Bus intelligent monitoring method and device

Abstract

The application provides a bus intelligent monitoring method and device, and relates to the technical field of bus monitoring.The method comprises the following steps: obtaining a specific frequency range excitation signal; collecting signals of a bus duct system and performing frequency domain transformation to obtain frequency domain characteristics of each monitoring point position; obtaining multi-point frequency domain characteristic data according to the frequency domain characteristics of each monitoring point position; comparing the multi-point frequency domain characteristic data with data in a preset baseline database, and calculating frequency domain characteristic deviations of each monitoring point at different frequencies; analyzing the distribution mode of the frequency domain characteristic deviations of each monitoring point in the vertical direction according to the frequency domain characteristic deviations of each monitoring point at different frequencies, and determining the fault position and the fault type.The method of the application overcomes the dependence of traditional methods on accurate physical models and high-precision synchronous sampling through frequency domain analysis and pattern recognition, and realizes rapid positioning of different fault types and fault positions of a vertical bus duct system of a high-rise building.

Description

Technical Field

[0001] This invention relates to the field of busbar monitoring technology, and more specifically, to a method and apparatus for intelligent busbar monitoring. Background Technology

[0002] As a vital component of modern cities, the reliability of the internal power supply system is crucial for high-rise buildings. In these buildings, the vertical busbar system is the primary backbone of power transmission, responsible for delivering electricity from the ground floor or transformer floor upwards to each floor. This system typically consists of dozens or even hundreds of standard-length busbar units, connected layer by layer along the building's vertical height via specially designed connectors, forming a long-distance power channel spanning multiple floors. On each floor, the busbar system supplies power to the distribution box or electrical equipment on that floor through branch interfaces. Therefore, the vertical busbar system is characterized by its enormous length, wide distribution along the building's height, and numerous connection points and branch interfaces.

[0003] Vertical busbar systems are typically installed in vertical shafts or electrical shafts within buildings. These shafts are usually narrow, relatively enclosed spaces with limited ventilation and heat dissipation. Under high loads or high ambient temperatures, heat can easily accumulate within the shafts, potentially causing the busbar temperature to rise. Traditional busbar system maintenance relies primarily on periodic manual inspections. However, due to the limitations of the shaft environment, manual inspections are inefficient and struggle to comprehensively and promptly identify potential problems, especially given the large number of connectors and branch interfaces distributed across dozens of floors. These connectors and branch interfaces are weak points in the electrical connections of the busbar system. Due to installation processes, material aging, thermal expansion and contraction, or vibration, the contact resistance at the connections may gradually increase over time, leading to localized heating and potential overheating faults. Such hidden faults are difficult to detect visually or through conventional testing methods in their early stages. Once they develop into serious faults, such as phase-to-phase short circuits or grounding faults, they are often accompanied by high-temperature arcs, posing a very high fire risk and a serious threat to the public safety of high-rise buildings.

[0004] Furthermore, high-rise buildings contain a wide variety and large number of electrical devices, including lighting, air conditioning, elevators, and information equipment. The frequent changes in the operating status of these devices cause dynamic fluctuations in the overall load of the busbar system. These load variations not only affect the temperature distribution of the busbar but can also generate harmonic currents and transient voltage disturbances. This interferes with traditional fault signal detection methods based on electrical parameter thresholds or simple waveform analysis, reducing the sensitivity and accuracy of detection, especially when detecting early-stage, weak, and concealed faults.

[0005] When a busbar system malfunctions, whether due to overheating, insulation degradation, or a short circuit, it is crucial to quickly and accurately locate the floor and specific busbar section where the fault occurs. This is essential for guiding emergency response (such as isolating the faulty section, activating backup power, and organizing personnel evacuation) and subsequent maintenance and repairs. However, due to the enormous vertical length and complex structure of the busbar system, as well as the limitations imposed by the narrow shaft space on sensor installation location, quantity, reliability, and signal transmission anti-interference capabilities, achieving rapid and accurate fault location presents numerous challenges. Signals propagating through long-distance vertical busbar systems are affected by multiple reflections, attenuation, refraction, and mode conversions, complicating the correlation between fault signal characteristics and fault location. Different types of faults (such as poor contact and insulation breakdown) produce different signal characteristics, and early, concealed faults often have very weak signals, making them difficult to distinguish from environmental noise and normal electrical interference. Traditional fault location methods, such as the traveling wave method, often require accurate physical model parameters of the busbar trunking and high-precision synchronous sampling. However, the actual parameters of the busbar trunking system in high-rise buildings are difficult to obtain accurately and may change over time. Large-scale, multi-floor high-precision synchronous sampling systems are costly to deploy and susceptible to interference.

[0006] Therefore, in the specific scenario of high-rise buildings, how to overcome the above-mentioned unfavorable factors and achieve rapid and accurate fault location of the busbar system is a technical problem that urgently needs to be solved. Summary of the Invention

[0007] The purpose of this invention is to provide a method and device for intelligent monitoring of busbars. By using frequency domain analysis and pattern recognition, it overcomes the dependence of traditional methods on accurate physical models and high-precision synchronous sampling, and realizes rapid location of different fault types and fault locations in the vertical busbar system of high-rise buildings.

[0008] In a first aspect, the present invention provides a busbar intelligent monitoring method, comprising the following steps: Obtain the excitation signal within a specific frequency range injected into the main incoming line end of the busbar system; The voltage or current response signals of the busbar system are collected by multiple monitoring points arranged along the vertical direction. The collected voltage and current response signals are transformed in the frequency domain to obtain the frequency domain characteristics of each monitoring point. Based on the frequency domain characteristics of each monitoring point, the input impedance of each monitoring point, the transmission impedance between different monitoring points, or the voltage-current cross ratio between different monitoring points are calculated to obtain multi-point frequency domain characteristic data. The multi-point frequency domain feature data is compared with the data in the preset baseline database to calculate the frequency domain feature deviation of each monitoring point at different frequencies; the baseline database is established based on multiple sets of multi-point frequency domain feature data collected during the normal operation of the bus trunking system. Based on the frequency domain characteristic deviation of each monitoring point at different frequencies, the distribution pattern of the frequency domain characteristic deviation of each monitoring point in the vertical direction is analyzed, and the fault location and fault type of the bus trunking system are determined according to the distribution pattern.

[0009] The intelligent busbar monitoring method provided by this invention injects a specific frequency domain excitation signal into the vertical busbar system, monitors the frequency domain response of the system at multiple points along the route, analyzes the deviation pattern between the system and the normal baseline, and combines the busbar structure topology to achieve rapid location of the fault area.

[0010] Furthermore, based on the frequency domain characteristic deviation of each monitoring point at different frequencies, the distribution pattern of the frequency domain characteristic deviation in the vertical direction of each monitoring point is analyzed, and based on the distribution pattern, the steps to determine the fault location and fault type of the bus trunking system include: For the vertical busbar system in the shaft of a high-rise building, a vertical coordinate system containing floor information and monitoring point location information is established. The frequency domain characteristic deviation of each monitoring point at different frequencies is mapped to the vertical coordinate system to obtain the discrete distribution of the frequency domain characteristic deviation in the vertical direction; Determine whether the difference between the vertical coordinates of any two adjacent discrete points in the discrete distribution is greater than a preset threshold. If the difference is greater than the preset threshold, then an interpolation algorithm is used to calculate the virtual frequency domain feature deviation value based on the frequency domain feature deviation value of the adjacent monitoring points and map it to the vertical coordinate system to obtain a new discrete distribution. A curve fitting algorithm is used to fit the new discrete distribution to obtain a continuous distribution curve of the frequency domain characteristic deviation along the vertical direction; Extract the characteristic parameters of the continuous distribution curve; the characteristic parameters include the location of extreme points, the rate of change of slope, and the curvature. Based on the aforementioned characteristic parameters, the fault location and fault type are determined.

[0011] Further, determining whether the difference between the vertical coordinates of any two adjacent discrete points in the discrete distribution is greater than a preset threshold; if the difference is greater than the preset threshold, then using an interpolation algorithm to calculate a virtual frequency domain feature deviation value based on the frequency domain feature deviation value of the adjacent monitoring points and mapping it to the vertical coordinate system to obtain a new discrete distribution includes the following steps: S1. Obtain the structural information of the busbar system; the structural information includes the location of the busbar branch interfaces; S2. Map the location of the busbar branch interface to the vertical coordinate system; S3. Determine whether the location of the busbar branch interface falls between two adjacent discrete points. If so, determine whether the difference between the vertical coordinates of the two adjacent discrete points is greater than the preset threshold. If so, execute step S4; otherwise, end the interpolation operation for the two adjacent discrete points. S4. Using a cubic spline interpolation algorithm, based on the frequency domain feature deviation value of the two adjacent discrete points, calculate the virtual frequency domain feature deviation value and map it to the vertical coordinate system to obtain the new discrete distribution.

[0012] Furthermore, the specific steps in step S4 include: Obtain the branch structure type information of the busbar branch interface, the branch structure type information including rated current and connection method; Based on the branch structure type information, a preset impedance correction coefficient table is queried to determine the impedance correction coefficient corresponding to the branch structure type information; Using a cubic spline interpolation algorithm, the frequency domain characteristic deviation value of two adjacent discrete points is corrected according to the impedance correction coefficient, and the corrected frequency domain characteristic deviation value is obtained. Based on the corrected frequency domain feature deviation value, the virtual frequency domain feature deviation value is calculated and mapped to the vertical coordinate system to obtain the new discrete distribution.

[0013] Furthermore, the steps of determining the fault location and fault type based on the aforementioned characteristic parameters include: Acquire fault feature data; the fault feature data includes multiple fault types and the frequency domain feature deviation distribution pattern corresponding to each fault type; The location of the fault is determined based on the location of the extreme point; Based on the slope change rate and curvature, the similarity score of the frequency domain feature deviation distribution pattern corresponding to each fault type is calculated using the fault feature data. The fault type is determined based on the similarity score.

[0014] Secondly, the present invention provides a busbar intelligent monitoring device, comprising: The first acquisition module is used to acquire the excitation signal of a specific frequency range injected into the main incoming end of the bus trunking system; The acquisition module is used to acquire voltage or current response signals of the busbar system through multiple monitoring points arranged in the vertical direction. The second acquisition module is used to perform frequency domain transformation on the acquired voltage response signal and current response signal to obtain the frequency domain characteristics of each monitoring point location; The first calculation module is used to calculate the input impedance of each monitoring point, the transmission impedance between different monitoring points, or the voltage-current cross ratio between different monitoring points based on the frequency domain characteristics of each monitoring point location, so as to obtain multi-point frequency domain characteristic data. The second calculation module is used to compare the multi-point frequency domain feature data with the data in the preset baseline database to calculate the frequency domain feature deviation of each monitoring point at different frequencies; the baseline database is established based on multiple sets of multi-point frequency domain feature data collected during the normal operation of the bus trunking system; The analysis module is used to analyze the distribution pattern of the frequency domain characteristic deviation of each monitoring point in the vertical direction based on the frequency domain characteristic deviation of each monitoring point at different frequencies, and to determine the fault location and fault type of the bus trunking system based on the distribution pattern.

[0015] Furthermore, the analysis module performs the following steps when analyzing the distribution pattern of the frequency domain characteristic deviation of each monitoring point in the vertical direction based on the frequency domain characteristic deviation at different frequencies, and determining the fault location and fault type of the bus trunking system based on the distribution pattern: For the vertical busbar system in the shaft of a high-rise building, a vertical coordinate system containing floor information and monitoring point location information is established. The frequency domain characteristic deviation of each monitoring point at different frequencies is mapped to the vertical coordinate system to obtain the discrete distribution of the frequency domain characteristic deviation in the vertical direction; Determine whether the difference between the vertical coordinates of any two adjacent discrete points in the discrete distribution is greater than a preset threshold. If the difference is greater than the preset threshold, then an interpolation algorithm is used to calculate the virtual frequency domain feature deviation value based on the frequency domain feature deviation value of the adjacent monitoring points and map it to the vertical coordinate system to obtain a new discrete distribution. A curve fitting algorithm is used to fit the new discrete distribution to obtain a continuous distribution curve of the frequency domain characteristic deviation along the vertical direction; Extract the characteristic parameters of the continuous distribution curve; the characteristic parameters include the location of extreme points, the rate of change of slope, and the curvature. Based on the aforementioned characteristic parameters, the fault location and fault type are determined.

[0016] Furthermore, the analysis module determines whether the difference between the vertical coordinates of any two adjacent discrete points in the discrete distribution is greater than a preset threshold. If the difference is greater than the preset threshold, an interpolation algorithm is used to calculate a virtual frequency domain feature deviation value based on the frequency domain feature deviation value of the adjacent monitoring points and map it to the vertical coordinate system to obtain a new discrete distribution. S1. Obtain the structural information of the busbar system; the structural information includes the location of the busbar branch interfaces; S2. Map the location of the busbar branch interface to the vertical coordinate system; S3. Determine whether the location of the busbar branch interface falls between two adjacent discrete points. If so, determine whether the difference between the vertical coordinates of the two adjacent discrete points is greater than the preset threshold. If so, execute step S4; otherwise, end the interpolation operation for the two adjacent discrete points. S4. Using a cubic spline interpolation algorithm, based on the frequency domain feature deviation value of the two adjacent discrete points, calculate the virtual frequency domain feature deviation value and map it to the vertical coordinate system to obtain the new discrete distribution.

[0017] Furthermore, the analysis module performs the following steps when using a cubic spline interpolation algorithm to calculate a virtual frequency domain feature deviation value based on the frequency domain feature deviation value between two adjacent discrete points and mapping it to the vertical coordinate system to obtain the new discrete distribution: Obtain the branch structure type information of the busbar branch interface, the branch structure type information including rated current and connection method; Based on the branch structure type information, a preset impedance correction coefficient table is queried to determine the impedance correction coefficient corresponding to the branch structure type information; Using a cubic spline interpolation algorithm, the frequency domain characteristic deviation value of two adjacent discrete points is corrected according to the impedance correction coefficient, and the corrected frequency domain characteristic deviation value is obtained. Based on the corrected frequency domain feature deviation value, the virtual frequency domain feature deviation value is calculated and mapped to the vertical coordinate system to obtain the new discrete distribution.

[0018] Furthermore, the analysis module performs the following when determining the fault location and fault type based on the aforementioned characteristic parameters: Acquire fault feature data; the fault feature data includes multiple fault types and the frequency domain feature deviation distribution pattern corresponding to each fault type; The location of the fault is determined based on the location of the extreme point; Based on the slope change rate and curvature, the similarity score of the frequency domain feature deviation distribution pattern corresponding to each fault type is calculated using the fault feature data. The fault type is determined based on the similarity score.

[0019] As can be seen from the above, the intelligent busbar monitoring method provided by this invention does not rely on a precise physical parameter model of the busbar system. Instead, it utilizes the differences in impedance spectrum or frequency domain transfer function characteristics of the busbar system under normal and fault conditions within a specific frequency range. By injecting excitation signals at specific locations in the busbar system, response signals are collected at multiple monitoring points along the vertical direction. The frequency domain characteristics of these signals are analyzed to establish a frequency domain characteristic baseline under normal conditions. When a system fault occurs, the fault point (e.g., increased contact resistance due to overheating, local capacitance / resistance changes due to insulation degradation) will change the local or even overall frequency domain characteristics of the busbar, causing the frequency domain characteristics of the response signals collected by the monitoring points along the route to deviate from the baseline. By analyzing the distribution pattern and frequency characteristics of these deviations at different monitoring points, combined with the vertical topology of the busbar system (segment connections, branch locations), the specific floor or busbar segment area where the fault occurs can be quickly located. This invention utilizes the unique influence of the multi-segment, multi-branch structure of high-rise vertical busbars on the propagation of specific frequency signals, avoiding dependence on high-precision synchronous sampling and precise physical models.

[0020] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing embodiments of the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings. Attached Figure Description

[0021] Figure 1 This is a flowchart of a bus intelligent monitoring method provided in an embodiment of the present invention.

[0022] Figure 2 This is a schematic diagram of a busbar intelligent monitoring device provided in an embodiment of the present invention.

[0023] Label Explanation: 100. First acquisition module; 200. Acquisition module; 300. Second acquisition module; 400. First calculation module; 500. Second calculation module; 600. Analysis module. Detailed Implementation

[0024] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0025] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this invention, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0026] Reference Appendix Figure 1 This invention provides a method for intelligent monitoring of busbars, comprising the following steps: Obtain the excitation signal within a specific frequency range injected into the main incoming line end of the busbar system; The voltage or current response signals of the busbar system are collected by multiple monitoring points arranged along the vertical direction. The collected voltage and current response signals are transformed in the frequency domain to obtain the frequency domain characteristics of each monitoring point. Based on the frequency domain characteristics of each monitoring point, the input impedance of each monitoring point, the transmission impedance between different monitoring points, or the voltage-current cross ratio between different monitoring points are calculated to obtain multi-point frequency domain characteristic data. The multi-point frequency domain characteristic data is compared with the data in the preset baseline database to calculate the frequency domain characteristic deviation of each monitoring point at different frequencies; the baseline database is established based on multiple sets of multi-point frequency domain characteristic data collected during the normal operation of the bus trunking system. Based on the frequency domain characteristic deviation of each monitoring point at different frequencies, the distribution pattern of the frequency domain characteristic deviation of each monitoring point in the vertical direction is analyzed, and the fault location and fault type of the bus trunking system are determined according to the distribution pattern.

[0027] The working principle of this invention is based on the differences in the frequency domain response characteristics of a busbar system under different operating conditions. By injecting an external excitation signal, the voltage and current responses at multiple locations along the vertical direction of the busbar system are measured and transformed into the frequency domain. A baseline of these frequency domain responses (such as impedance and transfer function) is established under normal conditions. When a fault occurs in the system, the fault point changes electrical parameters such as local impedance or dielectric constant, causing a change in the overall frequency domain response of the system. By comparing the deviations of the currently measured multi-point frequency domain responses from the baseline and analyzing the distribution patterns of these deviations in the vertical direction, the location of the fault causing the specific deviation pattern can be inferred using known structural information of the busbar (connection sections, branch points). For example, the fault point may become a reflection or attenuation center for a specific frequency signal, causing different changes in the frequency domain responses of its upstream and downstream monitoring points. Fault location is achieved by identifying this spatially distributed frequency domain change pattern.

[0028] This solution locates faults by analyzing impedance or transmission characteristic changes in the busbar system at different frequencies, overcoming existing technical limitations. First, the frequency domain analysis and pattern matching processing are fast, meeting the requirement for rapid fault location. Second, by injecting excitation signals at specific locations and monitoring multiple points, the solution eliminates the need for frequent manual inspections in narrow vertical shafts. Faults (especially overheating and insulation degradation) cause changes in local impedance or dielectric characteristics, which in turn affect the overall frequency domain response of the system at specific frequencies. The solution is sensitive to these changes, aiding in the detection of early, hidden faults. By analyzing the relative frequency domain response changes at multiple monitoring points, the solution effectively addresses the effects of signal attenuation, reflection, refraction, and environmental electromagnetic interference during long-distance vertical propagation (especially when combined with the adaptive filtering algorithm described below), distinguishing fault signals from background noise and interference caused by load changes. The solution establishes a baseline and performs pattern recognition based on actual measured frequency domain responses, avoiding reliance on precise physical parameter models and solving the problems of complex vertical busbar structures and the difficulty and variability of accurately obtained parameters. Furthermore, the solution design reduces the stringent requirements for high-precision synchronous sampling, lowering system deployment costs and complexity. By analyzing the response in different frequency ranges, the solution can differentiate between different types of faults (such as overheating and insulation degradation), further assisting in fault diagnosis.

[0029] The process involves acquiring a specific frequency range excitation signal injected into the main input terminal of the busbar system. This excitation signal can be generated by a signal generator and coupled to the main input terminal. The frequency range of the excitation signal is selected to cover the characteristic frequencies of the busbar system to fully stimulate its electrical response. Voltage or current response signals of the busbar system are acquired through multiple monitoring points arranged vertically. This can be achieved by installing voltage or current sensors at different floors or locations along the vertical height of the busbar system. These sensors acquire the system's response to the injected excitation signal. The acquired voltage and current response signals are then subjected to frequency domain transformation to obtain the frequency domain characteristics of each monitoring point. A Fast Fourier Transform (FFT) algorithm is typically used to convert the time-domain acquired signal into a frequency-domain representation, obtaining the amplitude and phase information of the signal at different frequencies. Based on the frequency domain characteristics of each monitoring point, the input impedance at each monitoring point, the transmission impedance between different monitoring points, or the voltage-current cross ratio (transmission admittance) between different monitoring points are calculated to obtain multi-point frequency domain characteristic data. For example, the input impedance can be calculated from the voltage and current frequency domain components of the same monitoring point, and the transmission impedance can be calculated from the voltage or current frequency domain components between different monitoring points. These calculation results constitute a multi-point dataset reflecting the overall electrical characteristics of the system. The multi-point frequency domain characteristic data is compared with data in a preset baseline database to calculate the frequency domain characteristic deviation of each monitoring point at different frequencies. The baseline database stores multi-point frequency domain characteristic data of the system in a healthy state. The comparison can be performed using difference or ratio calculations to quantify the difference between the current state and the normal state. Based on the frequency domain characteristic deviation of each monitoring point at different frequencies, the distribution pattern of the frequency domain characteristic deviation in the vertical direction of each monitoring point is analyzed. Based on the distribution pattern, the fault location and fault type of the busbar system are determined. For example, by observing the trend of the deviation value along the vertical direction and the peak position, it is matched with the deviation distribution pattern corresponding to the known fault type, thereby locating the fault and identifying its type.

[0030] Specifically, this method actively detects the electrical characteristics of a busbar system by injecting excitation signals within a specific frequency range into the system. In the vertical shafts of high-rise buildings, the complex structure of vertical busbar systems makes it difficult to effectively detect hidden faults using traditional methods. By collecting voltage or current response signals at multiple monitoring points arranged vertically, the system's response to the excitation signal at different vertical locations is obtained. Frequency domain transformation is performed on these response signals to obtain the frequency domain characteristics of each monitoring point, reflecting the system's impedance and transmission characteristics at different frequencies. Furthermore, parameters such as input impedance, transmission impedance, or voltage-current cross ratio are calculated to transform the frequency domain characteristics into more intuitive electrical parameters. These multi-point frequency domain characteristic data are compared with baseline data established under normal system operation to calculate the frequency domain characteristic deviation of each monitoring point at different frequencies. This deviation can sensitively reflect minute changes within the system, such as an increase in contact resistance at connection points or a localized decrease in insulation performance; these changes are typical manifestations of early hidden faults. By analyzing the distribution pattern of these frequency domain characteristic deviations at each monitoring point in the vertical direction, spatial characteristics related to the fault can be identified. Different types of faults and faults at different locations will produce specific deviation distribution patterns in the vertical direction. For example, localized overheating may cause a significant increase in impedance deviation at nearby monitoring points, while insulation degradation may affect the transmission characteristics between multiple monitoring points. Based on the identified distribution pattern, it is matched with a preset fault mode library to ultimately determine the specific location and type of the fault. Therefore, this method overcomes the challenges of fault detection and location in high-rise building busbar systems, achieving rapid and accurate detection and location of concealed faults, and improving the operational reliability and safety of the system.

[0031] In some specific implementations, a signal generator can be connected to the main incoming line terminal of the high-rise building's busbar system located on the basement floor, injecting a swept-frequency sinusoidal excitation signal with a frequency range of 1 kHz to 50 kHz. Current sensors are installed as monitoring points every five floors along the vertical direction of the building (e.g., the 5th, 10th, and 15th floors). The current response signal of each monitoring point under the excitation signal is collected. A Fast Fourier Transform is performed on the collected current response signal to obtain the current frequency domain component of each monitoring point in the 1 kHz to 50 kHz frequency range. The transmission admittance (the ratio of the current frequency domain component to the injected voltage frequency domain component) of each monitoring point relative to the main incoming line terminal is calculated. The currently calculated transmission admittance data for each monitoring point is compared with the baseline transmission admittance data collected and stored during normal system operation to calculate the transmission admittance deviation of each monitoring point at different frequencies. The distribution of these transmission admittance deviation values ​​along the vertical direction (i.e., floor height) is analyzed. For example, if the transmission admittance deviation shows a significant local peak in a specific frequency range at the monitoring point on the 15th floor, while the deviation is relatively small at the adjacent monitoring points on the 10th and 20th floors, it can be determined from this distribution pattern that the fault may be located in the bus section near the 15th floor. Furthermore, based on the specific frequency characteristics and amplitude of the deviation, it can be determined that the fault type may be an increase in local impedance caused by poor contact.

[0032] In some embodiments, the steps of analyzing the distribution pattern of the frequency domain characteristic deviation of each monitoring point in the vertical direction according to the frequency domain characteristic deviation of each monitoring point at different frequencies, and determining the fault location and fault type of the bus trunking system based on the distribution pattern include: For the vertical busbar system in the shaft of a high-rise building, a vertical coordinate system containing floor information and monitoring point location information is established. The frequency domain characteristic deviations of each monitoring point at different frequencies are mapped to the vertical coordinate system to obtain the discrete distribution of the frequency domain characteristic deviations in the vertical direction. The algorithm determines whether the difference between the vertical coordinates of any two adjacent discrete points in the discrete distribution is greater than a preset threshold. If the difference is greater than the preset threshold, an interpolation algorithm is used to calculate the virtual frequency domain characteristic deviation value based on the frequency domain characteristic deviation value of the adjacent monitoring points and map it to the vertical coordinate system to obtain a new discrete distribution. The preset threshold is determined based on the floor height of the busbar system and the monitoring accuracy requirements. A curve fitting algorithm is used to fit the new discrete distribution to obtain a continuous distribution curve of the frequency domain characteristic deviation along the vertical direction; Extract the characteristic parameters of the continuous distribution curve; the characteristic parameters include the location of extreme points, the rate of change of slope, and curvature. Based on the characteristic parameters, determine the fault location and fault type.

[0033] This embodiment aims to address the problem that discrete data collected from multiple monitoring points arranged vertically cannot fully reflect the continuous variation trend of frequency domain characteristic deviation along the vertical direction of the busbar system. First, a vertical coordinate system is established, assigning spatial location information to the frequency domain characteristic deviation values ​​of each monitoring point, forming a discrete distribution. Next, by checking the vertical distance between adjacent monitoring points, sparse data regions are identified. If the distance is greater than a preset threshold, it indicates that monitoring points are sparse in that region, potentially missing important deviation change information. In this case, an interpolation algorithm is used to calculate and add virtual deviation values ​​based on the known deviation values ​​of the monitoring points, increasing the data point density and obtaining a denser discrete distribution. The preset threshold ensures the necessity of the interpolation operation and matches the required monitoring accuracy. Then, a curve fitting algorithm is applied to the new discrete distribution, transforming the discrete data into a continuous function or curve expression, thereby comprehensively reflecting the variation trend of deviation along the entire vertical length and overcoming the limitations of relying solely on discrete point analysis. Characteristic parameters such as extreme point locations, slope change rate, and curvature are extracted from this continuous distribution curve. The extreme point locations directly indicate the locations of maximum or minimum deviation, and are usually related to the location of the fault. The rate of change of slope and curvature describe the shape characteristics of the curve, reflecting the influence patterns of different types of faults on the frequency domain characteristic deviation distribution. Finally, based on these extracted characteristic parameters, a comparative analysis is performed with preset fault characteristic data to determine the specific location and type of the fault. By converting discrete data into continuous curves and analyzing their characteristic parameters, this method can more accurately locate faults and identify fault types.

[0034] In some specific implementations, for a vertical busbar system in a 30-story high-rise building, monitoring points are arranged on the 1st, 15th, and 30th floors. The vertical coordinate system sets the 1st floor to 0 meters, the 15th floor to 45 meters, and the 30th floor to 90 meters (assuming each floor is 3 meters high). The collected frequency domain characteristic deviation values ​​are mapped to the 0-meter, 45-meter, and 90-meter positions. A preset threshold is set to 10 floors (30 meters). The distance between adjacent points is checked: the distance from 0 meters to 45 meters is 45 meters, which is greater than the 30-meter threshold; the distance from 45 meters to 90 meters is also 45 meters, which is greater than the 30-meter threshold. Since the distance between adjacent points is greater than the threshold, an interpolation algorithm is triggered. Using the interpolation algorithm, based on the deviation values ​​at the 0-meter and 45-meter positions, virtual deviation values ​​are calculated and added to positions between 0 meters and 45 meters, such as at the 15-meter and 30-meter positions. Similarly, based on the deviation values ​​at the 45-meter and 90-meter positions, virtual deviation values ​​are calculated and added to positions between 45 and 90 meters, such as at 60 and 75 meters. This yields a new discrete distribution containing both the original and virtual points. A curve fitting algorithm, such as multinomial fitting, is used to fit the new discrete distribution, resulting in a continuous distribution curve of the frequency domain characteristic deviation along the vertical direction. Feature parameters are extracted from this continuous curve; for example, a significant extreme point is found at approximately 55 meters, and the slope change rate and curvature near this extreme point exhibit a specific pattern. Based on the extreme point location of approximately 55 meters (corresponding to approximately the 19th layer), the fault location is determined to be near the 19th layer. Based on the slope change rate and curvature pattern, and by comparing with a fault feature database, the fault type is determined to be an overheating fault caused by poor contact. Interpolation increases the data density, and curve fitting yields a continuous distribution, improving the accuracy of fault location and making fault type determination more reliable.

[0035] In some embodiments, the step of determining whether the difference between the vertical coordinates of any two adjacent discrete points in the discrete distribution is greater than a preset threshold, and if the difference is greater than the preset threshold, then using an interpolation algorithm to calculate a virtual frequency domain feature deviation value based on the frequency domain feature deviation value of the adjacent monitoring points and mapping it to the vertical coordinate system to obtain a new discrete distribution includes: S1. Obtain busbar system structural information; structural information includes the location of busbar branch interfaces; S2. Map the location of the busbar branch interface to the vertical coordinate system; S3. Determine whether the location of the busbar branch interface falls between two adjacent discrete points. If so, determine whether the difference between the vertical coordinates of the two adjacent discrete points is greater than a preset threshold. If so, proceed to step S4; otherwise, end the interpolation operation for the two adjacent discrete points. S4. Using the cubic spline interpolation algorithm, based on the frequency domain feature deviation value of the two adjacent discrete points, the virtual frequency domain feature deviation value is calculated and mapped to the vertical coordinate system to obtain a new discrete distribution.

[0036] Obtaining structural information about the busbar system, including the locations of busbar branch interfaces, can be achieved by consulting the system's design drawings or conducting on-site surveys. This location information can be represented as the vertical distance relative to the system's starting point or the corresponding floor height. Mapping this location information to a vertical coordinate system converts these vertical distances or floor heights into corresponding numerical points. Determining whether a busbar branch interface location falls between adjacent discrete points is done by comparing the vertical coordinates of the branch interface with the vertical coordinates of two adjacent monitoring points (discrete points). If the coordinates of the branch interface are between the coordinates of two monitoring points, it is considered to fall between adjacent discrete points. Determining whether the difference between the vertical coordinates of adjacent discrete points exceeds a threshold involves comparing the vertical distance between adjacent monitoring points with a preset threshold. The threshold can be set based on the floor height of the busbar system, the density of monitoring point layout, and the required fault location accuracy. An interpolation algorithm is used to calculate a virtual frequency domain characteristic deviation value based on the frequency domain characteristic deviation values ​​of adjacent discrete points, generating new discrete points in the vertical coordinate system. The choice of interpolation algorithm can be determined based on actual needs, such as cubic spline interpolation.

[0037] Specifically, when analyzing the distribution pattern of frequency domain characteristic deviation along the vertical direction, this technical solution first establishes a vertical coordinate system and maps the frequency domain characteristic deviation values ​​collected and calculated through monitoring points onto this coordinate system, forming an initial discrete distribution. To improve the density and accuracy of the distribution, especially near structural change points, the solution acquires the structural information of the busbar system and identifies the specific vertical positions of the busbar branch interfaces. These branch interfaces are locations where the impedance characteristics of the busbar may change significantly. These branch interface positions are also mapped onto the vertical coordinate system. Subsequently, the solution traverses adjacent monitoring point pairs in the initial discrete distribution. For each pair of adjacent monitoring points, it first checks whether the difference between their vertical coordinates is greater than a preset threshold. If the difference is large, it indicates that the monitoring point density in that area is insufficient. Further, the solution determines whether there are busbar branch interfaces between these adjacent monitoring points. If a branch interface exists and the vertical distance between adjacent monitoring points is greater than a threshold, interpolation is considered necessary for this area to more accurately reflect the actual distribution (the impedance change at the branch interface of the busbar system is large, which may cause abrupt changes in the frequency domain characteristic deviation, potentially leading to discrepancies between the interpolation results and the actual situation, affecting the accuracy of subsequent curve fitting and fault diagnosis). In this case, the scheme uses an interpolation algorithm, such as cubic spline interpolation, to calculate virtual frequency domain characteristic deviation values ​​at the branch interface location and other possible locations between the branch interface and adjacent monitoring points using the frequency domain characteristic deviation values ​​of the two adjacent monitoring points. These virtual values ​​are added as new discrete points to the vertical coordinate system, resulting in a new discrete distribution containing more data points. By adding interpolation points near the branch interface location, this scheme ensures that the distribution of frequency domain characteristic deviations can be more accurately characterized at critical locations where the busbar system structure changes, thereby improving the accuracy of subsequent curve fitting and ultimately enhancing the accuracy of fault location and type determination based on distribution patterns. If there is no branch interface between adjacent monitoring points, or if a branch interface exists but the distance between adjacent monitoring points is less than a threshold, no interpolation operation is performed on this area.

[0038] In some specific embodiments, it is assumed that in a certain shaft of a high-rise building, a vertical busbar trunking system has a monitoring point each on the 5th floor and the 8th floor, and their vertical coordinates are Y5 and Y8 respectively, and the corresponding frequency-domain characteristic deviation values are D5 and D8 respectively. The preset threshold value of the vertical coordinate difference is the height of 3 floors. By referring to the structure drawing of the busbar trunking system, it is found that there is a busbar branch interface at the floor position of the 6th floor (vertical coordinate Y6). First, it is judged that the vertical coordinate difference |Y8 - Y5| between adjacent monitoring points (the 5th floor and the 8th floor) is equal to the height of 3 floors, which is equal to the preset threshold value. Then, it is judged whether the position of the branch interface (the 6th floor, Y6) falls between the monitoring points on the 5th floor and the 8th floor. Since Y5 < Y6 < Y8, the branch interface falls between the two. The condition that there is a branch interface between adjacent monitoring points and the vertical distance is not less than the threshold value is met. Thus, the cubic spline interpolation algorithm is used to calculate the virtual frequency-domain characteristic deviation value D6 at the branch interface position Y6 according to the frequency-domain characteristic deviation values D5 and D8 of monitoring point 5 and monitoring point 8. The virtual value D6 is added as a new discrete point (Y6, D6) to the vertical coordinate system. If necessary, more virtual points can also be calculated between Y5 and Y6 and between Y6 and Y8 according to the interpolation algorithm to further densify the distribution. By adding a data point at the branch interface position Y6, the new discrete distribution can more accurately reflect the influence of this structural change point on the frequency-domain characteristic distribution and provide a more refined data basis for subsequent curve fitting.

[0039] In some embodiments, the specific steps in step S4 include: Obtain the branch structure type information of the busbar branch interface, and the branch structure type information includes the rated current and the connection method; According to the branch structure type information, query the preset impedance correction coefficient table to determine the impedance correction coefficient corresponding to the branch structure type information; Use the cubic spline interpolation algorithm to correct the frequency-domain characteristic deviation values of the corresponding adjacent two discrete points according to the impedance correction coefficient to obtain the corrected frequency-domain characteristic deviation values; According to the corrected frequency-domain characteristic deviation values, calculate the virtual frequency-domain characteristic deviation value and map it to the vertical coordinate system to obtain a new discrete distribution.

[0040] The system retrieves branch structure type information for busbar trunking branch interfaces. This information can be retrieved from the busbar trunking system's design documents, installation records, or a pre-established equipment database. Branch structure type information includes the rated current rating and connection method of the branch interface, such as whether it is a bolted or plug-in branch box. A pre-defined impedance correction coefficient table is queried. This table is pre-established and stored in the system. Through experimental measurements, simulation calculations, or theoretical analysis, the influence of different branch structure types on the frequency domain characteristic deviation distribution of the busbar trunking system is determined, and this influence is quantified as impedance correction coefficients. This table uses the branch structure type information as an index to return the corresponding correction coefficients. A cubic spline interpolation algorithm is used to correct the frequency domain characteristic deviation values ​​of two adjacent discrete points based on the impedance correction coefficients. The correction method can be to adjust the input values ​​of the interpolation algorithm according to the correction coefficients, such as weighting or shifting the original frequency domain characteristic deviation values ​​of adjacent monitoring points, so that the corrected values ​​better reflect the local electrical characteristics at the branch interface. Based on the corrected frequency domain characteristic deviation value, a virtual frequency domain characteristic deviation value is calculated. Using a standard cubic spline interpolation mathematical model, with the vertical coordinates of adjacent monitoring points and the corrected frequency domain characteristic deviation value as input, the virtual frequency domain characteristic deviation value at the branch interface position or other vertical coordinates requiring interpolation is calculated. These virtual values, together with the deviation values ​​of the original monitoring points, constitute a new discrete distribution along the vertical direction.

[0041] Specifically, for vertical busbar systems within the shafts of high-rise buildings, after establishing a vertical coordinate system containing floor information and monitoring point location information, the frequency domain characteristic deviations of each monitoring point at different frequencies are mapped onto this coordinate system, resulting in a discrete distribution of the frequency domain characteristic deviations in the vertical direction. It is then determined whether the difference between the vertical coordinates of any two adjacent discrete points in this discrete distribution exceeds a preset threshold. If the difference exceeds the preset threshold, and there is a busbar branch interface between these adjacent discrete points, interpolation processing is required. To improve interpolation accuracy, this scheme first obtains the branch structure type information of the branch interface, such as its rated current and connection method. Then, based on this structure type information, a preset impedance correction coefficient table is consulted to obtain the impedance correction coefficient corresponding to the branch structure type. This correction coefficient reflects the influence of this type of branch interface on the frequency domain characteristics of the busbar system. Next, a cubic spline interpolation algorithm is used, and based on the obtained impedance correction coefficient, the frequency domain characteristic deviation values ​​of two adjacent discrete points are corrected. This correction operation ensures that the input data used for interpolation considers the local influence of the branch interface. Finally, based on the corrected frequency domain characteristic deviation values, a cubic spline interpolation algorithm is used to calculate virtual frequency domain characteristic deviation values, and these virtual values ​​are mapped to the vertical coordinate system. By inserting virtual data points that take into account branch characteristics at locations with branch interfaces and large monitoring point spacing, the resulting new discrete distribution can more accurately reflect the frequency domain characteristic deviation changes of the busbar system along the vertical direction, especially at critical locations such as branch interfaces. This provides a more accurate data foundation for subsequent continuous distribution curve fitting and improves the accuracy of fault location.

[0042] In some specific implementations, it is assumed that in the vertical coordinate system, monitoring point A is located at vertical coordinate Z_A, and its frequency domain characteristic deviation value is D_A; monitoring point B is located at vertical coordinate Z_B, and its frequency domain characteristic deviation value is D_B. Between Z_A and Z_B, at vertical coordinate Z_C, there exists a busbar branch interface. Z_B - Z_A is greater than a preset threshold. The branch structure type information of this branch interface is obtained, for example, the rated current is 630A, and the connection method is plug-in. A preset impedance correction coefficient table is consulted to find the impedance correction coefficient corresponding to "630A, plug-in," assuming the coefficient found is K = 1.15. A cubic spline interpolation algorithm is used, and D_A and D_B are corrected according to the correction coefficient K. For example, the corrected deviation values ​​D'_A = D_A * K, D'_B = D_B * K. Then, using cubic spline interpolation, with points (Z_A, D'_A) and (Z_B, D'_B) as input, the virtual frequency domain feature deviation value D_C at the vertical coordinate Z_C is calculated. The point (Z_C, D_C) is then mapped to the vertical coordinate system. This adds virtual data points that consider the influence of the branch interface to the original discrete distribution, forming a new discrete distribution where the data points near the branch interface location are closer to reality.

[0043] In some embodiments, the step of using a curve fitting algorithm to fit the new discrete distribution and obtain a continuous distribution curve of the frequency domain feature deviation along the vertical direction includes: Acquire electromagnetic environment monitoring data during the operation of the bus trunking system; the electromagnetic environment monitoring data includes electromagnetic interference frequency and intensity; Based on electromagnetic environment monitoring data, the new discrete distribution is decomposed using wavelet transform algorithm to obtain approximate components and detail components at multiple scales. Based on the frequency and intensity of electromagnetic interference, determine the detail components corresponding to the electromagnetic interference frequency, and calculate the energy value of the detail components; Determine whether the energy value of the detail component is greater than the preset energy threshold. If it is greater than the preset energy threshold, use an adaptive filtering algorithm to suppress the detail component and obtain the discrete distribution after suppressing the interference. A curve fitting algorithm is used to fit the discrete distribution after interference suppression, and the continuous distribution curve of frequency domain characteristic deviation along the vertical direction is obtained.

[0044] Electromagnetic environment monitoring data includes electromagnetic interference frequency and intensity. Wavelet transform algorithms are used to decompose the new discrete distribution into components of different scales. Detail components contain high-frequency elements of the signal. Electromagnetic interference frequency and intensity are used to identify detail components associated with the interference. The energy values ​​of the detail components are used to quantify the interference intensity. A preset energy threshold is used to determine whether the interference requires intervention. Adaptive filtering algorithms are used to suppress interference components in the detail components. The discrete distribution after interference suppression is used for subsequent curve fitting. Curve fitting algorithms are used to generate continuous distribution curves.

[0045] Specifically, after acquiring electromagnetic environment monitoring data during the operation of the busbar system, this data provides information on the frequency and intensity of electromagnetic interference (EMI). A wavelet transform algorithm is used to decompose the new discrete distribution of the frequency domain characteristic deviation, thereby obtaining approximate and detail components at multiple scales. Based on the EMI frequency and intensity in the electromagnetic environment monitoring data, the parts of the detail components obtained from the wavelet decomposition that correspond to these interference frequencies can be determined. The energy values ​​of these corresponding detail components are calculated to assess the significance of the interference. By comparing the calculated energy values ​​of the detail components with a preset energy threshold, it can be determined whether there is significant EMI requiring treatment. If the energy value of a detail component is greater than the preset energy threshold, it indicates the presence of significant interference. In this case, an adaptive filtering algorithm is used to suppress these affected detail components, thereby reducing or removing the interference components and obtaining a discrete distribution after interference suppression. Finally, a curve fitting algorithm is used to fit the discrete distribution data after interference suppression to obtain a continuous distribution curve of the frequency domain characteristic deviation along the vertical direction. Therefore, by identifying and suppressing EMI, the accuracy of the discrete distribution data is improved, which in turn enhances the accuracy of the continuous distribution curve fitting, providing a more reliable data foundation for subsequent fault diagnosis.

[0046] In some specific implementations, electromagnetic environment monitoring data can be acquired by deploying electromagnetic field sensors in or near the busbar shaft. For example, the sensors can periodically collect electromagnetic field intensity data within a specific frequency range. Wavelet transform can be performed using Daubechies wavelet or Symlets wavelet for multi-level decomposition. A preset energy threshold can be set based on historical data or experience, for example, as a percentage of the total energy of the detail components or a fixed value. Adaptive filtering algorithms can employ the Least Mean Square (LMS) algorithm or the Recursive Least Squares (RLS) algorithm, with filter parameters adjusted according to the characteristics of the interference signal. Curve fitting algorithms can employ methods such as polynomial fitting, spline fitting, or locally weighted regression. For example, after acquiring electromagnetic environment monitoring data, significant interference at frequencies of 50Hz and 150Hz is found. Wavelet decomposition is performed on the new discrete distribution to identify the detail components corresponding to the 50Hz and 150Hz frequency ranges. The energy values ​​of these detail components are calculated. If the energy values ​​exceed a preset threshold, an adaptive filter is constructed to filter these detail components. Then, the filtered detail components are reconstructed with the approximate components to obtain the discrete distribution after suppressing interference. Finally, the distribution is curve-fitted.

[0047] In some embodiments, the step of determining whether the energy value of the detail component is greater than a preset energy threshold, and if it is greater than the preset energy threshold, using an adaptive filtering algorithm to suppress the detail component to obtain a discrete distribution after interference suppression includes: Acquire electromagnetic environment monitoring data within a preset time window; the electromagnetic environment monitoring data includes electromagnetic interference frequency, intensity, and interference source type; Based on the frequency and intensity of electromagnetic interference, the detailed components corresponding to the electromagnetic interference frequency are determined, and the energy values ​​of the detailed components are calculated. At the same time, based on the type of interference source, a preset interference source characteristic library is queried to obtain the time-domain characteristics of the interference signal corresponding to the type of interference source. If the energy value of the detail component is greater than the preset energy threshold, an adaptive filter matching the time domain characteristics of the interference signal is constructed based on the time domain characteristics of the interference signal. The constructed adaptive filter is used to filter the detail component to obtain the discrete distribution after interference suppression.

[0048] Electromagnetic environment monitoring data is acquired through sensors deployed near the busbar system. These sensors monitor electromagnetic field strength within a specific frequency range and identify potential interference source types. Detail components are obtained by performing wavelet transform on the discrete distribution of frequency domain characteristic deviations in the vertical direction. Wavelet transform decomposes the signal into components of different scales, where detail components typically contain high-frequency information and may include interference signals. The interference source characteristic database is a database storing time-domain waveforms or patterns of typical interference signals generated by different types of interference sources (e.g., frequency converters, switching power supplies, wireless communication devices, etc.). The query process retrieves the corresponding time-domain characteristic data based on the detected interference source type. The adaptive filter construction process adjusts the filter's structure or parameters according to the acquired time-domain characteristics of the interference signal, enabling it to specifically suppress signal components with those time-domain characteristics.

[0049] Specifically, to address the issue of inaccurate interference suppression in adaptive filtering algorithms that rely solely on energy thresholds to determine interference signals, this solution further utilizes interference source type information when determining that the energy of detail components exceeds the threshold and requires suppression. First, electromagnetic environment monitoring data is continuously collected within a preset time window. This data includes not only the frequency and intensity of interference but also the identified interference source type. When detail components are obtained through wavelet transform and their energy is calculated, if the energy exceeds a preset threshold, the system searches a preset interference source characteristic library for time-domain characteristics of interference signals related to that type of interference source. For example, if the detected interference source is a frequency converter, the system queries the library for time-domain characteristics of typical harmonics or switching noise generated by frequency converters. Then, using the acquired interference signal time-domain characteristics, an adaptive filter matching these characteristics is constructed. This matching can involve adjusting the filter's order, coefficients, or structure to provide greater attenuation for signals with specific time-domain patterns. Finally, the constructed adaptive filter is applied to the detail components that need suppression for filtering. This allows for more precise separation and removal of electromagnetic interference components from the detailed components, resulting in purer discrete distribution data after interference suppression. This improves the accuracy of subsequent curve fitting, thereby enhancing the precision of fault location and type determination.

[0050] In some specific implementations, the preset time window is set to 1 second. Electromagnetic environment monitoring data is acquired through a spectrum analyzer and an interference source identification module. The spectrum analyzer monitors the frequency and intensity of the electromagnetic field, and the interference source identification module determines the type of interference source by analyzing signal characteristics or combining information from on-site equipment, for example, identifying the interference source as an elevator frequency converter. A three-level wavelet decomposition is performed on the discrete distribution to obtain multiple detail components. The energy of the detail component corresponding to the elevator frequency converter's operating frequency (e.g., a specific harmonic frequency) is calculated. A preset energy threshold is set to 0.1. If the calculated energy value of the detail component is 0.15, which is greater than the threshold, interference suppression is triggered. Based on the identified elevator frequency converter type, the interference source characteristic library is queried to obtain typical time-domain waveform characteristics generated by this type of frequency converter, such as the presence of specific frequency components and pulse characteristics. Based on these time-domain characteristics, an adaptive LMS (Least Mean Square) filter is constructed, and the filter's weight coefficients are adjusted to effectively track and cancel signals with these time-domain characteristics. This adaptive filter is applied to the detail components whose calculated energy exceeds the threshold for filtering processing. Thus, the inverter interference component in the detail component is effectively suppressed, resulting in a discrete distribution after interference suppression for subsequent curve fitting.

[0051] In some embodiments, the steps of determining the fault location and fault type based on characteristic parameters include: Acquire fault characteristic data; the fault characteristic data includes multiple fault types and the frequency domain characteristic deviation distribution pattern corresponding to each fault type; Determine the location of the fault based on the location of the extreme point; Based on the slope change rate and curvature, the similarity score of the frequency domain feature deviation distribution pattern corresponding to each fault type is calculated using fault feature data. The type of fault is determined based on the similarity score.

[0052] The fault feature data is a database storing different fault types and their corresponding frequency domain feature deviation distribution patterns in the vertical direction. This database provides a comparative reference for fault diagnosis. The fault location is determined based on the extreme points of the continuous distribution curve. Extreme points represent regions of drastic change in frequency domain feature deviation, which are associated with the physical location of the fault. By locating the extreme points, the vertical location of the fault is pinpointed. The fault type is determined based on the slope change rate and curvature of the continuous distribution curve. The slope change rate and curvature describe the local shape characteristics of the curve. Different fault types cause the frequency domain feature deviation curve to exhibit different trends and degrees of curvature. The slope change rate and curvature of the curve fitted from the actual collected data are compared with the distribution patterns corresponding to various fault types stored in the database to calculate a similarity score. The similarity score quantifies the degree of matching between the actual fault pattern and the known fault pattern. The fault type with the highest similarity score is the determined result.

[0053] Specifically, this technical solution addresses the problem of effectively associating extracted curve feature parameters with specific fault locations and types, thereby achieving accurate fault diagnosis. First, a fault feature database is acquired, storing different fault types and their corresponding frequency domain feature deviation distribution patterns in the vertical direction, providing known fault patterns as comparison references. Then, feature parameters extracted from the continuous distribution curve of the frequency domain feature deviation are used for fault diagnosis. Specifically, the fault location is determined based on the extreme points of the continuous distribution curve. Extreme points typically represent the regions with the most drastic changes in frequency domain feature deviation, and these regions are correlated with the physical location of the actual fault. Therefore, locating the extreme points allows pinpointing the vertical location of the fault. This is the technical means to solve the problem of fault location determination. Next, the slope change rate and curvature of the continuous distribution curve are used to determine the fault type. The slope change rate and curvature describe the local shape characteristics of the curve; different types of faults will cause the frequency domain feature deviation curve to exhibit different trends and degrees of curvature. By comparing the slope change rate and curvature of the fitted curve based on actual collected data with the distribution patterns corresponding to various fault types stored in the fault feature database, and calculating a similarity score, the degree of matching between the actual fault pattern and the known fault pattern can be quantified. The fault type with a high similarity score is the judgment result. By establishing a fault feature database, using extreme points to locate the fault location, and determining the fault type based on the slope change rate and curvature through similarity scoring, a specific fault diagnosis process is provided, solving the technical problem of how to transform abstract curve feature parameters into specific fault diagnosis results.

[0054] In some specific implementations, it is assumed that a continuous distribution curve of frequency domain characteristic deviation along the vertical direction is detected in the vertical busbar system of a high-rise building. This curve exhibits a significant peak at a vertical coordinate of 50 meters (corresponding to the 15th floor of the building), which is the extreme point. Based on the location of this extreme point, the fault location is determined to be at a vertical coordinate of 50 meters. Further, the slope change rate and curvature near this peak point are calculated. For example, the calculated slope change rate is denoted as X, and the curvature as Y. The values ​​X and Y are compared with data in a preset fault feature database. This database stores various fault types (e.g., poor contact, insulation degradation, phase-to-phase short circuit) and their corresponding slope change rate and curvature ranges or patterns. The similarity score between the actual curve feature (X, Y) and the corresponding pattern for each fault type in the database is calculated. For example, the similarity score with the "poor contact" type pattern is 0.95, with the "insulation degradation" type pattern is 0.60, and with the "phase-to-phase short circuit" type pattern is 0.30. Based on the similarity score, the fault type with the highest similarity was determined to be "poor contact". Therefore, the diagnosis is that a poor contact fault exists at a vertical coordinate of 50 meters.

[0055] Please refer to Figure 2 , Figure 2 This invention provides a busbar intelligent monitoring device in some embodiments, which is integrated into a back-end control device in the form of a computer program, comprising: The first acquisition module 100 is used to acquire the excitation signal of a specific frequency range injected into the main incoming end of the bus trunking system. The acquisition module 200 is used to acquire voltage response signals or current response signals of the busbar system through multiple monitoring points arranged in the vertical direction. The second acquisition module 300 is used to perform frequency domain transformation on the acquired voltage response signal and current response signal to obtain the frequency domain characteristics of each monitoring point location. The first calculation module 400 is used to calculate the input impedance of each monitoring point, the transmission impedance between different monitoring points, or the voltage-current cross ratio between different monitoring points based on the frequency domain characteristics of each monitoring point location, so as to obtain multi-point frequency domain characteristic data. The second calculation module 500 is used to compare the multi-point frequency domain feature data with the data in the preset baseline database and calculate the frequency domain feature deviation of each monitoring point at different frequencies; the baseline database is established based on multiple sets of multi-point frequency domain feature data collected during the normal operation of the bus trunking system; The analysis module 600 is used to analyze the distribution pattern of the frequency domain characteristic deviation of each monitoring point in the vertical direction based on the frequency domain characteristic deviation of each monitoring point at different frequencies, and to determine the fault location and fault type of the bus trunking system based on the distribution pattern.

[0056] In some embodiments, the analysis module 600 performs the following operations when analyzing the distribution pattern of the frequency domain characteristic deviation of each monitoring point in the vertical direction based on the frequency domain characteristic deviation of each monitoring point at different frequencies, and determining the fault location and fault type of the bus trunking system based on the distribution pattern: For the vertical busbar system in the shaft of a high-rise building, a vertical coordinate system containing floor information and monitoring point location information is established. The frequency domain characteristic deviations of each monitoring point at different frequencies are mapped to the vertical coordinate system to obtain the discrete distribution of the frequency domain characteristic deviations in the vertical direction. The algorithm determines whether the difference between the vertical coordinates of any two adjacent discrete points in the discrete distribution is greater than a preset threshold. If the difference is greater than the preset threshold, an interpolation algorithm is used to calculate the virtual frequency domain characteristic deviation value based on the frequency domain characteristic deviation value of the adjacent monitoring points and map it to the vertical coordinate system to obtain a new discrete distribution. The preset threshold is determined based on the floor height of the busbar system and the monitoring accuracy requirements. A curve fitting algorithm is used to fit the new discrete distribution to obtain a continuous distribution curve of the frequency domain characteristic deviation along the vertical direction; Extract the characteristic parameters of the continuous distribution curve; the characteristic parameters include the location of extreme points, the rate of change of slope, and curvature. Based on the characteristic parameters, determine the fault location and fault type.

[0057] In some embodiments, the analysis module 600 performs the following steps when determining whether the difference between the vertical coordinates of any two adjacent discrete points in the discrete distribution is greater than a preset threshold. If the difference is greater than the preset threshold, an interpolation algorithm is used to calculate a virtual frequency domain feature deviation value based on the frequency domain feature deviation value of the adjacent monitoring points and map it to the vertical coordinate system to obtain a new discrete distribution: S1. Obtain busbar system structural information; structural information includes the location of busbar branch interfaces; S2. Map the location of the busbar branch interface to the vertical coordinate system; S3. Determine whether the location of the busbar branch interface falls between two adjacent discrete points. If so, determine whether the difference between the vertical coordinates of the two adjacent discrete points is greater than a preset threshold. If so, proceed to step S4; otherwise, end the interpolation operation for the two adjacent discrete points. S4. Using the cubic spline interpolation algorithm, based on the frequency domain feature deviation value of the two adjacent discrete points, the virtual frequency domain feature deviation value is calculated and mapped to the vertical coordinate system to obtain a new discrete distribution.

[0058] In some embodiments, the analysis module 600 performs the following steps when using a cubic spline interpolation algorithm to calculate a virtual frequency domain feature deviation value based on the frequency domain feature deviation value of two adjacent discrete points and map it to the vertical coordinate system to obtain a new discrete distribution: Obtain the branch structure type information of the busbar branch interface, which includes the rated current and connection method; Based on the branch structure type information, query the preset impedance correction coefficient table to determine the impedance correction coefficient corresponding to the branch structure type information; The cubic spline interpolation algorithm is used to correct the frequency domain characteristic deviation value of two adjacent discrete points according to the impedance correction coefficient, so as to obtain the corrected frequency domain characteristic deviation value. Based on the corrected frequency domain characteristic deviation value, the virtual frequency domain characteristic deviation value is calculated and mapped to the vertical coordinate system to obtain a new discrete distribution.

[0059] In some embodiments, the analysis module 600 is executed when it is used to fit a new discrete distribution using a curve fitting algorithm to obtain a continuous distribution curve of the frequency domain feature deviation along the vertical direction: Acquire electromagnetic environment monitoring data during the operation of the bus trunking system; the electromagnetic environment monitoring data includes electromagnetic interference frequency and intensity; Based on electromagnetic environment monitoring data, the new discrete distribution is decomposed using wavelet transform algorithm to obtain approximate components and detail components at multiple scales. Based on the frequency and intensity of electromagnetic interference, determine the detail components corresponding to the electromagnetic interference frequency, and calculate the energy value of the detail components; Determine whether the energy value of the detail component is greater than the preset energy threshold. If it is greater than the preset energy threshold, use an adaptive filtering algorithm to suppress the detail component and obtain the discrete distribution after suppressing the interference. A curve fitting algorithm is used to fit the discrete distribution after interference suppression, and the continuous distribution curve of frequency domain characteristic deviation along the vertical direction is obtained.

[0060] In some embodiments, the analysis module 600 performs the following steps when determining whether the energy value of the detail component is greater than a preset energy threshold: if it is greater than the preset energy threshold, an adaptive filtering algorithm is used to suppress the detail component to obtain a discrete distribution after interference suppression. Acquire electromagnetic environment monitoring data within a preset time window; the electromagnetic environment monitoring data includes electromagnetic interference frequency, intensity, and interference source type; Based on the frequency and intensity of electromagnetic interference, the detailed components corresponding to the electromagnetic interference frequency are determined, and the energy values ​​of the detailed components are calculated. At the same time, based on the type of interference source, a preset interference source characteristic library is queried to obtain the time-domain characteristics of the interference signal corresponding to the type of interference source. If the energy value of the detail component is greater than the preset energy threshold, an adaptive filter matching the time domain characteristics of the interference signal is constructed based on the time domain characteristics of the interference signal. The constructed adaptive filter is used to filter the detail component to obtain the discrete distribution after interference suppression.

[0061] In some embodiments, the analysis module 600 performs the following when determining the fault location and fault type based on characteristic parameters: Acquire fault characteristic data; the fault characteristic data includes multiple fault types and the frequency domain characteristic deviation distribution pattern corresponding to each fault type; Determine the location of the fault based on the location of the extreme point; Based on the slope change rate and curvature, the similarity score of the frequency domain feature deviation distribution pattern corresponding to each fault type is calculated using fault feature data. The type of fault is determined based on the similarity score.

[0062] In this document, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, without necessarily requiring or implying any such actual relationship or order between these entities or operations.

[0063] The above description is merely an embodiment of the present invention and is not intended to limit the scope of protection of the present invention. For those skilled in the art, the present invention can have various modifications and variations. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for intelligent monitoring of busbars, characterized in that, Includes the following steps: Obtain the excitation signal within a specific frequency range injected into the main incoming line end of the busbar system; The voltage or current response signals of the busbar system are collected by multiple monitoring points arranged along the vertical direction. The collected voltage and current response signals are transformed in the frequency domain to obtain the frequency domain characteristics of each monitoring point. Based on the frequency domain characteristics of each monitoring point, the input impedance of each monitoring point, the transmission impedance between different monitoring points, or the voltage-current cross ratio between different monitoring points are calculated to obtain multi-point frequency domain characteristic data. The multi-point frequency domain feature data is compared with the data in the preset baseline database to calculate the frequency domain feature deviation of each monitoring point at different frequencies. The baseline database is established based on multiple sets of multi-point frequency domain feature data collected during the normal operation of the bus trunking system. Based on the frequency domain characteristic deviation of each monitoring point at different frequencies, the distribution pattern of the frequency domain characteristic deviation of each monitoring point in the vertical direction is analyzed, and the fault location and fault type of the bus trunking system are determined according to the distribution pattern.

2. The intelligent busbar monitoring method according to claim 1, characterized in that, The steps for analyzing the distribution pattern of frequency domain characteristic deviations at different frequencies of each monitoring point in the vertical direction, and determining the fault location and fault type of the busbar system based on the distribution pattern, include: For the vertical busbar system in the shaft of a high-rise building, a vertical coordinate system containing floor information and monitoring point location information is established. The frequency domain characteristic deviation of each monitoring point at different frequencies is mapped to the vertical coordinate system to obtain the discrete distribution of the frequency domain characteristic deviation in the vertical direction; Determine whether the difference between the vertical coordinates of any two adjacent discrete points in the discrete distribution is greater than a preset threshold. If the difference is greater than the preset threshold, then an interpolation algorithm is used to calculate the virtual frequency domain feature deviation value based on the frequency domain feature deviation value of the adjacent monitoring points and map it to the vertical coordinate system to obtain a new discrete distribution. A curve fitting algorithm is used to fit the new discrete distribution to obtain a continuous distribution curve of the frequency domain characteristic deviation along the vertical direction; Extract the characteristic parameters of the continuous distribution curve; the characteristic parameters include the location of extreme points, the rate of change of slope, and the curvature. Based on the aforementioned characteristic parameters, the fault location and fault type are determined.

3. The intelligent busbar monitoring method according to claim 2, characterized in that, The steps of determining whether the difference between the vertical coordinates of any two adjacent discrete points in the discrete distribution is greater than a preset threshold, and if the difference is greater than the preset threshold, then using an interpolation algorithm to calculate a virtual frequency domain feature deviation value based on the frequency domain feature deviation value of the adjacent monitoring points and mapping it to the vertical coordinate system to obtain a new discrete distribution include: S1. Obtain the structural information of the busbar system; the structural information includes the location of the busbar branch interfaces; S2. Map the location of the busbar branch interface to the vertical coordinate system; S3. Determine whether the location of the busbar branch interface falls between two adjacent discrete points. If so, determine whether the difference between the vertical coordinates of the two adjacent discrete points is greater than the preset threshold. If so, execute step S4; otherwise, end the interpolation operation for the two adjacent discrete points. S4. Using a cubic spline interpolation algorithm, based on the frequency domain feature deviation value of the two adjacent discrete points, calculate the virtual frequency domain feature deviation value and map it to the vertical coordinate system to obtain the new discrete distribution.

4. The intelligent busbar monitoring method according to claim 3, characterized in that, The specific steps in step S4 include: Obtain the branch structure type information of the busbar branch interface, the branch structure type information including rated current and connection method; Based on the branch structure type information, a preset impedance correction coefficient table is queried to determine the impedance correction coefficient corresponding to the branch structure type information; Using a cubic spline interpolation algorithm, the frequency domain characteristic deviation value of two adjacent discrete points is corrected according to the impedance correction coefficient, and the corrected frequency domain characteristic deviation value is obtained. Based on the corrected frequency domain feature deviation value, the virtual frequency domain feature deviation value is calculated and mapped to the vertical coordinate system to obtain the new discrete distribution.

5. The intelligent busbar monitoring method according to claim 2, characterized in that, The steps for determining the fault location and fault type based on the aforementioned characteristic parameters include: Acquire fault feature data; the fault feature data includes multiple fault types and the frequency domain feature deviation distribution pattern corresponding to each fault type; The location of the fault is determined based on the location of the extreme point; Based on the slope change rate and curvature, the similarity score of the frequency domain feature deviation distribution pattern corresponding to each fault type is calculated using the fault feature data. The fault type is determined based on the similarity score.

6. A busbar intelligent monitoring device, characterized in that, include: The first acquisition module is used to acquire the excitation signal of a specific frequency range injected into the main incoming end of the bus trunking system; The acquisition module is used to acquire voltage or current response signals of the busbar system through multiple monitoring points arranged in the vertical direction. The second acquisition module is used to perform frequency domain transformation on the acquired voltage response signal and current response signal to obtain the frequency domain characteristics of each monitoring point location; The first calculation module is used to calculate the input impedance of each monitoring point, the transmission impedance between different monitoring points, or the voltage-current cross ratio between different monitoring points based on the frequency domain characteristics of each monitoring point location, so as to obtain multi-point frequency domain characteristic data. The second calculation module is used to compare the multi-point frequency domain feature data with the data in the preset baseline database and calculate the frequency domain feature deviation of each monitoring point at different frequencies. The baseline database is established based on multiple sets of multi-point frequency domain feature data collected during the normal operation of the bus trunking system. The analysis module is used to analyze the distribution pattern of the frequency domain characteristic deviation of each monitoring point in the vertical direction based on the frequency domain characteristic deviation of each monitoring point at different frequencies, and to determine the fault location and fault type of the bus trunking system based on the distribution pattern.

7. The intelligent busbar monitoring device according to claim 6, characterized in that, The analysis module performs the following steps when analyzing the distribution pattern of frequency domain characteristic deviations of each monitoring point in the vertical direction based on the frequency domain characteristic deviations at different frequencies, and determining the fault location and fault type of the busbar system based on the distribution pattern: For the vertical busbar system in the shaft of a high-rise building, a vertical coordinate system containing floor information and monitoring point location information is established. The frequency domain characteristic deviation of each monitoring point at different frequencies is mapped to the vertical coordinate system to obtain the discrete distribution of the frequency domain characteristic deviation in the vertical direction; Determine whether the difference between the vertical coordinates of any two adjacent discrete points in the discrete distribution is greater than a preset threshold. If the difference is greater than the preset threshold, then an interpolation algorithm is used to calculate the virtual frequency domain feature deviation value based on the frequency domain feature deviation value of the adjacent monitoring points and map it to the vertical coordinate system to obtain a new discrete distribution. A curve fitting algorithm is used to fit the new discrete distribution to obtain a continuous distribution curve of the frequency domain characteristic deviation along the vertical direction; Extract the characteristic parameters of the continuous distribution curve; the characteristic parameters include the location of extreme points, the rate of change of slope, and the curvature. Based on the aforementioned characteristic parameters, the fault location and fault type are determined.

8. The intelligent busbar monitoring device according to claim 7, characterized in that, The analysis module determines whether the difference between the vertical coordinates of any two adjacent discrete points in the discrete distribution is greater than a preset threshold. If the difference is greater than the preset threshold, an interpolation algorithm is used to calculate a virtual frequency domain feature deviation value based on the frequency domain feature deviation value of the adjacent monitoring points and map it to the vertical coordinate system to obtain a new discrete distribution. S1. Obtain the structural information of the busbar system; the structural information includes the location of the busbar branch interfaces; S2. Map the location of the busbar branch interface to the vertical coordinate system; S3. Determine whether the location of the busbar branch interface falls between two adjacent discrete points. If so, determine whether the difference between the vertical coordinates of the two adjacent discrete points is greater than the preset threshold. If so, execute step S4; otherwise, end the interpolation operation for the two adjacent discrete points. S4. Using a cubic spline interpolation algorithm, based on the frequency domain feature deviation value of the two adjacent discrete points, calculate the virtual frequency domain feature deviation value and map it to the vertical coordinate system to obtain the new discrete distribution.

9. The intelligent busbar monitoring device according to claim 8, characterized in that, The analysis module executes the following steps when using a cubic spline interpolation algorithm to calculate a virtual frequency domain feature deviation value based on the frequency domain feature deviation value between two adjacent discrete points and map it to the vertical coordinate system to obtain the new discrete distribution: Obtain the branch structure type information of the busbar branch interface, the branch structure type information including rated current and connection method; Based on the branch structure type information, a preset impedance correction coefficient table is queried to determine the impedance correction coefficient corresponding to the branch structure type information; Using a cubic spline interpolation algorithm, the frequency domain characteristic deviation value of two adjacent discrete points is corrected according to the impedance correction coefficient, and the corrected frequency domain characteristic deviation value is obtained. Based on the corrected frequency domain feature deviation value, the virtual frequency domain feature deviation value is calculated and mapped to the vertical coordinate system to obtain the new discrete distribution.

10. The intelligent busbar monitoring device according to claim 7, characterized in that, The analysis module performs the following when determining the fault location and fault type based on the aforementioned characteristic parameters: Acquire fault feature data; the fault feature data includes multiple fault types and the frequency domain feature deviation distribution pattern corresponding to each fault type; The location of the fault is determined based on the location of the extreme point; Based on the slope change rate and curvature, the similarity score of the frequency domain feature deviation distribution pattern corresponding to each fault type is calculated using the fault feature data. The fault type is determined based on the similarity score.

Images