions, and calculates and obtains the thermal alignment deviation degree, gas alignment deviation degree, and optical alignment deviation degree; performs arc prediction respectively according to the thermal parameter distribution, gas parameter distribution, and optical parameter distribution to obtain the thermal predicted arc parameters, gas predicted arc parameters, and optical predicted arc parameters; according to the thermal alignment deviation degree, gas alignment deviation degree, and optical alignment deviation degree, performs weighted calculation on the predicted arc parameter set to obtain the arc parameters as the monitoring and recognition result. This invention solves the technical problems of low accuracy in the early identification of electrical fire faults and lag in early warning response in the prior art. Through multi-dimensional parameter monitoring and discrimination analysis, it generates the parameter distributions and alignment deviation degrees of heat, gas, and light, combines machine learning to predict the heat, gas, and optical arc parameters, and performs weighted calculation on the comprehensive arc parameters to achieve accurate identification and early warning of arc faults, and achieves the technical effect of improving the accuracy and reliability of the identification of electrical fire faults. Description of the Drawings

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0010] Figure 1 Schematic structural diagram of the electrical fire early warning sign monitoring and recognition device for multi-fault linkage simulation provided by the embodiment of the present application;

[0011] Figure 2 Flow schematic diagram of the combined processing module in the electrical fire early warning sign monitoring and recognition device for multi-fault linkage simulation provided by the embodiment of the present application.

[0012] Explanation of reference numerals: Parameter monitoring and analysis module 11, combined processing module 12, arc prediction module 13, monitoring and recognition result acquisition module 14. Detailed implementation manners

[0013] By providing an electrical fire early warning sign monitoring and recognition device for multi-fault linkage simulation, the present application aims to solve the technical problems of low accuracy in early identification of electrical fire faults and lag in early warning response in the prior art. Through multi-dimensional parameter monitoring and discrimination analysis, the parameter distributions and alignment deviation degrees of heat, gas, and light are generated, and machine learning is combined to predict the heat, gas, and photoelectric arc parameters, and the comprehensive arc parameters are calculated by weighting, so as to achieve accurate identification and early warning of arc faults, and achieve the technical effects of improving the accuracy and reliability of electrical fire fault identification.

[0014] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative efforts shall fall within the protection scope of the present application.

[0015] It should be noted that any variations of the terms "include" and "have" are intended to cover non-exclusive inclusions. For example, a process, method, device, product, or server that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or are inherent to these processes, methods, products, or devices.

[0016] Embodiment, as Figure 1 shown, the embodiment of the present application provides an electrical fire early warning sign monitoring and recognition device for multi-fault linkage simulation, and the device includes:

[0017] A parameter monitoring and analysis module 11, configured to monitor the multi-dimensional early features of the arc of the target line, obtain the heat parameter distribution, gas parameter distribution, and light parameter distribution, and respectively perform discrimination analysis to obtain the heat discrimination distribution, gas discrimination distribution, and light discrimination distribution.

[0018] In an embodiment of the present application, when the parameter monitoring and analysis module 11 monitors the multi-dimensional characteristics of the early stage of an arc on a target line, it respectively collects the early-stage thermal parameters, gas parameters, and optical parameters of the arc at multiple monitoring points by means of the sensor distribution configured at the multiple monitoring points of the current target line; then, according to the point coordinates of the multiple monitoring points, it arranges the collected thermal parameters, gas parameters, and optical parameters to generate a thermal parameter distribution, a gas parameter distribution, and an optical parameter distribution.

[0019] Next, a discrimination analysis is performed on the thermal parameter distribution, the gas parameter distribution, and the optical parameter distribution. For example, in the thermal parameter distribution, by randomly selecting several thermal parameter points, calculating the thermal deviation ratio of their mean value to the target thermal parameter, the thermal discrimination degree of each monitoring point is obtained, and then an overall thermal discrimination degree distribution is formed. Similarly, a discrimination analysis is performed on the gas parameter distribution and the optical parameter distribution to respectively obtain a gas discrimination degree distribution and an optical discrimination degree distribution.

[0020] Furthermore, in the device provided in the embodiment of the application, the parameter monitoring and analysis module 11 is further configured to:

[0021] Configure a sensor distribution at multiple monitoring points of the target line; use the sensor distribution to collect the early-stage thermal parameters, gas parameters, and optical parameters of the arc at the multiple monitoring points; and arrange the thermal parameters, gas parameters, and optical parameters of the multiple monitoring points according to the point coordinates of the multiple monitoring points to obtain a thermal parameter distribution, a gas parameter distribution, and an optical parameter distribution.

[0022] In an embodiment of the present application, first, a sensor distribution is configured at multiple monitoring points preset in the target line, and a dedicated sensor is configured at each monitoring point, including an infrared temperature sensor, a gas sensor, and a photodetector. Among them, the infrared temperature sensor is used to collect the temperature data of the early stage of the arc in real time, the gas sensor is used to detect the change in gas composition, and the photodetector is used to monitor the optical signal generated by the arc.

[0023] Through the configured sensors, the multi-dimensional parameters of the early stage of the arc from multiple monitoring points are collected, including thermal parameters, gas parameters, and optical parameters. The thermal parameter is an index reflecting the thermal characteristics of the arc, the gas parameter includes the component concentration of the gas, etc., and the optical parameter includes the light intensity generated by the arc. Through this process, the early-stage thermal parameters, gas parameters, and optical parameters of the arc at multiple monitoring points are obtained.

[0024] After the early-stage thermal parameters, gas parameters, and optical parameters of the arc at multiple monitoring points are collected, they are arranged according to the geographical coordinates of each monitoring point, and through the spatial grid mapping technology, the thermal parameters, gas parameters, and optical parameters are bound to specific monitoring points to generate a thermal parameter distribution, a gas parameter distribution, and an optical parameter distribution.

[0025] Furthermore, in the device provided in the embodiment of the application, the parameter monitoring and analysis module 11 is further configured to:

[0026] Select the first thermal parameter of the first monitoring point within the thermal parameter distribution, perform discrimination analysis on the first thermal parameter to obtain the first thermal discrimination; continue to perform discrimination analysis on other thermal parameters within the thermal parameter distribution to obtain the thermal discrimination distribution; continue to perform discrimination analysis on the gas parameter distribution and the optical parameter distribution to obtain the gas discrimination distribution and the optical discrimination distribution.

[0027] In the embodiment of the present application, during the process of performing discrimination analysis within the thermal parameter distribution, first select a monitoring point in the thermal parameter distribution, referred to as the first monitoring point, and its thermal parameter is referred to as the first thermal parameter. Next, within the thermal parameter distribution, randomly select several monitoring points from the surroundings, collect the thermal parameters of these points, and calculate their mean value as the reference value. Then calculate the absolute value of the difference between the first thermal parameter and the mean value, and divide it by the mean value to obtain the difference ratio between the two, that is, the thermal deviation ratio. This thermal deviation ratio reflects the abnormality degree of the first thermal parameter relative to the surrounding monitoring points, and this thermal deviation ratio is used as the first thermal discrimination of this point.

[0028] After completing the analysis of the first monitoring point, repeat the above steps for other monitoring points within the thermal parameter distribution in sequence, and calculate the thermal discrimination of each monitoring point one by one. Finally, integrate the thermal discriminations of all monitoring points to generate the thermal discrimination distribution of the entire target line.

[0029] The discrimination analysis will also be extended to data in other dimensions. In the gas parameter distribution, apply a similar method to the gas parameters of each monitoring point, calculate its difference relative to the surrounding monitoring points, and generate the gas discrimination distribution to reveal the abnormal areas of the early gas component changes in the arc. Similarly, in the optical parameter distribution, analyze the optical parameters of each monitoring point to obtain the optical discrimination distribution to reflect the abnormal areas of the arc light signal.

[0030] Through the above steps, the thermal discrimination distribution, the gas discrimination distribution, and the optical discrimination distribution are obtained respectively.

[0031] Furthermore, in the device provided in the embodiment of the application, the parameter monitoring and analysis module 11 is further configured to:

[0032] Randomly select several random thermal parameters within the thermal parameter distribution; calculate the thermal deviation ratio between the mean value of the several random thermal parameters and the first thermal parameter to obtain the first thermal discrimination.

[0033] In the embodiment of the present application, first randomly select several random thermal parameters within the thermal parameter distribution as the reference data set. Next, perform statistical processing on the selected random thermal parameters and calculate their mean value, which represents the overall characteristics of the thermal parameters in the current area.

[0034] Subsequently, the first thermal parameter of the target point is compared with the mean of the random thermal parameters, the absolute value of the difference between the first thermal parameter and the mean of the random thermal parameters is calculated, and the mean of the random thermal parameters is removed to obtain the thermal deviation ratio, and the calculated thermal deviation ratio is used as the first thermal discrimination degree.

[0035] The combination processing module 12 is configured to combine the thermal discrimination degree distribution, the gas discrimination degree distribution, and the light discrimination degree distribution to obtain a plurality of discrimination degree distribution combinations, perform discrimination degree alignment processing respectively to obtain a plurality of alignment position distributions, and calculate and obtain a thermal alignment deviation degree, a gas alignment deviation degree, and a light alignment deviation degree.

[0036] In the embodiment of the present application, the combination processing module 12 forms a plurality of discrimination degree distribution combinations by pairwise combining the thermal discrimination degree distribution, the gas discrimination degree distribution, and the light discrimination degree distribution. Each combination includes a thermal-gas discrimination degree distribution combination, a thermal-light discrimination degree distribution combination, a gas-thermal discrimination degree distribution combination, a gas-light discrimination degree distribution combination, a light-thermal discrimination degree distribution combination, and a light-gas discrimination degree distribution combination.

[0037] Next, for each discrimination degree distribution combination, discrimination degree alignment processing is performed. Taking the thermal-gas discrimination degree distribution combination as an example, first, in the thermal discrimination degree distribution, each thermal monitoring point is selected, and the deviation between it and the adjacent thermal monitoring point is calculated to obtain the thermal discrimination degree deviation distribution; similarly, in the gas discrimination degree distribution, each gas monitoring point is selected, and the deviation between it and the adjacent gas monitoring point is calculated to obtain the gas discrimination degree deviation distribution. Through these two steps of calculation, a thermal-gas alignment position distribution is generated. The same processing method is applicable to the thermal-light discrimination degree distribution combination, so as to obtain the thermal-light alignment position distribution.

[0038] Then, according to the thermal-gas alignment position distribution and the thermal-light alignment position distribution, the deviation distance between each pair of thermal monitoring points and gas monitoring points is calculated to obtain the thermal-gas monitoring point deviation degree. By calculating the mean of all thermal-gas monitoring point deviation degrees, the thermal-gas alignment deviation degree is obtained. Similarly, for the thermal-light alignment position distribution, the deviation distance between each pair of thermal monitoring points and light monitoring points is calculated to obtain the thermal-light monitoring point deviation degree, and the thermal-light alignment deviation degree is obtained by calculating its mean. Finally, the mean of the thermal-gas alignment deviation degree and the thermal-light alignment deviation degree is calculated to obtain the thermal alignment deviation degree as the spatial consistency index of the overall thermal parameter.

[0039] Similarly, the combination processing module continues to perform discrimination degree alignment processing and alignment deviation degree calculation on other discrimination degree distribution combinations, such as the gas-thermal discrimination degree distribution combination, the gas-light discrimination degree distribution combination, the light-thermal discrimination degree distribution combination, and the light-gas discrimination degree distribution combination, and finally obtains the gas alignment deviation degree and the light alignment deviation degree.

[0040] Further, as Figure 2As shown, in the device provided by the application embodiment, the combined processing module 12 is further configured to:

[0041] Pairwise combine the thermal discrimination distribution, the gas discrimination distribution, and the optical discrimination distribution to obtain multiple discrimination distribution combinations; select the thermal-gas discrimination distribution combination and the thermal-optical discrimination distribution combination within the multiple discrimination distribution combinations, and perform discrimination alignment processing respectively to obtain a thermal-gas alignment position distribution and a thermal-optical alignment position distribution; calculate a thermal alignment deviation degree according to the thermal-gas alignment position distribution and the thermal-optical alignment position distribution; continue to perform discrimination alignment processing and alignment deviation degree calculation on other gas-thermal discrimination distribution combinations, gas-optical discrimination distribution combinations, optical-thermal discrimination distribution combinations, and optical-gas discrimination distribution combinations to obtain a gas alignment deviation degree and an optical alignment deviation degree.

[0042] In the application embodiment, when processing the thermal discrimination distribution, the gas discrimination distribution, and the optical discrimination distribution, pairwise combination is first performed to associate the discrimination characteristics of different dimensions and form multiple discrimination distribution combinations. These combinations include thermal-gas discrimination distribution combinations, thermal-optical discrimination distribution combinations, gas-thermal discrimination distribution combinations, gas-optical discrimination distribution combinations, optical-thermal discrimination distribution combinations, and optical-gas discrimination distribution combinations.

[0043] Among the multiple discrimination distribution combinations obtained, select the thermal-gas discrimination distribution combination and the thermal-optical discrimination distribution combination, and perform discrimination alignment processing respectively. For the thermal-gas discrimination distribution combination, select each monitoring point in the thermal discrimination distribution, calculate its deviation from the adjacent monitoring points, and generate a thermal discrimination deviation distribution. In the gas discrimination distribution, use the same method to generate a gas discrimination deviation distribution. Then, for each thermal monitoring point in the thermal discrimination deviation distribution, calculate the absolute value of its deviation from all gas monitoring points in the gas discrimination deviation distribution, select the gas monitoring point with the smallest absolute deviation value, and establish an alignment relationship with the corresponding thermal monitoring point to generate a thermal-gas alignment position distribution. Similarly, by performing the same processing steps on the thermal-optical discrimination distribution combination, a thermal-optical alignment position distribution is generated.

[0044] Next, calculate the thermal alignment deviation degree according to the generated thermal-gas alignment position distribution and thermal-optical alignment position distribution. Specifically, first calculate the deviation distance between each pair of thermal monitoring points and gas monitoring points to obtain a thermal-gas monitoring point deviation degree, and average all the deviation degrees to obtain a thermal-gas alignment deviation degree; then perform the same calculation on the thermal-optical alignment position distribution to obtain a thermal-optical alignment deviation degree. Finally, calculate the average value of the thermal-gas alignment deviation degree and the thermal-optical alignment deviation degree to obtain the thermal alignment deviation degree, which reflects the consistency of the thermal discrimination in the multi-dimensional combination.

[0045] Meanwhile, continue to perform the same alignment processing steps on other distinctiveness distribution combinations (gas-thermal, gas-optical, optical-thermal, optical-gas), and calculate the corresponding gas alignment deviation and optical alignment deviation. These alignment deviations respectively reflect the accuracy of each type of parameter corresponding to the other two types of parameters.

[0046] Furthermore, in the device provided by the application embodiment, the combination processing module 12 is further configured to:

[0047] In the thermal distinctiveness distribution within the thermal-gas distinctiveness distribution combination, select the neighboring thermal distinctiveness of each thermal distinctiveness to obtain a plurality of neighboring thermal distinctiveness; calculate the deviation between each thermal distinctiveness and its neighboring thermal distinctiveness to obtain a thermal distinctiveness deviation distribution; in the gas distinctiveness distribution within the thermal-gas distinctiveness distribution combination, select the neighboring gas distinctiveness of each gas distinctiveness to obtain a plurality of neighboring gas distinctiveness; calculate the deviation between each gas distinctiveness and its neighboring gas distinctiveness to obtain a gas distinctiveness deviation distribution; for each thermal distinctiveness deviation at each thermal monitoring point within the thermal distinctiveness deviation distribution, calculate the difference between the plurality of gas distinctiveness deviations at all the gas monitoring points within the gas distinctiveness deviation distribution and each thermal distinctiveness deviation, select the gas monitoring point with the smallest difference, construct the alignment relationship with the thermal monitoring point, and obtain multiple pairs of alignment relationships between the thermal monitoring points and the gas monitoring points as the thermal-gas alignment position distribution; perform distinctiveness alignment processing according to the thermal-optical distinctiveness distribution combination to obtain the thermal-optical alignment position distribution.

[0048] In the embodiment of the present application, first, in the thermal distinctiveness distribution, select the neighboring thermal distinctiveness of each monitoring point. Since the monitoring points are continuous, the selection of the neighboring thermal distinctiveness is based on the spatial order, that is, starting from the current monitoring point, sequentially select the thermal distinctiveness of the next monitoring point as its neighboring thermal distinctiveness, and through this process, obtain a plurality of neighboring thermal distinctiveness. Next, calculate the deviation between each thermal distinctiveness and its neighboring thermal distinctiveness, that is, directly calculate the difference between the two, to obtain the thermal distinctiveness deviation distribution. This step helps to reveal the change trend of the thermal distinctiveness in space, especially the change of the thermal parameters between adjacent points.

[0049] Similarly, in the gas distinctiveness distribution, select the neighboring gas distinctiveness of each gas monitoring point, that is, starting from the current gas monitoring point, sequentially select the gas distinctiveness of the next gas monitoring point as its neighboring gas distinctiveness, obtain a plurality of neighboring gas distinctiveness, and calculate the deviation between each gas distinctiveness and its neighboring gas distinctiveness to obtain the gas distinctiveness deviation distribution.

[0050] Next, for each thermal monitoring point and its corresponding thermal discrimination deviation in the thermal discrimination deviation distribution, calculate the deviation difference between it and all gas monitoring points in the gas discrimination deviation distribution. Specifically, calculate the absolute value of the difference between the thermal discrimination deviation of each thermal monitoring point and the gas discrimination deviation of all gas monitoring points. By selecting the gas monitoring point with the smallest absolute value of the difference, establish the alignment relationship between the thermal monitoring point and the gas monitoring point. If the thermal monitoring point and the gas monitoring point are the same point, it indicates that the thermal parameters and gas parameters collected by these two monitoring points are relatively accurate, meaning they are spatially consistent. If the two monitoring points are not the same point and are far apart, it shows that there is a large deviation in the collection results of the thermal monitoring point and the gas monitoring point, which may be caused by sensor errors. Through this method, generate the thermal-gas alignment position distribution, representing the spatial alignment between the thermal discrimination and the gas discrimination.

[0051] For the thermal-optical discrimination distribution combination, adopt the same alignment processing method. Select the adjacent thermal discrimination of the thermal monitoring point and the adjacent optical discrimination of the optical monitoring point for comparison, and generate the thermal-optical alignment position distribution.

[0052] Furthermore, in the device provided by the application embodiment, the combination processing module 12 is further configured to:

[0053] According to the alignment relationships of multiple pairs of thermal monitoring points and gas monitoring points in the thermal-gas alignment position distribution, calculate the monitoring point deviation distances of each pair of thermal monitoring points and gas monitoring points to obtain multiple thermal-gas monitoring point deviation degrees; calculate the average value of the multiple thermal-gas monitoring point deviation degrees to obtain the thermal-gas alignment deviation degree; according to the alignment relationships of multiple pairs of thermal monitoring points and optical monitoring points in the thermal-optical alignment position distribution, calculate the monitoring point deviation distances of each pair of thermal monitoring points and optical monitoring points to obtain multiple thermal-optical monitoring point deviation degrees; calculate the average value of the multiple thermal-optical monitoring point deviation degrees to obtain the thermal-optical alignment deviation degree; calculate the average value of the thermal-gas alignment deviation degree and the thermal-optical alignment deviation degree to obtain the thermal alignment deviation degree.

[0054] In the embodiment of the present application, first, according to the alignment relationships of multiple pairs of thermal monitoring points and gas monitoring points in the thermal-gas alignment position distribution, calculate the monitoring point deviation distance between each thermal monitoring point and the corresponding gas monitoring point pair by pair. The calculation of the deviation distance is based on whether the thermal monitoring point and the gas monitoring point are the same monitoring point. If they are the same monitoring point, it indicates that the collection results of the thermal parameters and the gas parameters are highly corresponding, and at this time, the deviation distance is 0; if they are not the same monitoring point, the deviation distance is measured according to their spatial offset amount, for example, it can be directly represented by the offset number in the monitoring point sequence (such as if the number of offset monitoring points is 1, the deviation distance is 1). In this way, calculate the thermal-gas monitoring point deviation degrees of each pair of thermal monitoring points and gas monitoring points, and take the average value of all deviation degrees to generate the overall thermal-gas alignment deviation degree.

[0055] Next, for the thermal-optical alignment position distribution, the same logic is adopted. The monitoring point deviation distances between the thermal monitoring points and the corresponding optical monitoring points are calculated pair by pair. If the thermal monitoring point and the optical monitoring point are the same monitoring point, the deviation distance is 0; otherwise, the deviation distance is calculated according to the number of offsets in the monitoring point sequence. After summarizing these deviation distances, multiple thermal-optical monitoring point deviation degrees are generated, and by calculating their mean value, the thermal-optical alignment deviation degree is obtained.

[0056] Finally, by combining the thermal-gas alignment deviation degree and the thermal-optical alignment deviation degree and calculating their mean value, the final thermal alignment deviation degree is generated.

[0057] The arc prediction module 13 is used to perform arc prediction respectively according to the thermal parameter distribution, gas parameter distribution and optical parameter distribution, and obtain a set of predicted arc parameters.

[0058] In the embodiment of the present application, the arc prediction module 13 collects a set of sample thermal parameter distributions, a set of sample gas parameter distributions and a set of sample optical parameter distributions based on the arc monitoring data within the historical time, and at the same time collects the arc parameters when the arc is generated to form a set of arc parameters. Using these data, the thermal arc prediction path, gas arc prediction path and optical arc prediction path are trained respectively, and after integration, an arc predictor is formed. Finally, the current thermal parameter distribution, gas parameter distribution and optical parameter distribution are input into the arc predictor to obtain the thermal predicted arc parameters, gas predicted arc parameters and optical predicted arc parameters, and the thermal predicted arc parameters, gas predicted arc parameters and optical predicted arc parameters together form a set of predicted arc parameters.

[0059] Furthermore, in the device provided by the embodiment of the application, the arc prediction module 13 is further used for:

[0060] Collecting a set of sample thermal parameter distributions, a set of sample gas parameter distributions and a set of sample optical parameter distributions according to the early arc monitoring data within the historical time, and collecting the arc parameters when the arc is generated to obtain a set of arc parameters; respectively using the set of sample thermal parameter distributions, the set of sample gas parameter distributions and the set of sample optical parameter distributions as input data, and using the set of arc parameters as output supervision data to train the thermal arc prediction path, gas arc prediction path and optical arc prediction path; integrating the trained thermal arc prediction path, gas arc prediction path and optical arc prediction path to obtain an arc predictor; respectively inputting the thermal parameter distribution, gas parameter distribution and optical parameter distribution into the arc predictor, predicting and outputting to obtain the thermal predicted arc parameters, gas predicted arc parameters and optical predicted arc parameters, and using the thermal predicted arc parameters, gas predicted arc parameters and optical predicted arc parameters as the set of predicted arc parameters.

[0061] In the embodiments of the present application, first, sample data is extracted from historical monitoring data, and a sample thermal parameter distribution set, a sample gas parameter distribution set, and a sample optical parameter distribution set are respectively constructed. These sets respectively include the thermal, gas, and optical parameter distributions during the occurrence of an electric arc. For example, the sample thermal parameter distribution set contains the temperature change data during the occurrence of the electric arc, the sample gas parameter distribution set includes the data of gas concentration changes, and the sample optical parameter distribution set records the light intensity and wavelength characteristics during the electric arc discharge. At the same time, the corresponding arc parameters during the occurrence of the electric arc are collected from the historical monitoring data, including the intensity, duration, etc. of the electric arc, to form an arc parameter set.

[0062] Next, the sample thermal parameter distribution set, the sample gas parameter distribution set, and the sample optical parameter distribution set are respectively used as input data, and the arc parameter set is used as output supervision data to train the thermal arc prediction path, the gas arc prediction path, and the optical arc prediction path. Specifically, during the training process, the sample thermal parameter distribution set, the sample gas parameter distribution set, and the sample optical parameter distribution set are respectively used as inputs, and the arc parameter set is used as the output. Using a supervised learning method, such as a support vector machine, the thermal arc prediction path, the gas arc prediction path, and the optical arc prediction path are respectively trained. After the training is completed, the trained thermal arc prediction path, the gas arc prediction path, and the optical arc prediction path are integrated into a unified arc predictor, which can simultaneously process the data inputs in the three dimensions of heat, gas, and light.

[0063] Finally, the previously obtained thermal parameter distribution, gas parameter distribution, and optical parameter distribution are respectively input into the arc predictor, and the predicted output obtains the thermal predicted arc parameters, the gas predicted arc parameters, and the optical predicted arc parameters. The thermal predicted arc parameters, the gas predicted arc parameters, and the optical predicted arc parameters are used as the predicted arc parameter set.

[0064] The monitoring and recognition result acquisition module 14 is used to perform a weighted calculation on the predicted arc parameter set according to the thermal alignment deviation degree, the gas alignment deviation degree, and the optical alignment deviation degree to obtain arc parameters as the monitoring and recognition result.

[0065] In the embodiments of the present application, when the monitoring and recognition result acquisition module 14 performs a weighted calculation on the thermal predicted arc parameters, the gas predicted arc parameters, and the optical predicted arc parameters according to the thermal alignment deviation degree, the gas alignment deviation degree, and the optical alignment deviation degree, first, according to the thermal alignment deviation degree, the gas alignment deviation degree, and the optical alignment deviation degree, the ratio of each of them to the total deviation degree is calculated to obtain three deviation ratios. Subsequently, the reciprocal of each deviation ratio is calculated, and each reciprocal is divided by the sum of the reciprocals of the three deviation ratios to generate the thermal prediction weight, the gas prediction weight, and the optical prediction weight.

[0066] Next, the calculated thermal prediction weight, gas prediction weight, and light prediction weight are used to perform weighted calculations on the thermal prediction arc parameters, gas prediction arc parameters, and light prediction arc parameters. By multiplying the thermal prediction arc parameters by the thermal prediction weight, the gas prediction arc parameters by the gas prediction weight, the light prediction arc parameters by the light prediction weight, and adding these three parts together, comprehensive arc parameters are generated.

[0067] Finally, the obtained arc parameters are used as the monitoring and identification results.

[0068] Furthermore, in the device provided by the application embodiment, the monitoring and identification result acquisition module 14 is further configured to:

[0069] Calculate the ratios of the thermal alignment deviation degree, gas alignment deviation degree, and light alignment deviation degree to the sum of the thermal alignment deviation degree, gas alignment deviation degree, and light alignment deviation degree respectively to obtain three deviation ratios; calculate the ratios of the reciprocals of each deviation ratio to the sum of the reciprocals of the three deviation ratios to obtain the thermal prediction weight, gas prediction weight, and light prediction weight; use the thermal prediction weight, gas prediction weight, and light prediction weight to perform weighted calculations on the set of predicted arc parameters to obtain arc parameters as the monitoring and identification results.

[0070] In the embodiment of the present application, first, normalization processing is performed on the thermal alignment deviation degree, gas alignment deviation degree, and light alignment deviation degree. By calculating the ratio of each deviation degree to the sum of all deviation degrees, three deviation ratios are obtained.

[0071] Subsequently, the reciprocal of each deviation ratio is taken to reflect that the data dimension with a smaller deviation has higher importance. Then, each reciprocal is divided by the sum of the three reciprocals to generate the thermal prediction weight, gas prediction weight, and light prediction weight.

[0072] After the weight calculation is completed, weighted calculations are performed on the thermal prediction arc parameters, gas prediction arc parameters, and light prediction arc parameters. By multiplying the thermal prediction arc parameters by the thermal prediction weight, the gas prediction arc parameters by the gas prediction weight, the light prediction arc parameters by the light prediction weight, and finally adding these three weighted results together, the final arc parameters are generated. This comprehensive arc parameter combines the prediction information of the three-dimensional data of heat, gas, and light, accurately reflecting the possibility and intensity of arc occurrence. Finally, the generated arc parameters are used as the monitoring and identification results for the monitoring and identification of early signs of electrical fires. Through this process, the risk of arc faults can be effectively evaluated, and the risk of fire occurrence can be determined based on the intensity and possibility of arc occurrence. If the risk of arc faults is relatively high, the system can issue an early warning in a timely manner to help take protective measures to prevent the occurrence of fires.

[0073] In the embodiment of the present application, in summary, the embodiment of the present application has at least the following technical effects:

[0074] This application monitors the multi-dimensional characteristics of early-stage arcs in the target line, obtains the thermal parameter distribution, gas parameter distribution, and optical parameter distribution, respectively conducts discrimination analysis to obtain the thermal discrimination distribution, gas discrimination distribution, and optical discrimination distribution; combines the thermal discrimination distribution, gas discrimination distribution, and optical discrimination distribution to obtain multiple discrimination distribution combinations, respectively conducts discrimination alignment processing to obtain multiple alignment position distributions, and calculates and obtains the thermal alignment deviation degree, gas alignment deviation degree, and optical alignment deviation degree; respectively conducts arc prediction based on the thermal parameter distribution, gas parameter distribution, and optical parameter distribution to obtain the thermal predicted arc parameters, gas predicted arc parameters, and optical predicted arc parameters; based on the thermal alignment deviation degree, gas alignment deviation degree, and optical alignment deviation degree, conducts weighted calculation on the set of predicted arc parameters to obtain the arc parameters as the monitoring and recognition result. The present invention solves the technical problems of low accuracy in early identification of electrical fire faults and lag in early warning response in the prior art. Through multi-dimensional parameter monitoring and discrimination analysis, it generates the parameter distributions and alignment deviation degrees of heat, gas, and light, combines machine learning to predict the heat, gas, and optical arc parameters, and conducts weighted calculation of the comprehensive arc parameters to achieve accurate identification and early warning of arc faults, and achieves the technical effect of improving the accuracy and reliability of electrical fire fault identification.

[0075] It should be noted that the above sequence of embodiments of the present application is only for description and does not represent the superiority or inferiority of the embodiments. And the above describes specific embodiments of this specification. The processes depicted in the drawings do not necessarily require the specific order and continuous order shown to achieve the desired result. In some embodiments, multi-tasking and parallel processing are also possible or may be advantageous.

[0076] The above are only the preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included within the protection scope of the present application.

[0077] This specification and the drawings are only exemplary descriptions of the present application and are considered to have covered any and all modifications, variations, combinations, or equivalents within the scope of the present application. Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalent technologies, the present application is intended to include these changes and modifications.

Claims

1. An electrical fire early sign monitoring and identification device with multi-fault linkage simulation, characterized in that: The device comprises: The parameter monitoring and analysis module is used to monitor the early multi-dimensional characteristics of the arc on the target line, obtain the thermal parameter distribution, gas parameter distribution and optical parameter distribution, and perform discrimination analysis to obtain the thermal discrimination distribution, gas discrimination distribution and optical discrimination distribution; A combination processing module, used for combining the thermal distinctiveness distribution, the gas distinctiveness distribution and the light distinctiveness distribution to obtain a plurality of distinctiveness distribution combinations, performing distinctiveness alignment processing respectively to obtain a plurality of alignment position distributions, and calculating and obtaining a thermal alignment deviation, a gas alignment deviation and a light alignment deviation; An arc prediction module, used to perform arc prediction according to the thermal parameter distribution, gas parameter distribution and light parameter distribution, respectively, to obtain a predicted arc parameter set; The monitoring and identification result acquisition module is used to perform weighted calculation on the predicted arc parameter set according to the thermal alignment deviation, the gas alignment deviation and the light alignment deviation to obtain arc parameters as monitoring and identification results.

2. The electrical fire early sign monitoring and identification device with multi-fault linkage simulation according to claim 1 is characterized in that: Parameter monitoring and analysis module, used for: Configure sensor distribution at multiple monitoring points on the target line; The sensors are distributed to collect the thermal parameters, gas parameters and optical parameters of the arc at the early stage at the plurality of monitoring points; According to the point coordinates of the multiple monitoring points, the thermal parameters, gas parameters and light parameters of the multiple monitoring points are arranged to obtain the thermal parameter distribution, gas parameter distribution and light parameter distribution.

3. The electrical fire early sign monitoring and identification device with multi-fault linkage simulation according to claim 2 is characterized in that: Parameter monitoring and analysis module, used for: Selecting a first thermal parameter of a first monitoring point in the thermal parameter distribution, performing a discrimination analysis on the first thermal parameter, and obtaining a first thermal discrimination; Continuing to perform discrimination analysis on other thermal parameters in the thermal parameter distribution to obtain a thermal discrimination distribution; Continue to perform discrimination analysis on the gas parameter distribution and the light parameter distribution to obtain gas discrimination distribution and light discrimination distribution.

4. The electrical fire early sign monitoring and identification device with multi-fault linkage simulation according to claim 3 is characterized in that: Parameter monitoring and analysis module, used for: randomly selecting a number of random thermal parameters within the thermal parameter distribution; The thermal deviation ratio of the mean value of the plurality of random thermal parameters to the first thermal parameter is calculated to obtain a first thermal distinctiveness.

5. The electrical fire early sign monitoring and identification device with multi-fault linkage simulation according to claim 1 is characterized in that: Combined processing modules for: Combining the thermal discrimination distribution, the gas discrimination distribution and the light discrimination distribution in pairs to obtain a plurality of discrimination distribution combinations; Selecting a heat-gas distinguishability distribution combination and a heat-light distinguishability distribution combination from among the plurality of distinguishability distribution combinations, and performing distinguishability alignment processing on each combination to obtain a heat-gas aligned position distribution and a heat-light aligned position distribution; Calculating a thermal alignment deviation according to the thermal-gas alignment position distribution and the thermal-optical alignment position distribution; Continue to perform discrimination alignment processing and alignment deviation calculation on other gas-heat discrimination distribution combinations, gas-light discrimination distribution combinations, light-heat discrimination distribution combinations, and light-gas discrimination distribution combinations to obtain gas alignment deviation and light alignment deviation.

6. The electrical fire early sign monitoring and identification device with multi-fault linkage simulation according to claim 5 is characterized in that: Combined processing modules for: In the heat discrimination distribution within the heat-gas discrimination distribution combination, selecting the adjacent heat discrimination of each heat discrimination to obtain a plurality of adjacent heat discriminations; Calculate the deviation between each thermal discrimination and the adjacent thermal discrimination to obtain the thermal discrimination deviation distribution; In the gas discrimination degree distribution within the heat-gas discrimination degree distribution combination, selecting the adjacent gas discrimination degree of each gas discrimination degree to obtain a plurality of adjacent gas discrimination degrees; Calculate the deviation between each air discrimination and the adjacent air discrimination to obtain the air discrimination deviation distribution; For each thermal discrimination deviation of each thermal monitoring point in the thermal discrimination deviation distribution, the difference between the multiple gas discrimination deviations of all the multiple gas monitoring points in the gas discrimination deviation distribution and each thermal discrimination deviation is calculated, and the gas monitoring point with the smallest difference is selected to establish an alignment relationship with the thermal monitoring point, so as to obtain alignment relationships between multiple pairs of thermal monitoring points and gas monitoring points as a thermal-gas alignment position distribution; According to the combination of the thermal-optical distinctiveness distribution, distinctiveness alignment processing is performed to obtain the thermal-optical alignment position distribution.

7. The electrical fire early sign monitoring and identification device with multi-fault linkage simulation according to claim 6 is characterized in that: Combined processing modules for: Calculating the monitoring point deviation distance of each pair of thermal monitoring points and gas monitoring points according to the alignment relationship of multiple pairs of thermal monitoring points and gas monitoring points in the thermal-gas alignment position distribution to obtain multiple thermal-gas monitoring point deviation degrees; Calculating the average of the deviations of the plurality of heat-gas monitoring points to obtain a heat-gas alignment deviation; Calculating the monitoring point deviation distance of each pair of thermal monitoring points and optical monitoring points according to the alignment relationship of multiple pairs of thermal monitoring points and optical monitoring points in the thermal-optical alignment position distribution to obtain multiple thermal-optical monitoring point deviation degrees; Calculating the average of the deviations of the plurality of thermal-optical monitoring points to obtain a thermal-optical alignment deviation; The average of the heat-gas alignment deviation and the heat-light alignment deviation is calculated to obtain the heat alignment deviation.

8. The electrical fire early sign monitoring and identification device with multi-fault linkage simulation according to claim 1 is characterized in that: Arc prediction module for: According to the arc early monitoring data in the historical time, a sample thermal parameter distribution set, a sample gas parameter distribution set and a sample light parameter distribution set are collected, and arc parameters when the arc is generated are collected to obtain an arc parameter set; The sample thermal parameter distribution set, the sample gas parameter distribution set and the sample light parameter distribution set are respectively used as input data, and the arc parameter set is used as output supervision data to train the thermal arc prediction path, the gas arc prediction path and the light arc prediction path; Integrate the trained thermal arc prediction path, gas arc prediction path and photoelectric arc prediction path to obtain an arc predictor; The thermal parameter distribution, gas parameter distribution and light parameter distribution are respectively input into the arc predictor, and the prediction output obtains the thermal predicted arc parameters, gas predicted arc parameters and light predicted arc parameters, and the thermal predicted arc parameters, gas predicted arc parameters and light predicted arc parameters are used as the predicted arc parameter set.

9. The electrical fire early sign monitoring and identification device with multi-fault linkage simulation according to claim 1 is characterized in that: Monitoring and recognition result acquisition module, used for: Respectively calculating the ratios of the thermal alignment deviation, the gas alignment deviation and the light alignment deviation to the sum of the thermal alignment deviation, the gas alignment deviation and the light alignment deviation to obtain three deviation ratios; Calculating the ratio of the inverse of each deviation ratio to the sum of the inverses of the three deviation ratios to obtain a heat prediction weight, a gas prediction weight, and a light prediction weight; The heat prediction weight, the gas prediction weight and the light prediction weight are used to perform weighted calculation on the predicted arc parameter set to obtain arc parameters as monitoring and identification results.

Images