Intelligent detection system based on low-voltage line

Abstract

The invention discloses an intelligent detection system based on a low-voltage line, and the system comprises a multi-modal data collection module which is used for collecting the multi-modal data of the low-voltage line in real time; the signal transmission verification module is used for transmitting the data converted into the digital signal to the data preprocessing module through an internal channel; the data preprocessing module is used for eliminating abnormal data and uniformly mapping data left after elimination to an interval of [0, 1]; the fault prediction analysis module is used for fusing the mapped voltage data, current data and surrounding environment data to obtain comprehensive characteristics, predicting the fault occurrence probability of the low-voltage line by using a chaos mixed model based on the comprehensive characteristics, and determining that the low-voltage line is in a fault state when the probability exceeds a preset threshold value theta old; the threshold value dynamic adjusting module is used for dynamically adjusting a preset threshold value; and the state feedback warning module is used for displaying a key event and a fault state through an indicator light when the key event and the fault state are detected, and reporting the key event and the fault state to a monitoring center.

Description

Technical Field

[0001] This application relates to the technical field of power detection, and particularly to an intelligent detection system based on low-voltage lines. Background Art

[0002] In the field of power systems, low-voltage lines, as important infrastructure for power transmission and distribution, their operating status is directly related to the safety and stability of the entire power system. With the acceleration of urbanization and the continuous growth of power demand, the complexity and importance of low-voltage lines have become increasingly prominent. To ensure the reliable operation of the power system, real-time monitoring and fault prediction of low-voltage lines have become a crucial task. In recent years, with the rapid development of Internet of Things, big data, and artificial intelligence technologies, it has provided strong technical support for the research and development of intelligent detection systems for low-voltage lines.

[0003] However, traditional low-voltage line detection technologies have many deficiencies. On the one hand, traditional technologies mainly rely on manual inspections and regular detections. This method is not only inefficient but also difficult to achieve real-time monitoring of the line operating status. On the other hand, traditional technologies often rely on empirical judgments and simple statistics in fault prediction, lacking scientificity and accuracy. In addition, traditional technologies are also easily interfered by environmental factors, resulting in unstable and unreliable detection results. These problems seriously affect the safe operation of low-voltage lines and the efficiency of fault handling, and there is an urgent need for a new intelligent detection system to solve these problems. Summary of the Invention

[0004] In view of this, the embodiments of this application provide an intelligent detection system based on low-voltage lines. By integrating multiple modules such as multi-modal data acquisition, signal transmission verification, data preprocessing, fault prediction analysis, threshold dynamic adjustment, and status feedback warning, it realizes comprehensive and real-time monitoring of the operating status of low-voltage lines. This system can accurately predict the probability of faults, determine the type of faults, and improve the accuracy of detection through adaptive threshold adjustment. At the same time, the status feedback warning module ensures timely reporting of key events, providing strong guarantee for the stable operation of the power system.

[0005] According to one aspect of this application, an intelligent detection system based on low-voltage lines is provided, including: a multi-modal data acquisition module, a signal transmission verification module, a data preprocessing module, a fault prediction analysis module, a threshold dynamic adjustment module, and a status feedback warning module;

[0006] The multi-modal data acquisition module is used to collect multi-modal data of low-voltage lines in real time, including: measuring voltage data through a voltage sensor, measuring current data in different current ranges through a current transformer, collecting line operating data and surrounding environment data through various environmental sensors, and converting all the collected data from analog signals to digital signals;

[0007] The signal transmission verification module is used to transmit the data converted into digital signals to the data preprocessing module through an internal channel by means of differential transmission technology. Among them, a check bit is added to the transmitted data at the transmitting end, and the integrity of the transmitted data is verified according to the check bit at the receiving end;

[0008] The data preprocessing module is used to identify and eliminate abnormal data caused by sensor failures and instantaneous strong electromagnetic interference based on the local outlier factor, and use a nonlinear transformation function to uniformly map the remaining data after elimination to the interval [0, 1];

[0009] The fault prediction and analysis module is used to fuse the mapped voltage data, current data, and surrounding environment data by means of a multi-modal data fusion feature extraction formula to obtain comprehensive features, and predict the probability of a fault occurring in the low-voltage line based on the comprehensive features using a chaotic hybrid model. When the probability exceeds the preset threshold θ old it is determined that the low-voltage line is in a fault state, and the type of line fault is judged according to the comprehensive features;

[0010] The threshold dynamic adjustment module is used to dynamically adjust the preset threshold based on the real-time collected multi-modal data and historical multi-modal data using an adaptive threshold dynamic adjustment formula;

[0011] The status feedback warning module is provided with an indicator light, which is used to detect key events based on voltage data, current data, and line operation data. When a key event and a fault state are detected, it is presented through the indicator light, reported to the monitoring center and relevant personnel, and the event record is stored at the same time.

[0012] By means of the above technical solutions, an intelligent detection system based on low-voltage lines provided by an embodiment of the present application has the following beneficial effects:

[0013] First, by constructing a multi-modal data acquisition module, the present application realizes a comprehensive monitoring of the operation state of low-voltage lines. It can not only accurately measure three-phase voltage and current, but also integrates various environmental sensors such as temperature sensors, humidity sensors, and line vibration sensors, and can collect line operation and surrounding environment data in real time. This multi-modal data acquisition method provides a rich and accurate data basis for subsequent fault prediction and analysis. At the same time, through the signal transmission verification module and the data preprocessing module, the present application effectively guarantees the integrity of data transmission and the accuracy of data processing, and further improves the reliability and stability of the system.

[0014] Second, by introducing a multi-modal data fusion feature extraction formula and a chaotic mixing model, this application can accurately predict the probability of a fault occurring in a low-voltage line and accurately determine the specific type of the line fault. This fault prediction method based on multi-modal data fusion and chaos theory not only improves the accuracy and reliability of fault prediction but also provides a scientific basis for the preventive maintenance and fault handling of the line. In addition, this application also sets up a threshold dynamic adjustment module, which can dynamically adjust the detection threshold according to real-time and historical data, so as to adapt to the line operation state under different working conditions. This dynamic adjustment mechanism enables the intelligent detection system of this application to more flexibly handle various complex situations and further improves the practicality and adaptability of the system.

[0015] The above description is only an overview of the technical solution of this application. In order to be able to understand the technical means of this application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of this application more obvious and understandable, the following specific embodiments of this application are specifically given. Brief Description of the Drawings

[0016] The drawings described herein are used to provide a further understanding of this application and constitute a part of this application. The schematic embodiments of this application and their descriptions are used to explain this application and do not constitute an improper limitation of this application. In the drawings:

[0017] Figure 1 It shows a schematic structural diagram of an intelligent detection system based on a low-voltage line provided by an embodiment of this application. Detailed Description of the Embodiment

[0018] In the following, this application will be described in detail with reference to the drawings and in combination with the embodiments. It should be noted that, without conflict, the embodiments and features in the embodiments of this application can be combined with each other.

[0019] In this embodiment, an intelligent detection system based on a low-voltage line is provided, as Figure 1 shown, including: a multi-modal data acquisition module, a signal transmission verification module, a data preprocessing module, a fault prediction and analysis module, a threshold dynamic adjustment module, and a status feedback warning module.

[0020] The multi-modal data acquisition module is used to collect multi-modal data of the low-voltage line in real time, including: measuring voltage data through a voltage sensor, measuring current data in different current ranges through a current transformer, collecting line operation data and surrounding environment data through various environmental sensors, and converting all the collected data from analog signals to digital signals;

[0021] The signal transmission verification module is used to transmit the data converted into digital signals to the data preprocessing module through an internal channel by means of differential transmission technology. Among them, a verification bit is added to the transmitted data at the transmission end, and the integrity of the transmitted data is verified according to the verification bit at the receiving end;

[0022] The data preprocessing module is used to identify and remove abnormal data caused by sensor failures and instantaneous strong electromagnetic interference based on the local outlier factor, and uniformly map the remaining data after removal to the [0, 1] interval by using a non-linear transformation function;

[0023] The fault prediction and analysis module is used to fuse the mapped voltage data, current data, and surrounding environment data by means of a multi-modal data fusion feature extraction formula to obtain comprehensive features, and based on the comprehensive features, use a chaotic hybrid model to predict the probability of a fault occurring in the low-voltage line. When the probability exceeds the preset threshold θ old it is determined that the low-voltage line is in a fault state, and the type of line fault is judged according to the comprehensive features;

[0024] The threshold dynamic adjustment module is used to dynamically adjust the preset threshold based on the real-time collected multi-modal data and historical multi-modal data by using an adaptive threshold dynamic adjustment formula;

[0025] The status feedback warning module is provided with an indicator light, which is used to detect key events according to the voltage data, current data, and line operation data. When a key event and a fault state are detected, it is presented through the indicator light, reported to the monitoring center and relevant personnel, and the event record is stored at the same time.

[0026] In an alternative embodiment, the low-voltage line in the urban residential area is detected by the above-mentioned intelligent detection system based on the low-voltage line. In a medium-sized urban residential area, the power supply is crucial for the daily life of residents, and an intelligent detection system based on the low-voltage line is installed here to ensure the stable operation of the line.

[0027] The voltage sensor accurately measures the three-phase voltage within the measurement range of 0.7Un ≤ U ≤ 1.2Un (±1V), and monitors in real time whether the voltage of the community power grid is stable within the normal range, ensuring that the residential electrical equipment will not be damaged due to abnormal voltage. For example, during the peak electricity consumption period in summer, the voltage sensor can timely detect whether the voltage has a downward trend to ensure the normal operation of high-power electrical appliances.

[0028] The current transformer measures the three-phase current in different current ranges within the measurement ranges of 5 - 50A (±0.5A) and 50 - 600A (±1%). When a resident suddenly increases the use of high-power electrical appliances at home, the current transformer can accurately capture the current change and prevent the line from failing due to overcurrent.

[0029] The temperature sensor in the environmental sensor monitors the ambient temperature around the line, the humidity sensor detects the humidity, and the line vibration sensor can sense whether the line generates abnormal vibration due to external forces (such as wind blowing, tree branches touching). In the hot summer, the temperature sensor can detect the increase in temperature near the line. If the temperature approaches or exceeds the critical temperature for the safe operation of the line, the system can give an early warning.

[0030] The signal transmission verification module uses differential transmission technology to efficiently transmit the data collected by each sensor to the data preprocessing module through an internal high-speed and anti-interference channel. During the transmission process, a verification bit is added to each data packet, just like attaching a unique "identity label" to each "data package". At the receiving end, the system carefully verifies the integrity of the data based on the verification bit. If the data is changed due to external interference during the transmission process, the receiving end can quickly detect the problem by virtue of the verification mechanism and request retransmission, thus ensuring the accuracy of the data and providing a reliable basis for subsequent analysis and judgment.

[0031] The data preprocessing module first identifies and eliminates abnormal data. For example, at a certain moment, due to the instantaneous strong electromagnetic interference generated by nearby construction, some sensor data shows abnormal fluctuations. The system will identify and eliminate abnormal data based on the local outlier factor, screening out and eliminating these data that significantly deviate from the normal range. For each data type of voltage data, current data, temperature data, and humidity data, for each data point x i , calculate its local reachability density ρ i and the local outlier factor LOF i , and the formula is: where k is the preset number of neighboring points, N k (x i ) represents the set composed of the k nearest neighbors of the data point x i , d reach (x i , x j ) is the reachable distance from x i to x j , and the formula is: d reach (x i , x j ) = max{d(x i , x j ), k - dist(x j )}, d(x i , x j ) is the Euclidean distance between x i and x j , k - dist(x j ) is the distance from x j to its k-th nearest neighbor, and the calculation formula for the local outlier factor LOF i is: When the LOF i exceeds the preset local outlier threshold T lof the data point is determined to be abnormal data and removed from the dataset. After that, for the differences in the dimensions and value ranges of different data, then, using a non-linear transformation function, the data with different dimensions and value ranges are uniformly mapped to the interval [0, 1]. For a set of data X = {x1, x2, …, x n}, calculate its minimum value x min , maximum value x max and median x med . Define the non-linear transformation function f(x), and the calculation formula is: where β is an adjustable parameter determined by actual data, β takes values between 2 and 5, e is the natural logarithm, and the normalized data y i The calculation formula is: so that various types of data can be analyzed and compared under the same standard, effectively improving the accuracy of fault prediction.

[0032] The fault prediction analysis module will fuse voltage, current, temperature, and humidity multimodal data by means of a multimodal data fusion feature extraction formula and extract comprehensive features. Let the voltage be V, the current be I, the temperature be T, and the humidity be H. The calculation formula is: where x1 = V, x2 = I, x3 = T, x4 = H, ω i is the weight coefficient corresponding to the data type, satisfying is the mean value of the i-th type of data, is the standard deviation of the i-th type of data, is an exponential parameter set according to the characteristics of the data type. Use the chaotic hybrid model to analyze and predict the probability P fault of a low-voltage line failure, and the calculation formula is: where P fault is the probability of failure, m represents the number of multimodal fusion feature values, M j-τ is the multimodal fusion feature value at time point j after time delay τ, α j is the feature weight corresponding to time point j, β1 is the chaotic term coefficient, and Chaos(j) is a chaotic sequence generated based on the Logistic map. After transformation, a chaotic sequence value related to the multimodal fusion feature value at time point j is obtained. Assume that the calculated probability of failure P fault , when it exceeds the preset threshold θ old, the specific type of line fault can be further determined by combining the extracted features. For example, within a certain period of time, if the system detects that the voltage remains low continuously, the current increases to a certain extent, and the temperature also rises slightly, by combining these features, the system may determine that there is an overload and voltage drop problem in the line, and it is very likely that there is poor contact or potential local short - circuit in a certain section of the line, so as to take preventive and repair measures in advance.

[0033] In addition, for the preset threshold θ old , it can be obtained in the following way: collect historical multi - modal data of low - voltage lines under different working conditions, and calculate the historical fault occurrence probability P′ of each working condition through a chaotic mixing model fault , calculate the mean value of the probability values and the standard deviation σ P . The formula is: where n is the number of working conditions, is the historical fault occurrence probability of the i - th working condition; calculate the preset threshold as θ old . The calculation formula is: where k0 is an empirical coefficient, and its value ranges from 1 to 3. Of course, the preset threshold can also be determined based on preset empirical values, which are not limited here.

[0034] The threshold dynamic adjustment module will collect real - time multi - modal data and historical multi - modal data. In the case where the electricity load in residential areas changes significantly with seasons and time, such as when a large number of heating devices are enabled in winter, the line current and power increase significantly. The system will use an adaptive adjustment mechanism to dynamically change the detection threshold according to these changes. The calculation formula is: where, θ new is the adjusted preset threshold, θ old is the preset threshold before adjustment, γ is the adjustment coefficient, which controls the amplitude of threshold adjustment, ΔM is the change amount of each group of adjacent - time multi - modal fusion feature values among multiple multi - modal fusion feature values within a preset time period, is the mean value of the change amounts of each group of adjacent - time multi - modal fusion feature values among multiple multi - modal fusion feature values within a preset time period, and δ is an exponential parameter dynamically adjusted according to environmental stability factors, making the preset threshold more suitable for the actual operation situation and reducing false alarms and missed detections.

[0035] The status feedback warning module is provided with a variety of indicator lights, including overvoltage, undervoltage, overcurrent warning indicator lights corresponding to each of the three phases A, B, and C in the low-voltage line, warning indicator lights for abnormal line vibration events, status indicator lights for the fault status of the low-voltage line, as well as power indicator lights and communication indicator lights. The key events presented by the indicator lights include overvoltage events, undervoltage events, overcurrent events corresponding to each of the three phases A, B, and C in the low-voltage line, and abnormal line vibration events. Among them, the abnormal line vibration event refers to an event where the line vibration data is greater than a preset vibration threshold. When the system detects that the voltage of a certain phase is too high, the corresponding overvoltage indicator light immediately lights up. At the same time, the system will quickly report the detailed event information to the monitoring center of the community property management office and the mobile terminals of relevant maintenance personnel through a preset wireless method, such as Wi-Fi or 4G network, and automatically store the event record. After receiving the information, the maintenance personnel can conduct inspections and repairs in a timely manner to ensure that the safety of residents' electricity use is not affected.

[0036] In summary, in the scenario of low-voltage line detection in urban residential areas, the intelligent detection system of the present application demonstrates strong practicability. Through the precise monitoring of voltage, current, temperature, and humidity data by the multimodal data acquisition module, it can promptly capture line abnormalities during peak electricity consumption and special situations. The signal transmission verification module ensures accurate data transmission. The data preprocessing module eliminates abnormal data and standardizes the data range. The fault prediction analysis module and the threshold dynamic adjustment module effectively cooperate to accurately judge faults and reasonably adjust the threshold. The status feedback warning module promptly notifies relevant personnel to handle problems. Overall, the system greatly improves the safety and stability of low-voltage lines in residential areas, ensures worry-free electricity use for residents, and provides an efficient and reliable solution for community power management.

[0037] In another alternative embodiment, the low-voltage lines in industrial factories are detected by the above intelligent detection system based on low-voltage lines. In an industrial factory with multiple production lines, the production process highly depends on stable power supply, and this low-voltage line intelligent detection system plays a crucial role.

[0038] The voltage sensor closely monitors the three-phase voltage, and its measurement range ensures that it can accurately capture the voltage fluctuations that may occur during industrial production. For example, when large equipment starts or stops, the voltage may change instantaneously, and the voltage sensor can accurately measure and provide timely feedback. The current transformer can adapt to a large range of current changes in the industrial plant, from 5 - 50 A (±0.5 A) to 50 - 600 A (±1%), and can accurately measure the working current of each device when the production line is running at full load, so as to promptly detect whether there are problems such as abnormal energy consumption or short circuits in the equipment. The temperature sensor in the environmental sensor can monitor the ambient temperature around the production line. Since heat may be generated during industrial production, the temperature sensor can detect the increase in temperature and prevent faults caused by overheating of the circuit. The humidity sensor can detect the environmental humidity to avoid a decrease in the insulation performance of the circuit due to high humidity. The line vibration sensor can sense whether the operation of the equipment has a large vibration impact on the line to ensure the stability of the line connection.

[0039] The signal transmission and verification module also uses differential transmission technology and an internal high-speed, anti-interference channel to transmit the data collected by the sensors to the data preprocessing module and adds a verification bit for data integrity verification. In a complex electromagnetic environment such as an industrial plant, there are many electromagnetic interference sources, such as large motors and welding machine equipment. The verification mechanism can effectively resist these interferences, ensure the accuracy of the data during transmission, and avoid misjudgments caused by data errors.

[0040] The data preprocessing module will identify and eliminate abnormal data. For example, at a certain moment, due to the strong electromagnetic interference generated during the operation of the welding machine, some current data becomes abnormal. The system will identify and eliminate the abnormal data caused by sensor failures and instantaneous strong electromagnetic interference based on the local outlier factor. For each data point x i , calculate its local reachability density ρ i and local outlier factor LOF i , and the formula is:

[0041] The calculation formula for the local outlier factor LOF i is: Then, use a non-linear transformation function to normalize the collected voltage, current, temperature, and humidity data. For a set of data X = {x1, x2,..., x n}, calculate its minimum value x min , maximum value x max and median x med , and define the non-linear transformation function f(x), and the calculation formula is: Linearly scale the data after non-linear transformation so that it is mapped to the [0, 1] interval. The normalized data y iThe calculation formula is as follows: It enables analysis to be carried out on the same scale, improving the accuracy of fault prediction.

[0042] The fault prediction analysis module makes a judgment by integrating multi-modal data. With the help of the multi-modal data fusion feature extraction formula, it fuses the multi-modal data of voltage, current, temperature, and humidity, and extracts comprehensive features. Let the voltage be V, the current be I, the temperature be T, and the humidity be H. The calculation formula is as follows: The probability of a fault occurring in a low-voltage line is analyzed and predicted using a chaotic hybrid model. The calculation formula is as follows: When the probability P fault exceeds the preset threshold θ old , combined with the extracted features, the specific type of line fault is judged. Suppose that during a certain production period, the system discovers that the current of a certain production line suddenly increases significantly, and at the same time the temperature rises sharply. Combining these features, the system may judge that a short-circuit fault has occurred in a certain device on the production line, and it is necessary to stop the machine for maintenance in time to avoid the expansion of the fault and affecting the entire production process.

[0043] The threshold dynamic adjustment module collects the line operation data, environmental parameters, and historical data in real time according to the production characteristics of the industrial plant, and dynamically adjusts the detection threshold. The calculation formula is as follows:

[0044] For example, when switching between different production processes, the load and operating state of the line will change significantly. The system analyzes the changes in the multi-modal fusion feature values and uses the adaptive threshold dynamic adjustment formula to adjust the threshold. During the peak production season, when the equipment runs at a high load for a long time, the system appropriately increases the threshold to adapt to the normal fluctuations in the production process and reduce false alarms. During the equipment maintenance or debugging stage, when the line load is light, the system correspondingly reduces the threshold to improve the sensitivity of fault detection.

[0045] The indicator light of the status feedback warning module can visually present the line operation and fault status. When it detects that there is an overcurrent or undervoltage problem in the line of a certain production line, the corresponding indicator light lights up, and immediately reports the detailed event information to the central monitoring room of the factory and the maintenance personnel responsible for the production line by wired means, such as industrial Ethernet. At the same time, the system stores the event record for subsequent analysis and summary, providing data support for optimizing the power supply and equipment maintenance of the production line, and ensuring the continuity and stability of industrial production.

[0046] In summary, for the detection of low-voltage lines in industrial plants, the system of this application is equally crucial. The multi-modal data acquisition module adapts to the complex industrial environment, keenly senses the changes in voltage and current and the fluctuations of environmental factors. The signal transmission and verification module ensures the integrity of data in the plant with strong electromagnetic interference. Data preprocessing lays the foundation for subsequent analysis. Fault prediction and analysis can quickly locate faults such as equipment short circuits. Threshold dynamic adjustment fits the load changes brought about by the switching of production processes. The status feedback and warning module efficiently conveys fault information. This system effectively guarantees the power continuity of industrial production, reduces the risk of equipment damage, and improves production efficiency, becoming an indispensable and powerful tool for industrial power maintenance.

[0047] By applying the technical solution of this embodiment, the following beneficial effects can be achieved:

[0048] First, by constructing a multi-modal data acquisition module, this application realizes the comprehensive monitoring of the operating state of low-voltage lines. It can not only accurately measure three-phase voltage and current, but also integrates various environmental sensors such as temperature sensors, humidity sensors, and line vibration sensors, enabling real-time collection of line operation and surrounding environment data. This multi-modal data acquisition method provides a rich and accurate data basis for subsequent fault prediction and analysis. At the same time, through the signal transmission verification module and data preprocessing module, this application effectively ensures the integrity of data transmission and the accuracy of data processing, further improving the reliability and stability of the system.

[0049] Second, by introducing the multi-modal data fusion feature extraction formula and the chaotic hybrid model, this application can accurately predict the probability of faults occurring in low-voltage lines and accurately determine the specific types of line faults. This fault prediction method based on multi-modal data fusion and chaos theory not only improves the accuracy and reliability of fault prediction, but also provides a scientific basis for the preventive maintenance and fault handling of lines. In addition, this application also sets up a threshold dynamic adjustment module, which can dynamically adjust the detection threshold according to real-time and historical data, so as to adapt to the line operating state under different working conditions. This dynamic adjustment mechanism enables the intelligent detection system of this application to more flexibly handle various complex situations, further improving the practicality and adaptability of the system.

[0050] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, 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, it should be considered as the scope recorded in this specification.

[0051] The above-described embodiments merely represent several implementation manners of the present application. The description thereof is relatively specific and detailed, but it should not be construed as a limitation to the patent scope of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.

Claims

1. An intelligent detection system based on low-voltage lines, characterized in that, Including: A multi-modal data acquisition module, a signal transmission verification module, a data preprocessing module, a fault prediction and analysis module, a threshold dynamic adjustment module, and a status feedback warning module; The multi-modal data acquisition module is used to collect multi-modal data of low-voltage lines in real time, including: measuring voltage data through a voltage sensor, measuring current data in different current ranges through a current transformer, collecting line operation data and surrounding environment data through various environmental sensors, and converting all the collected data from analog signals into digital signals; The signal transmission verification module is used to transmit the data converted into digital signals to the data preprocessing module through an internal channel by means of differential transmission technology. Among them, a check bit is added to the transmitted data at the transmission end, and the integrity of the transmitted data is verified according to the check bit at the receiving end; The data preprocessing module is used to identify and eliminate abnormal data caused by sensor failures and instantaneous strong electromagnetic interference based on the local outlier factor for voltage data, current data, and surrounding environment data, and use a non-linear transformation function to uniformly map the remaining data after elimination to the [0, 1] interval; The fault prediction and analysis module is used to fuse the mapped voltage data, current data, and surrounding environment data by means of a multi-modal data fusion feature extraction formula to obtain comprehensive features, and based on the comprehensive features, use a chaotic hybrid model to predict the probability of a fault occurring in the low-voltage line. When the probability exceeds the preset threshold θ old it is determined that the low-voltage line is in a fault state, and the type of line fault is judged according to the comprehensive features; The threshold dynamic adjustment module is used to dynamically adjust the preset threshold based on the real-time collected multi-modal data and historical multi-modal data using an adaptive threshold dynamic adjustment formula; The status feedback warning module is provided with an indicator light, which is used to detect key events based on voltage data, current data, and line operation data. When key events and fault states are detected, they are presented through the indicator light, reported to the monitoring center and relevant personnel, and event records are stored at the same time.

2. An intelligent detection system based on low-voltage lines according to claim 1, characterized in that, The measurement range of the voltage sensor in the multi-modal data acquisition module is: 0.7Un~1.2Un; the measurement range of the current transformer is: 5~600A.

3. An intelligent detection system based on a low-voltage line according to claim 1, characterized in that, The environmental sensors used in the multi-modal data acquisition module include a temperature sensor, a humidity sensor, and a line vibration sensor. The line operation data includes line vibration data, and the surrounding environment data includes the temperature data collected by the temperature sensor within the corresponding collection range and the humidity data collected by the humidity sensor within the corresponding collection range.

4. An intelligent detection system based on a low-voltage line according to claim 3, characterized in that, The data preprocessing module is specifically used to identify and eliminate abnormal data caused by sensor failures and instantaneous strong electromagnetic interference based on the local outlier factor through the following steps: For each data point x in each type of data i , calculate its local reachability density ρ i . The formula is: where k is the preset number of neighboring points, and N k (x i ) represents the set composed of the k nearest neighbors of the data point x i . And d reach (x i , x j ) is the reachability distance from x i to x j . The formula is: d reach (x i , x j ) = max{d(x i , x j ), k - dist(x j )}, where d(x i , x j ) is the Euclidean distance between x i and x j , and k - dist(x j ) is the distance from x j to its k-th nearest neighbor. Calculate the local outlier factor LOF i according to the local reachability density ρ i . The calculation formula for the local outlier factor LOF i is: When LOF i exceeds the preset local outlier threshold T lof , the data point x i is determined as an abnormal data and removed from the data set.

5. An intelligent detection system based on a low-voltage line according to claim 3, characterized in that, The data preprocessing module is specifically used to uniformly map the remaining data after elimination to the [0, 1] interval using a non-linear transformation function through the following steps: For each type of data, for a set of data X = {x1, x2, …, x n}, calculate its minimum value x min , maximum value x max and median x med . Define the non-linear transformation function f(x), and the calculation formula is: where β is an adjustable parameter, e is the natural logarithm, and the formula for the normalized data y i is:

6. The intelligent detection system based on a low-voltage line according to claim 3, characterized in that, The fault prediction and analysis module is specifically used to obtain comprehensive features by fusing the mapped voltage data, current data, temperature data, and humidity data through the following steps using a multi-modal data fusion feature extraction formula: For the mapped voltage data, current data, temperature data, and humidity data corresponding to each time point, let the voltage be V, the current be I, the temperature be T, and the humidity be H. The multi-modal data fusion feature extraction formula is as follows: where x1 = V, x2 = I, x3 = T, x4 = H, ω i is the weight coefficient corresponding to the data type, satisfying is the mean value of the i-th type of data, is the standard deviation of the i-th type of data, is the exponential parameter of the i-th type of data.

7. An intelligent detection system based on a low-voltage line according to claim 1, characterized in that, The fault prediction and analysis module is specifically used to predict the probability of fault occurrence of low-voltage lines using a chaotic hybrid model through the following steps: The calculation formula is as follows: where P fault is the probability of fault occurrence, m represents the number of multi-modal fusion feature values, M j-τ is the multi-modal fusion feature value at time point j after a time delay τ, α j is the feature weight corresponding to time point j, β1 is the chaos term coefficient, Chaos(j) is the chaos sequence generated based on the Logistic map, and after transformation, the chaos sequence value related to the multi-modal fusion feature value at time point j is obtained.

8. An intelligent detection system based on a low-voltage line according to claim 1, characterized in that, The fault prediction and analysis module is further configured to determine a preset threshold θ through the following steps old : Collect historical multi-modal data of low-voltage lines under different working conditions, and calculate the historical fault occurrence probability P′ under each working condition through a chaotic mixing model fault , calculate the mean value of the probability values and the standard deviation σ P , the formula is: where n is the number of working conditions, is the historical fault occurrence probability under the i-th working condition; Calculate the preset threshold as θ old , and the calculation formula is as follows: where k0 is an empirical coefficient.

9. An intelligent detection system based on a low-voltage line according to claim 1, characterized in that The threshold dynamic adjustment module is further configured to implement dynamic adjustment of a preset threshold based on real-time collected multi-modal data and historical multi-modal data by using an adaptive threshold dynamic adjustment formula through the following steps. The calculation formula is: where θ new is the adjusted preset threshold, θ old is the preset threshold before adjustment, γ is an adjustment coefficient that controls the amplitude of threshold adjustment, ΔM is the change amount of each pair of adjacent multi-modal fusion feature values in time among multiple multi-modal fusion feature values M within a preset time period, is the mean value of the change amounts of each pair of adjacent multi-modal fusion feature values in time among multiple multi-modal fusion feature values M within a preset time period, and δ is an exponential parameter dynamically adjusted according to environmental stability factors.

10. An intelligent detection system based on a low-voltage line according to claim 3, characterized in that, The state feedback warning module is specifically used to present key events through indicator lights, including overvoltage events, undervoltage events, overcurrent events corresponding to each of the A, B, and C phases in the low-voltage line, and abnormal line vibration events. The indicator lights set therein include: warning indicator lights for overvoltage, undervoltage, and overcurrent corresponding to each of the A, B, and C phases in the low-voltage line, a warning indicator light for abnormal line vibration events, and a status indicator light for the fault status of the low-voltage line.

Images