Low-voltage power supply and distribution system bus duct wiring fault prediction method and system

Abstract

The invention provides a low-voltage power supply and distribution system bus duct wiring fault prediction method and system, and relates to the technical field of cable wiring, and the method comprises the steps: constructing a bus duct topological structure; receiving a new equipment demand; according to the position of the newly added equipment, carrying out clamped node analysis on the bus duct topological structure, and positioning a plurality of alternative segments; constructing a plurality of segmented twin models, performing wiring fault prediction, and outputting a plurality of segmented safety coefficients; positioning a target bus duct segment in the plurality of alternative segments; and in the low-voltage power distribution system, taking the clamping alternative node of the target bus duct segment as a target clamping node, and carrying out bus duct clamping construction of the newly added plug-in box. According to the method and the device, the technical problem that the low-voltage power supply and distribution system is difficult to identify the risk in advance when facing the newly-added equipment, so that the fault or the equipment is damaged is solved, and the clamping position of the newly-added plug-in box is positioned by predicting the fault risk of different clamping positions, so that the operation and maintenance efficiency of the low-voltage power distribution system is improved.

Description

Technical Field

[0001] The present application relates to the field of cable wiring technology, and in particular to a method and system for predicting bus duct wiring faults in a low-voltage power supply and distribution system. Background Art

[0002] In low-voltage power supply and distribution systems, busbars serve as important power distribution equipment. Through the design and wiring of busbars, electrical energy is transmitted from distribution cabinets or substations to various load devices. Existing busbar wiring methods are usually relatively fixed and lack flexibility, making it difficult to meet the needs of new equipment, resulting in poor scalability and difficulty in quickly locating suitable segments for the connection of new equipment. Because the low-voltage power supply and distribution system does not reserve sufficient capacity or take into account the load characteristics of new equipment (such as the starting current, continuous load, load fluctuations, etc. of the equipment) during design, directly adding new equipment may cause system overload, voltage drop, heat accumulation and other problems, which in turn cause equipment damage or failure. Existing fault prediction methods usually perform load analysis through static calculations and estimates, but these methods cannot take into account the dynamic changes of equipment in real time, resulting in the inability to detect potential risks in a timely manner.

[0003] In summary, the existing technology has a technical problem in that it is difficult for the low-voltage power supply and distribution system to identify risks in advance when new equipment is added, resulting in failures or equipment damage, which further affects the safety of the low-voltage power supply and distribution system. Summary of the Invention

[0004] The purpose of this application is to provide a method and system for predicting bus duct wiring faults in a low-voltage power supply and distribution system, so as to solve the technical problem in the prior art that it is difficult for the low-voltage power supply and distribution system to identify risks in advance when facing new equipment, resulting in faults or equipment damage, which further affects the safety of the low-voltage power supply and distribution system.

[0005] In view of the above problems, the present application provides a method and system for predicting bus duct wiring faults in a low-voltage power supply and distribution system.

[0006] In the first aspect, the present application provides a method for predicting bus duct wiring faults in a low-voltage power supply and distribution system, wherein the method is implemented by a low-voltage power supply and distribution system bus duct wiring fault prediction system, wherein the method comprises: constructing a bus duct topology structure according to bus duct wiring information in a low-voltage power distribution system; receiving new equipment requirements, wherein the new equipment requirements include new equipment location information and new equipment load characteristics; performing a card-in node analysis in the bus duct topology structure according to the new equipment location information, and locating multiple alternative Bus duct segmentation, wherein the alternative bus duct segment has a position identification of a plug-in alternative node; after obtaining multiple segmented twin models based on the historical operation digital twin modeling of the multiple alternative bus duct segments, the load characteristics of the newly added equipment are used to drive the multiple segmented twin models to predict bus duct wiring faults and output multiple segment safety factors; the target bus duct segment is positioned in the multiple alternative bus duct segments according to the multiple segment safety factors; in the low-voltage power distribution system, the plug-in alternative node of the target bus duct segment is used as the target plug-in node to perform bus duct plug-in construction of the newly added plug-in box.

[0007] Optionally, the power consumption position is calibrated in the low-voltage power supply and distribution system according to the location information of the newly added equipment, and the plug-in node analysis is performed in the bus duct topology structure according to the calibration results to obtain multiple plug-in candidate nodes; the multiple alternative bus duct segments are obtained by locating the segments to which the multiple plug-in candidate nodes belong in the bus duct topology structure; the positions of the multiple plug-in candidate nodes are mapped and identified in the multiple alternative bus duct segments; and the electrical load characteristics of multiple groups of equipment in the multiple alternative bus duct segments are obtained by calling local data.

[0008] Optionally, multiple safety plug-in association information of the multiple alternative bus duct segments are called from the design information of the low-voltage power supply and distribution system; based on the multiple safety plug-in association information and the multiple groups of equipment electrical load characteristics of the multiple alternative bus duct segments, the accessibility of the plug-in boxes of the multiple alternative bus duct segments is mapped and quantified to obtain multiple accessible safety factors; the multiple accessible safety factors are traversed using the accessible safety threshold to screen and obtain K alternative bus duct segments; based on the segment ID, the K historical power distribution operation data of the K alternative bus duct segments are called from the local operation log, and K segment twin models are constructed based on the K historical power distribution operation data; the newly added equipment load characteristics are used to drive the operation of the K segment twin models to perform bus duct wiring fault prediction, and quantify and output K segment safety factors.

[0009] Optionally, the second historical power distribution operation data of the second alternative bus duct segment is divided according to the change node of the access plug-in box to obtain multiple power distribution operation sub-data, wherein H temperature measurement key nodes are arranged on the second alternative bus duct segment; the first power distribution operation sub-data is extracted from the multiple power distribution operation sub-data, wherein the first power distribution operation sub-data includes current-related time series characteristics, W device load time series characteristics of W connected devices, and H node temperature time series characteristics of H temperature measurement key nodes; a random forest regression algorithm is used, with load-current as input feature and temperature data as label, and the current-related time series characteristics, W device load time series characteristics, and H node temperature time series characteristics as training data to train and obtain a first sub-temperature prediction model; by analogy, after constructing multiple sub-temperature prediction models based on the multiple power distribution operation sub-data, the multiple sub-temperature prediction models are feature-fused by introducing an attention mechanism to output a second segmented twin model; by analogy, the K segmented twin models are constructed based on the K historical power distribution operation data.

[0010] Optionally, new load limits and new overshoot currents are extracted from the load characteristics of the new equipment; after the power supply load disturbance is performed on the K segmented twin models, the new load limits and new overshoot currents are input into the K segmented twin models to predict the key node temperatures, and K node predicted temperature time series arrays are obtained; using a preset temperature threshold, overtemperature frequency statistics are performed on the K node predicted temperature time series arrays to obtain K first-dimensional safety factors; in the process of inputting the new load limits and new overshoot currents into the K segmented twin models to predict the key node temperatures, real-time load statistics are performed to obtain K segmented load time series data; based on the load limit, overload frequency statistics are performed on the K segmented load time series data to obtain K second-dimensional safety factors; the K first-dimensional safety factors and the K second-dimensional safety factors are normalized, and the K segmented safety factors are output.

[0011] Optionally, N plug-in coordinates, N device load limits and N device overshoot currents of N connected devices are extracted from the electrical load characteristics of the first group of devices; the first plug-in box capacity limit, the first load limit and the first overcurrent capacity are extracted from the first safety plug-in association information; the newly added load limit and the newly added overshoot current are extracted from the load characteristics of the newly added devices; the plug-in box capacity is calculated based on the first plug-in alternative node and the N plug-in coordinates of the first alternative bus duct segment to obtain the real-time plug-in capacity characteristics; if the real-time plug-in capacity characteristics meet the first plug-in box capacity limit, the plug-in box accessibility is quantified based on the N device load limits, N device overshoot currents, the newly added load limit, the newly added overshoot current, the first load limit and the first overcurrent capacity, and the first accessibility safety factor is output.

[0012] Optionally, after adding the load limits of the N devices and the newly added load limit, the load deviation between the summed result and the first load limit is calculated to obtain a first load deviation; after adding the overshoot currents of the N devices and the newly added overshoot current, the current deviation between the summed result and the first overcurrent capacity is calculated to obtain a first overcurrent deviation; if the first load deviation and the first overcurrent deviation are both positive values, the first load deviation and the first overcurrent deviation are normalized and added based on a preset weighting rule, and the first accessible safety factor is output.

[0013] Optionally, if the first load deviation and / or the first overcurrent deviation is a negative value, the first accessible safety factor is set to zero.

[0014] Optionally, with the first alternative bus duct segmentation as a spatial constraint, the plug-in box density is calculated according to the first plug-in alternative node and N plug-in coordinates to obtain a real-time plug-in density feature; with the first alternative bus duct segmentation as a spatial constraint, the plug-in box spacing is solved according to the first plug-in alternative node and N plug-in coordinates, and a real-time minimum plug-in spacing feature is output, wherein the real-time plug-in density feature and the real-time minimum plug-in spacing feature constitute the real-time plug-in capacity feature; if the mapping of the real-time plug-in density feature and the real-time minimum plug-in spacing feature satisfies the first plug-in density limit and the first plug-in spacing limit in the first plug-in box capacity limit, it is judged that the real-time plug-in capacity feature satisfies the first plug-in box capacity limit.

[0015] In the second aspect, the present application also provides a low-voltage power supply and distribution system bus duct wiring fault prediction system, which is used to execute the low-voltage power supply and distribution system bus duct wiring fault prediction method as described in the first aspect, wherein the low-voltage power supply and distribution system bus duct wiring fault prediction system includes: a topology structure construction module, which is used to construct a bus duct topology structure according to the bus duct wiring information in the low-voltage power distribution system; an equipment demand receiving module, which is used to receive new equipment requirements, wherein the new equipment requirements include new equipment location information and new equipment load characteristics; a card-in node analysis module, which is used to perform card-in node analysis in the bus duct topology structure according to the new equipment location information, and locate multiple alternative bus duct distribution points. Segment, wherein the alternative bus duct segment has a position identifier for snapping into an alternative node; a fault prediction module, which is used to obtain multiple segmented twin models based on the historical operation digital twin modeling of the multiple alternative bus duct segments, and then use the load characteristics of the newly added equipment to drive the multiple segmented twin models to predict bus duct wiring faults and output multiple segment safety factors; a segment positioning module, which is used to locate the target bus duct segment in the multiple alternative bus duct segments according to the multiple segment safety factors; a snap-in node determination module, which is used to use the snap-in alternative node of the target bus duct segment as the target snap-in node in the low-voltage power distribution system to perform bus duct snap-in construction of the newly added plug-in box.

[0016] One or more technical solutions provided in this application have at least the following beneficial effects:

[0017] A bus duct topology structure is constructed according to the bus duct wiring information in the low-voltage power distribution system; new equipment requirements are received, wherein the new equipment requirements include new equipment location information and new equipment load characteristics; a snap-in node analysis is performed on the bus duct topology structure according to the new equipment location information, and multiple alternative bus duct segments are located, wherein the alternative bus duct segments have location identifiers of snap-in alternative nodes; after digital twin modeling is performed based on the historical operation of the multiple alternative bus duct segments to obtain multiple segmented twin models, the load characteristics of the new equipment are used to drive the multiple segmented twin models to predict bus duct wiring faults, and multiple segmented safety factors are output; a target bus duct segment is located in the multiple alternative bus duct segments according to the multiple segmented safety factors; in the low-voltage power distribution system, the snap-in alternative node of the target bus duct segment is used as the target snap-in node, and the bus duct snap-in construction of the new plug-in box is performed. That is to say, after receiving the demand for new plug-in equipment, the fault risk of plug-in boxes in different positions is predicted based on the current plug-in status of the plug-in boxes in the bus duct trunk system and the power consumption characteristics of the plug-in boxes, and the bus duct sections that meet the plug-in requirements are screened out to locate the plug-in position of the new plug-in box, thereby reducing the time spent on positioning and completing the access and wiring of the new equipment, identifying potential problems in advance, reducing operation and maintenance costs, and improving the safety and stability of the low-voltage power supply and distribution system.

[0018] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, which can be implemented in accordance with the contents of the description, and to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically listed below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in this application or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely exemplary, and a person of ordinary skill in the art can obtain other drawings based on the provided drawings without creative work.

[0020] Figure 1 This is a flow chart of a method for predicting bus duct wiring faults in a low-voltage power supply and distribution system according to the present application;

[0021] Figure 2 This is a structural diagram of the bus duct wiring fault prediction system for the low-voltage power supply and distribution system of this application.

[0022] Description of the accompanying drawings: topology structure construction module 11, equipment demand receiving module 12, stuck node analysis module 13, fault prediction module 14, segment positioning module 15, stuck node determination module 16. DETAILED DESCRIPTION

[0023] This application solves the technical problem in the prior art that it is difficult for the low-voltage power supply and distribution system to identify risks in advance when facing new equipment, which leads to failures or equipment damage, and further affects the safety of the low-voltage power supply and distribution system, by providing a method and system for predicting bus duct wiring faults in a low-voltage power supply and distribution system. After receiving the demand for newly inserted equipment, the fault risk of plug-in boxes at different positions is predicted based on the current plug-in status of the plug-in box in the bus duct trunk system and the power consumption characteristics of the plug-in box. The bus duct segments that meet the plug-in requirements are screened out to locate the plug-in position of the newly added plug-in box, which reduces the time spent on positioning, thereby completing the access and wiring of the newly added equipment, identifying potential problems in advance, reducing operation and maintenance costs, and improving the safety and stability of the low-voltage power supply and distribution system.

[0024] Below, the technical solutions in this application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of this application, rather than all the embodiments of this application. It should be understood that this application is not limited to the example embodiments described herein. Based on the embodiments of this application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application. It should also be noted that, for the convenience of description, only the parts related to this application, rather than all of them, are shown in the accompanying drawings.

[0025] For example, see the attached Figure 1 The present application provides a method for predicting bus duct wiring faults in a low-voltage power supply and distribution system. The method is executed by a low-voltage power supply and distribution system bus duct wiring fault prediction system. The method specifically includes the following steps:

[0026] S100: Construct a bus duct topology structure according to bus duct wiring information in the low-voltage power distribution system.

[0027] Specifically, the existing bus duct wiring information in the low-voltage distribution system is collected through electrical design drawings, site surveys, and real-time monitoring data. This includes detailed information about each bus duct segment, such as the physical dimensions (e.g., length, cross-sectional area), rated electrical characteristics (e.g., rated current, impedance, allowable voltage drop threshold), the spatial location of connection nodes, and the presence of branches in the low-voltage distribution system. Based on this collected wiring information, power system modeling software or computer-aided design (CAD) software is used to draw a bus duct topology diagram based on the collected bus duct wiring information. This diagram includes the connection method of the bus duct segments, the current transmission paths between segments, branch points, and node locations. Each bus duct segment is connected by virtual connection nodes, at which parameters such as current and voltage can be set. The model should include the physical characteristics (e.g., length, cross-sectional area) and electrical characteristics (e.g., impedance, rated current) of the bus duct.

[0028] Topology refers to the connections and arrangement of elements within a system. Busway topology refers to the connections between busway segments, including the layout of each segment, node configuration, and branching. The constructed busway topology is further analyzed and optimized to ensure full consistency with the busway wiring information in the low-voltage distribution system. By constructing a busway topology, the layout and connections of the busway in the entire low-voltage distribution system are determined, facilitating system analysis and troubleshooting.

[0029] S200: Receive a new device requirement, wherein the new device requirement includes new device location information and new device load characteristics.

[0030] Specifically, based on the equipment location diagram, equipment installation guide or technical specification of the newly added equipment, the corresponding equipment access requirements are obtained, including key data such as the location information and load characteristics of the newly added equipment, to help evaluate the impact of the newly added equipment on the low-voltage power distribution system and optimize and adjust the low-voltage power distribution system. The load characteristic information of the equipment is extracted, including the rated power, current, voltage requirements, power factor, etc. of the equipment, to determine its impact on the low-voltage power distribution system and ensure that the bus duct and power equipment will not have problems such as overload and excessive voltage drop after the new load is connected. The location information of the newly added equipment refers to the accessible location of the newly added equipment in the low-voltage power distribution system, including the specific location where the equipment is installed, the bus duct section to which it is connected, and the adjacent load equipment.

[0031] Based on the device's location information and load characteristics, we analyze the potential impact of new equipment on the low-voltage distribution system, assess the current load changes after the new equipment is connected, check whether the voltage meets requirements, and predict whether overload, excessive voltage drop, and other problems will occur. By receiving requests for new equipment and analyzing the device's location information and load characteristics, we accurately assess the impact of new equipment on the low-voltage distribution system and ensure the safe and stable connection of new equipment.

[0032] S300: performing a snap-in node analysis on the bus duct topology structure according to the location information of the newly added device, and locating a plurality of candidate bus duct segments, wherein the candidate bus duct segments have position identifiers of snap-in candidate nodes.

[0033] Furthermore, the present application S300 includes:

[0034] According to the location information of the newly added equipment, the power consumption position is calibrated in the low-voltage power supply and distribution system, and according to the calibration results, the plug-in node analysis is performed on the bus duct topology structure to obtain multiple plug-in alternative nodes; by locating the segments to which the multiple plug-in alternative nodes belong in the bus duct topology structure, the multiple alternative bus duct segments are obtained; the positions of the multiple plug-in alternative nodes are mapped and identified in the multiple alternative bus duct segments; and the electrical load characteristics of multiple groups of equipment in the multiple alternative bus duct segments are obtained by calling local data.

[0035] Specifically, based on the location information of the newly added equipment, the specific location where the newly added equipment will be connected to the power distribution system is identified and calibrated to determine the equipment access point and the area where it is located. On the bus duct topology diagram, the power consumption location calibration results are analyzed to determine the possible insertion nodes, that is, the points where the new equipment can be connected. Based on the equipment location, multiple possible insertion nodes are analyzed (such as the connection nodes of bus duct C, bus duct D, and bus duct E). These nodes can be branch connection points, trunk line access points or endpoints, etc. Insertion node analysis refers to analyzing which node in the low-voltage power distribution system (i.e., the connection point of the bus duct) the new equipment can be connected to, such as the branch, access point or specific connection area of the bus duct, to determine multiple possible access points (insertion nodes).

[0036] The multiple bus duct segments described as the candidate nodes are located in the bus duct topology, i.e., multiple candidate bus duct segments, i.e., multiple bus duct segments that may receive new equipment, representing candidate areas that can accommodate the load of the new equipment. The multiple candidate nodes are mapped and identified in the multiple candidate bus duct segments, and the candidate nodes and their specific locations are clearly marked in the topology diagram of each segment.

[0037] By calling local data, the equipment information, load characteristics and operating status of the distribution system are obtained from the local storage system (such as database, data warehouse, real-time monitoring system, etc.). The electrical load characteristics of the equipment in each alternative busbar segment are obtained, that is, the load characteristics of the connected equipment in each segment, including equipment location, equipment load limit and equipment overshoot current, etc. The electrical load characteristics of some equipment connected to the low-voltage distribution system, such as motors and air-conditioning units, will be recorded in the database, including data such as the equipment's rated power, starting current, continuous working current, load limit, overshoot current, etc.

[0038] By receiving the location information of newly added equipment and performing power location calibration and card-in node analysis, the reasonable access points and appropriate bus duct segmentation of multiple newly added equipment are determined to avoid problems such as overload and voltage drop after the equipment is connected.

[0039] S400: After obtaining multiple segmented twin models based on the historical operation digital twin modeling of the multiple alternative bus duct segments, the load characteristics of the newly added equipment are used to drive the multiple segmented twin models to predict bus duct wiring faults and output multiple segmented safety factors.

[0040] Specifically, the historical operating data of multiple alternative bus duct segments are obtained respectively, and digital twin modeling is performed based on this data. Digital twin is a method of simulating a physical entity or system through real-time data, which can reflect the operating status, performance and behavior of the physical system through real-time data. Each twin model takes into account the different characteristics, load conditions and operating environment of each segment. According to the load characteristics of the newly added equipment, it is input into multiple segmented twin models to predict the bus duct wiring fault and obtain multiple segment safety factors. The detailed process is shown in the corresponding explanation below. For the sake of brevity of the specification, it will not be described in detail here.

[0041] Furthermore, the present application S400 includes:

[0042] From the design information of the low-voltage power supply and distribution system, multiple safety plug-in association information of the multiple alternative bus duct segments are called; based on the multiple safety plug-in association information and the multiple groups of equipment electrical load characteristics of the multiple alternative bus duct segments, the accessibility of the plug-in boxes of the multiple alternative bus duct segments is mapped and quantified to obtain multiple accessible safety factors; the accessible safety threshold is used to traverse the multiple accessible safety factors to screen and obtain K alternative bus duct segments; according to the segment ID, the K historical power distribution operation data of the K alternative bus duct segments are called from the local operation log, and K segment twin models are constructed based on the K historical power distribution operation data; the newly added equipment load characteristics are used to drive the operation of the K segment twin models to perform bus duct wiring fault prediction, and quantify the output of K segment safety factors.

[0043] Specifically, the design information of the low-voltage power supply and distribution system is obtained, and all relevant electrical parameters and configuration data set for the low-voltage power distribution system during the design phase, including but not limited to the topology of the distribution system (including bus duct layout, equipment location, etc.), the design capacity, connection method, and plug-in point location of each segmented bus duct, the electrical parameters of the bus duct (such as rated current, maximum current, rated voltage, etc.), the design capacity, load characteristics, and overcurrent capacity of each plug-in box, etc.

[0044] In a low-voltage power distribution system, each section of the bus duct is equipped with a plug-in point or access device. Based on the system design information, the safety plug-in associated information of each alternative section can be extracted, including load characteristics, plug-in box capacity limitations and overcurrent capabilities. From the multiple sets of equipment electrical load characteristics of multiple alternative bus duct sections, the coordinates, load limits and overshoot currents of multiple connected devices in each section are extracted; the plug-in box capacity limitations, load limits and maximum overcurrent capabilities corresponding to each section are extracted from multiple safety plug-in associated information; and the newly added load limits and newly added overshoot currents are extracted from the load characteristics of the newly added devices. The plug-in box capacity is calculated based on the connected devices and newly added devices in each section to obtain the plug-in capacity characteristics, which are compared with the plug-in box capacity limitations corresponding to each section. If this limitation is met, it means that it can be safely inserted. At this time, the accessibility of the plug-in box is quantified for each section based on the newly added equipment, and multiple accessibility safety factors are obtained. The detailed process explanation is in the corresponding section below, and for the sake of brevity of the specification, it will not be described in detail here.

[0045] Obtaining the accessibility safety threshold, a predetermined safety value, determines whether each candidate busbar segment meets the safety requirements for connecting new equipment. Typically, the safety threshold is set based on factors such as electrical load, overcurrent capability, and equipment tolerance. Only segments with a safety factor greater than or equal to the accessibility safety threshold pass the screening process. Multiple accessibility safety factors are compared with the accessibility safety threshold to screen out the candidate busbar segments that meet the accessibility safety threshold, resulting in K candidate busbar segments. By screening the candidate busbar segments for K segments that meet the accessibility safety threshold, the time required to locate the node is indirectly shortened.

[0046] According to the ID of each segment, K historical distribution operation data corresponding to K alternative busbar segments are called from the local operation log. Based on the current correlation time series characteristics, load time series characteristics of connected equipment, and temperature time series characteristics of key nodes in the historical distribution operation data, multiple temperature prediction models for each alternative busbar segment are trained. The attention mechanism is introduced to perform feature fusion on multiple sub-temperature prediction models, and K segmented twin models of the K alternative busbar segments are output.

[0047] Based on the load characteristics of the newly added equipment, K segmented twin models are driven to predict busbar wiring faults for K candidate busbar segments. These K segmented twin models are used to predict the temperatures of key nodes in these K candidate busbar segments. This temperature prediction process is monitored in real time, and real-time load statistics are compiled. By calculating the frequency of overtemperature and overload events, K segmented safety factors are calculated for each of the K selected candidate busbar segments, quantifying and evaluating the safety of each busbar segment after the addition of new equipment.

[0048] Furthermore, the present application further comprises the following steps:

[0049] N plug-in coordinates, N device load limits and N device overshoot currents of N connected devices are extracted from the electrical load characteristics of the first group of devices; the first plug-in box capacity limit, the first load limit and the first overcurrent capacity are extracted from the first safety plug-in association information; the newly added load limit and the newly added overshoot current are extracted from the load characteristics of the newly added devices; the plug-in box capacity is calculated based on the first plug-in alternative node of the first alternative bus duct segment and the N plug-in coordinates to obtain the real-time plug-in capacity characteristics; if the real-time plug-in capacity characteristics meet the first plug-in box capacity limit, the plug-in box accessibility is quantified based on the N device load limits, N device overshoot currents, the newly added load limit, the newly added overshoot current, the first load limit and the first overcurrent capacity, and the first accessibility safety factor is output.

[0050] Taking the first alternative bus duct segment as a spatial constraint, the plug-in box density is calculated according to the first plug-in alternative node and N plug-in coordinates to obtain a real-time plug-in density feature; taking the first alternative bus duct segment as a spatial constraint, the plug-in box spacing is solved according to the first plug-in alternative node and N plug-in coordinates, and a real-time minimum plug-in spacing feature is output, wherein the real-time plug-in density feature and the real-time minimum plug-in spacing feature constitute the real-time plug-in capacity feature; if the mapping of the real-time plug-in density feature and the real-time minimum plug-in spacing feature satisfies the first plug-in density limit and the first plug-in spacing limit in the first plug-in box capacity limit, it is determined that the real-time plug-in capacity feature satisfies the first plug-in box capacity limit.

[0051] After adding the load limits of the N devices and the newly added load limit, the load deviation between the summed result and the first load limit is calculated to obtain a first load deviation; after adding the overshoot currents of the N devices and the newly added overshoot current, the current deviation between the summed result and the first overcurrent capacity is calculated to obtain a first overcurrent deviation; if the first load deviation and the first overcurrent deviation are both positive values, the first load deviation and the first overcurrent deviation are normalized and added based on a preset weighting rule, and the first accessible safety factor is output.

[0052] If the first load deviation and / or the first overcurrent deviation is a negative value, the first accessible safety factor is set to zero.

[0053] Specifically, a first candidate bus duct segment is randomly selected from a plurality of candidate bus duct segments, and the corresponding first safety plug association information is matched in a plurality of safety plug association information, and the corresponding first group of device electrical load characteristics are matched in a plurality of groups of device electrical load characteristics. From the first group of device electrical load characteristics, the plug coordinates of N connected devices are obtained, and the plug coordinates of each device represent its position in the electrical system. Each connected device has its own load limit and overshoot current, which are usually described in the technical specifications of the device and stored in the system database. The extracted plug coordinates, device load limit and overshoot current are integrated to form a data table or data set about the connected devices. Each device will have corresponding plug coordinates, load limit and overshoot current, as shown in Table 1:

[0054] Table 1 Equipment electrical load characteristics

[0055] Device Name Plug-in coordinates Load limit Overshoot current Device A (1,5) 10kW 50A Device B (1,7) 15kW 60A Device C (2,6) 27kW 82A

[0056] Safety parameters related to the plug-in box or plug-in point in the first alternative bus duct segment are extracted from the first safety plug-in association information, including the first plug-in box capacity limit, the first load limit, and the first overcurrent capacity. The plug-in box capacity limit refers to the maximum current or power value that the plug-in box can withstand, which determines how much load the plug-in box can safely carry. Exceeding this capacity may cause the plug-in box to be overloaded, damaged, or malfunction. The load limit refers to the maximum load characteristic that the plug-in box or bus duct system can continuously carry, usually including the power value of the equipment. If the load exceeds this limit, the equipment may not operate safely or cause system failure. Overcurrent capacity refers to the maximum current value that the plug-in box can withstand and its duration in the event of an overload current. By considering the plug-in box capacity limit, load limit, and overcurrent capacity, the risk of failure caused by improper access can be reduced.

[0057] The new load limit and the new overshoot current are extracted from the load characteristics of the new device, that is, the maximum load requirement and the maximum overshoot current of the new device.

[0058] Determine the first alternative bus duct segment's first snap-in alternative node, that is, an alternative node position determined in the segment. The snap-in node is the connection point when the device is connected to the bus duct. Use the first alternative bus duct segment as a spatial constraint, that is, perform spatial restrictions according to the bus duct length, cross-sectional area, etc. of the first alternative bus duct segment, perform plug-in box density calculation on the first snap-in alternative node and N plug-in coordinates, and obtain real-time plug-in density characteristics. Typically, the number of connected devices of the first alternative bus duct segment is determined by the sum of the first snap-in alternative node and the N plug-in coordinates, and the real-time plug-in density characteristics of the first alternative bus duct segment are obtained based on the ratio of the number of connected devices to the bus duct length. For example, assuming that the length of the first alternative busbar segment is 100 meters, 5 devices have been installed, and the plug-in coordinates of the devices are (x1, y1), (x2, y2), (x3, y3), (x4, y4), (x5, y5), combined with the first plug-in alternative node, the calculated plug-in box density is 6 / 100 = 0.06 devices / meter.

[0059] Using the first candidate busbar segmentation as a spatial constraint, the first candidate node and N plug-in coordinates are used to solve the plug-in box spacing, calculate the spacing between devices, and calculate the physical distance between devices using the plug-in coordinates between each two devices. The minimum plug-in spacing refers to the shortest distance between two adjacent plug-in boxes. Based on the real-time plug-in density characteristics and the real-time minimum plug-in spacing characteristics, the real-time plug-in capacity characteristics are determined. This is a comprehensive quantitative characteristic of the current status of the plug-in box, considering multiple factors such as the plug-in box density and plug-in spacing, to evaluate whether the device to be inserted meets the capacity restrictions of the plug-in box.

[0060] The real-time plug density and minimum spacing characteristics are obtained after the first candidate node is inserted into the plug-in box. When the real-time plug density and minimum spacing characteristics meet the first plug density and spacing limits, the first candidate busbar segment's safety requirements, including capacity limits, load limits, and overcurrent capacity, are determined. Only when both the density and minimum spacing characteristics after insertion meet the limits can the real-time plug capacity characteristics be determined to meet the first plug-in box capacity limits, indicating that the newly added device can now be safely accommodated.

[0061] The N device load limits of the N connected devices and the newly added load limit of the newly added device are added together to obtain the summed result. The load deviation between this summed result and the first load limit is calculated to obtain the first load deviation. The difference between the summed device load limits (including the load limits of the connected devices and the newly added devices) and the maximum load limit set by the low-voltage power distribution system is calculated. If the deviation is a positive value, it means that the low-voltage power distribution system has capacity to support the newly added devices. If it is a negative value, it means that the load limit has been exceeded and the low-voltage power distribution system may be overloaded.

[0062] Similarly, the N device overshoot currents of the N connected devices and the newly added overshoot current of the newly added device are added together to obtain the summed result. The current deviation between this summed result and the first overcurrent capability is calculated to obtain the first overcurrent deviation. For example, if the load limit of connected device 1 is 50kW and the overshoot current is 20A, the load limit of device 2 is 70kW and the overshoot current is 30A, the load limit of the newly added device is 30kW and the overshoot current is 15A, the first load limit is 200kW, and the first overcurrent capability is 100A, then the summed load limit result is 50+70+30=150kW, and the first load deviation is 200-150=50kW; the summed current result is 20+30+15=65A, and the first overcurrent deviation is 35A. Both are positive values, indicating that this device can be accepted.

[0063] When both the load deviation and the overcurrent deviation are positive, it indicates that the low-voltage power distribution system has the ability to accommodate new equipment. According to the preset weighting rule (for example, the weights are 0.6 of the load deviation and 0.4 of the overcurrent deviation), the first load deviation and the first overcurrent deviation are normalized and added together to obtain the first accessible safety factor. For example, the calculation formula for the first accessible safety factor is: Among them, α and β are the weight coefficients of the first load deviation and the first overcurrent deviation respectively, L1 is the first load deviation, L max is the first load limit (i.e., the maximum load limit), A1 is the first overcurrent deviation, A max is the first overcurrent capacity (i.e., the maximum overcurrent capacity), and the first accessible safety factor is

[0064] The same steps are repeated for multiple candidate busbar sections, quantitatively evaluating their safety performance when connecting new equipment, resulting in multiple acceptable safety factors. For each candidate busbar section, a quantitative analysis of the plug-in box capacity, load characteristics, and overcurrent capability is performed based on its electrical load characteristics and safety plug-in information. The acceptable safety factor is calculated by integrating various parameters (such as load deviation and overcurrent deviation).

[0065] If either the first load deviation or the first overcurrent deviation is negative, the current operating status of the plug-in box or busbar segment does not meet the requirements for safe access. The load capacity or current carrying capacity of the busbar segment is insufficient to ensure the safe access of new equipment. The access safety factor for that segment is set to zero, indicating that it is unsuitable for any new equipment. By setting the safety factor to zero when the load deviation and overcurrent deviation are negative, the busbar segment is protected from overload or overcurrent failures caused by the inability to support new equipment.

[0066] By quantifying the accessibility of plug-in boxes in multiple alternative bus duct segments, we ensure that the selection of access points for new equipment is accurate and safe, reduce the risk of failure due to inappropriate access, and thus improve the safety and stability of the low-voltage distribution system.

[0067] Furthermore, the present application further comprises the following steps:

[0068] The second historical power distribution operation data of the second alternative bus duct segment is divided according to the change node of the access plug-in box to obtain multiple power distribution operation sub-data, wherein H temperature measurement key nodes are arranged on the second alternative bus duct segment; the first power distribution operation sub-data is extracted from the multiple power distribution operation sub-data, wherein the first power distribution operation sub-data includes current-related time series features, W device load time series features of W connected devices, and H node temperature time series features of H temperature measurement key nodes; a random forest regression algorithm is adopted, with load-current as input feature and temperature data as label, and the current-related time series features, W device load time series features, and H node temperature time series features as training data to train and obtain a first sub-temperature prediction model; by analogy, after constructing multiple sub-temperature prediction models based on the multiple power distribution operation sub-data, the multiple sub-temperature prediction models are feature-fused by introducing an attention mechanism to output a second segmented twin model; by analogy, the K segmented twin models are constructed based on the K historical power distribution operation data.

[0069] Specifically, each alternative bus duct segment is provided with a plurality of key temperature measurement nodes for monitoring temperature data. Taking the second alternative bus duct segment as an example, H key temperature measurement nodes are provided on it for real-time monitoring of the temperature of the bus duct. Determine how to divide the second alternative bus duct segment by the change in the access position of the plug-in box. The second historical power distribution operation data corresponding to the second alternative bus duct segment is divided according to the plug-in box access change node to obtain a plurality of power distribution operation sub-data. The second historical power distribution operation data is the power distribution operation data of the second alternative bus duct segment in the past period of time, which records the power distribution operation status of the segment under different time and load conditions.

[0070] A random selection is made from multiple power distribution operation sub-data as the first power distribution operation sub-data, including current-related timing characteristics, W device load timing characteristics of W connected devices, and H node temperature timing characteristics of H key temperature measurement nodes. The current-related timing characteristics are characteristics of current changes over time, usually describing characteristic data such as the waveform, periodic changes, and impact current of the current. The W device load timing characteristics of the W connected devices are the load changes of the connected devices on the small segment within a specific time period, including the changes in the device load over time, the amplitude and frequency of the load fluctuations, etc. The H node temperature timing characteristics of the H key temperature measurement nodes are the temperature changes of the H key nodes (nodes equipped with temperature sensors) in the bus duct segment within a specific time period.

[0071] Random forest is an ensemble learning method consisting of multiple decision trees. In regression tasks, random forest regression improves prediction accuracy by training multiple decision trees and averaging their results. Each decision tree is trained based on a random selection of sample subsets and features, and the final output result is the average of the predicted values of all trees. Determine the input features and target variables, use the current correlation time series features and equipment load time series features as input features, and the node temperature time series features as labels. Perform data preprocessing on the first distribution operation sub-data, including data cleaning and standardization, to convert it into features that can be recognized by the model. The processed data set is divided into a training set and a validation set. Usually 80% of the training set is used for training and 20% of the validation set is used for validation.

[0072] The random forest regression algorithm learns the relationship between current, load, and temperature from the training data. During training, the model constructs multiple decision trees, each trained on a different subset of data and random features, resulting in multiple models. Finally, the random forest regression algorithm averages the outputs of these decision trees to produce the final prediction result. This training results in a temperature prediction model based on current and device load, known as the first sub-temperature prediction model. This model can predict node temperature changes based on input data such as current and load. The trained model's prediction accuracy should be evaluated through cross-validation or a test set. If the temperature prediction error is large, the model's hyperparameters (such as the number of trees and depth) can be adjusted to optimize the model. For example, the mean squared error between the true and predicted values can be calculated to measure the difference between the predicted and actual temperatures. Model convergence criteria are set, such as a validation set loss change of less than 0.01 for five consecutive rounds or a training set accuracy of 95%. When the model reaches convergence, training is terminated, resulting in the first sub-temperature prediction model.

[0073] This process is repeated, training multiple sub-temperature prediction models based on multiple distribution operation sub-data. Each model corresponds to a segment segmented by the node change associated with the plug-in box. An attention mechanism is introduced to fuse the features of these sub-temperature prediction models and perform a weighted average of their prediction results. Weights are automatically assigned by the attention mechanism based on the importance of each model. The attention mechanism mimics how humans focus on the most relevant components of multiple inputs. By assigning weights to input features, it emphasizes features with a greater impact on the model, thereby enhancing the model's representational capabilities. Feature fusion combines features from different sources or models and leverages this comprehensive information to improve the model's prediction accuracy. In this context, feature fusion involves fusing the outputs of multiple sub-temperature prediction models through the attention mechanism to form a final twin model. The attention mechanism assigns different weights to each sub-model output based on the importance of the input features (such as current and load timing characteristics). By fusing the outputs of these multiple sub-temperature prediction models, a second segmented twin model is ultimately generated, capable of accurately predicting changes in parameters such as temperature within the segment.

[0074] Similarly, based on K historical distribution operation data for K candidate busbar segments, K segmented twin models are constructed, each representing the operating status of the corresponding busbar segment. By extracting features from multiple distribution operation sub-data, a temperature prediction sub-model is constructed using a random forest regression algorithm. An attention mechanism is introduced for feature fusion to construct a busbar twin model. This enables accurate temperature prediction and load analysis, improving the reliability of the low-voltage power supply and distribution system and providing more accurate operational status predictions.

[0075] Furthermore, the present application further comprises the following steps:

[0076] Extract new load limits and new overshoot currents from the load characteristics of the new equipment; after performing power supply load disturbance on the K segmented twin models, input the new load limits and new overshoot currents into the K segmented twin models to predict the key node temperatures, and obtain K node predicted temperature time series arrays; use a preset temperature threshold to perform overtemperature frequency statistics on the K node predicted temperature time series arrays, and obtain K first-dimensional safety factors; in the process of inputting the new load limits and new overshoot currents into the K segmented twin models to predict the key node temperatures, perform real-time load statistics to obtain K segmented load time series data; based on the load limit, perform overload frequency statistics on the K segmented load time series data to obtain K second-dimensional safety factors; normalize the K first-dimensional safety factors and the K second-dimensional safety factors, and output the K segmented safety factors.

[0077] Specifically, the new load limit and new overshoot current of the newly added equipment are extracted from the load characteristics of the newly added equipment. The power supply load disturbance is applied to the K segmented twin models to simulate the changes in the distribution system load after the new equipment is connected, and the system response under the load disturbance is examined, especially the changes in temperature and load. The load limit and overshoot current of the newly added equipment are added as input disturbances to the K segmented twin models. Based on the input information, the K segmented twin models simulate the temperature prediction of each segment under the newly added load, output the temperature prediction data of multiple key nodes in each segment, form a time series, and obtain the K node predicted temperature time series array.

[0078] A temperature threshold (the upper temperature limit) is preset. When a node's temperature exceeds this threshold, it is considered an overtemperature condition. The number of times each node's temperature exceeds the preset threshold is counted. This means the number of times each temperature sequence in the K node predicted temperature time series array exceeds the preset threshold. This indicates whether temperature overshoots are frequent during operation, potentially affecting device stability. Based on this overtemperature frequency count, K first-dimensional safety factors are derived.

[0079] When the newly added load limit and overshoot current are input into the twin model for temperature prediction of key nodes, real-time load statistics are collected, load changes are monitored in real time, and K segmented load time series data are generated. Similarly, a load limit is preset, i.e., the upper limit of the load is set. When the node load exceeds this threshold, an overload occurs. The number of times the load limit is exceeded in the K segmented load time series data, i.e., the overload frequency, is counted to obtain K second-dimensional safety factors.

[0080] The thermal safety factor and the load safety factor are combined and normalized to produce the final segmental safety factor. Specifically, for each of the K candidate busbar segments, the first- and second-dimensional safety factors are calculated based on the frequency of overtemperature and overload events. These safety factors are then normalized, converting the different dimensions of the safety factors to a value between 0 and 1 to ensure they are compared on the same scale. The normalized safety factors can then be combined using methods such as weighted averaging to produce the final safety factor for each segment.

[0081] By comprehensively considering the safety factors of both temperature and load, the safety margin of each segment of the power distribution system can be comprehensively assessed, and potential overload or overheating risks can be effectively predicted. By inputting the load characteristics of newly added equipment into the segmented twin model for temperature prediction and load statistics, combined with frequency statistics of overtemperature and overload, the safety factor of each segment of the low-voltage power supply and distribution system can be accurately assessed. Generally, overheating or overload will lead to a reduced safety factor. In other words, the higher the segment safety factor, the higher the safety, and it can be targeted and inserted into the node. Through accurate temperature prediction and load statistics, potential problems in the system after the addition of new equipment can be promptly discovered, thereby reducing the risk of failure.

[0082] S500: Locating a target bus duct segment among the multiple candidate bus duct segments according to the multiple segment safety factors.

[0083] S600: In the low-voltage power distribution system, the plug-in candidate node of the target bus duct segment is used as the target plug-in node to perform bus duct plug-in construction for the newly added plug-in box.

[0084] Specifically, based on the multiple segment safety factors obtained after the bus duct wiring fault prediction, the highest segment safety factor is determined, and the bus duct segment corresponding to the highest segment safety factor is matched among multiple alternative bus duct segments as the target bus duct segment. One or more corresponding bus duct segments with higher safety factors can be selected to avoid special situations. When the highest alternative bus duct segment is unsafe or other special situations occur, the second best one can also be selected as the target bus duct segment. The ultimate goal of locating the target segment is to determine which segments are most suitable for connecting new equipment to avoid problems such as overload or overheating. The selected target segment should have a higher safety factor and be able to effectively share the new load. By evaluating and locating the safety factors of multiple segments, it is ensured that the connection of new equipment will not cause overload or other faults, thereby enhancing the stability of the distribution system.

[0085] The target busbar segment is the busbar segment in the low-voltage distribution system that has been safety-assessed and selected based on requirements to accommodate the newly added power distribution box. It possesses a high safety factor and load capacity to support the connection of new equipment. The target insertion node is the evaluated and determined location on the selected target busbar segment for inserting the newly added power distribution box. After the target busbar segment is determined, the candidate insertion node corresponding to that segment is designated as the target insertion node.

[0086] After the target insertion node is selected, the bus duct insertion construction is carried out. Connect the plug-in box to the target insertion node through cables, conductors, etc. to ensure that electrical signals and currents can pass smoothly. Ensure a stable connection between the plug-in box and the bus duct at the target insertion node to avoid the risk of looseness or poor contact. After the insertion is completed, perform an electrical test to check whether the connection complies with electrical specifications and ensure that there are no safety hazards. By clarifying the location of the target bus duct segment and the insertion node, ensure that the access of the plug-in box will not cause overload, poor contact or electrical failure, improve the installation efficiency of the plug-in box, and avoid unnecessary downtime or extended construction period.

[0087] In summary, the method for predicting bus duct wiring faults in a low-voltage power supply and distribution system provided by this application has the following beneficial effects:

[0088] A bus duct topology structure is constructed according to the bus duct wiring information in the low-voltage power distribution system; new equipment requirements are received, wherein the new equipment requirements include new equipment location information and new equipment load characteristics; a snap-in node analysis is performed on the bus duct topology structure according to the new equipment location information, and multiple alternative bus duct segments are located, wherein the alternative bus duct segments have location identifiers of snap-in alternative nodes; after digital twin modeling is performed based on the historical operation of the multiple alternative bus duct segments to obtain multiple segmented twin models, the load characteristics of the new equipment are used to drive the multiple segmented twin models to predict bus duct wiring faults, and multiple segmented safety factors are output; a target bus duct segment is located in the multiple alternative bus duct segments according to the multiple segmented safety factors; in the low-voltage power distribution system, the snap-in alternative node of the target bus duct segment is used as the target snap-in node, and the bus duct snap-in construction of the new plug-in box is performed. That is to say, after receiving the demand for new plug-in equipment, the fault risk of plug-in boxes in different positions is predicted based on the current plug-in status of the plug-in boxes in the bus duct trunk system and the power consumption characteristics of the plug-in boxes, and the bus duct sections that meet the plug-in requirements are screened out to locate the plug-in position of the new plug-in box, thereby reducing the time spent on positioning and completing the access and wiring of the new equipment, identifying potential problems in advance, reducing operation and maintenance costs, and improving the safety and stability of the low-voltage power supply and distribution system.

[0089] Example 2, based on the same inventive concept as the low-voltage power supply and distribution system bus duct wiring fault prediction method in the above-mentioned Example 1, this application also provides a low-voltage power supply and distribution system bus duct wiring fault prediction system, please refer to the attached Figure 2 , the low-voltage power supply and distribution system bus duct wiring fault prediction system includes:

[0090] A topology construction module 11 is used to construct a bus duct topology structure according to bus duct wiring information in a low-voltage power distribution system; an equipment demand receiving module 12 is used to receive new equipment requirements, wherein the new equipment requirements include new equipment location information and new equipment load characteristics; a card-in node analysis module 13 is used to perform card-in node analysis in the bus duct topology structure according to the new equipment location information, and locate multiple candidate bus duct segments, wherein the candidate bus duct segments have position identifiers for card-in candidate nodes; a fault prediction module 14 is used to perform card-in node analysis in the bus duct topology structure according to the new equipment location information, and locate multiple candidate bus duct segments, wherein the candidate bus duct segments have position identifiers for card-in candidate nodes; a fault prediction module 14 is used to perform card-in node analysis in the bus duct topology structure according to the new equipment location information, and locate multiple candidate bus duct segments according to the new equipment location information ... and locate multiple candidate bus duct segments according to the new equipment location information, and locate multiple candidate bus duct segments according to the new equipment location information; a fault prediction module 14 is used to perform card-in node analysis in the bus duct topology structure according to the new equipment location information, and locate multiple candidate bus duct segments according to the new equipment location information, and locate multiple candidate bus duct segments according to the new equipment location information; a fault prediction module 14 is used to perform card-in node analysis in the bus duct topology structure according to the new equipment location information, and locate multiple candidate bus duct segments according to the new equipment location information, and locate multiple candidate bus duct segments according to the new equipment location information; a fault prediction module 14 is used to perform card-in node analysis in the bus duct topology structure according to the new equipment location information, and locate multiple candidate bus duct segments according to the new equipment location information; a The historical operation digital twin modeling of multiple alternative bus duct segments is carried out to obtain multiple segmented twin models, and the load characteristics of the newly added equipment are used to drive the multiple segmented twin models to predict bus duct wiring faults and output multiple segmented safety factors; a segmented positioning module 15 is used to locate the target bus duct segment in the multiple alternative bus duct segments according to the multiple segmented safety factors; a card-in node determination module 16 is used to use the card-in alternative node of the target bus duct segment as the target card-in node in the low-voltage power distribution system to perform bus duct card-in construction of the newly added plug-in box.

[0091] Furthermore, the stuck node analysis module 13 in the low-voltage power supply and distribution system bus duct wiring fault prediction system is further configured to:

[0092] According to the location information of the newly added equipment, the power consumption position is calibrated in the low-voltage power supply and distribution system, and according to the calibration results, the plug-in node analysis is performed on the bus duct topology structure to obtain multiple plug-in alternative nodes; by locating the segments to which the multiple plug-in alternative nodes belong in the bus duct topology structure, the multiple alternative bus duct segments are obtained; the positions of the multiple plug-in alternative nodes are mapped and identified in the multiple alternative bus duct segments; and the electrical load characteristics of multiple groups of equipment in the multiple alternative bus duct segments are obtained by calling local data.

[0093] Furthermore, the fault prediction module 14 in the low-voltage power supply and distribution system bus duct wiring fault prediction system is further configured to:

[0094] From the design information of the low-voltage power supply and distribution system, multiple safety plug-in association information of the multiple alternative bus duct segments are called; based on the multiple safety plug-in association information and the multiple groups of equipment electrical load characteristics of the multiple alternative bus duct segments, the accessibility of the plug-in boxes of the multiple alternative bus duct segments is mapped and quantified to obtain multiple accessible safety factors; the accessible safety threshold is used to traverse the multiple accessible safety factors to screen and obtain K alternative bus duct segments; according to the segment ID, the K historical power distribution operation data of the K alternative bus duct segments are called from the local operation log, and K segment twin models are constructed based on the K historical power distribution operation data; the newly added equipment load characteristics are used to drive the operation of the K segment twin models to perform bus duct wiring fault prediction, and quantify the output of K segment safety factors.

[0095] Furthermore, the fault prediction module 14 in the low-voltage power supply and distribution system bus duct wiring fault prediction system is further configured to:

[0096] N plug-in coordinates, N device load limits and N device overshoot currents of N connected devices are extracted from the electrical load characteristics of the first group of devices; the first plug-in box capacity limit, the first load limit and the first overcurrent capacity are extracted from the first safety plug-in association information; the newly added load limit and the newly added overshoot current are extracted from the load characteristics of the newly added devices; the plug-in box capacity is calculated based on the first plug-in alternative node of the first alternative bus duct segment and the N plug-in coordinates to obtain the real-time plug-in capacity characteristics; if the real-time plug-in capacity characteristics meet the first plug-in box capacity limit, the plug-in box accessibility is quantified based on the N device load limits, N device overshoot currents, the newly added load limit, the newly added overshoot current, the first load limit and the first overcurrent capacity, and the first accessibility safety factor is output.

[0097] Furthermore, the fault prediction module 14 in the low-voltage power supply and distribution system bus duct wiring fault prediction system is further configured to:

[0098] Taking the first alternative bus duct segment as a spatial constraint, the plug-in box density is calculated according to the first plug-in alternative node and N plug-in coordinates to obtain a real-time plug-in density feature; taking the first alternative bus duct segment as a spatial constraint, the plug-in box spacing is solved according to the first plug-in alternative node and N plug-in coordinates, and a real-time minimum plug-in spacing feature is output, wherein the real-time plug-in density feature and the real-time minimum plug-in spacing feature constitute the real-time plug-in capacity feature; if the mapping of the real-time plug-in density feature and the real-time minimum plug-in spacing feature satisfies the first plug-in density limit and the first plug-in spacing limit in the first plug-in box capacity limit, it is determined that the real-time plug-in capacity feature satisfies the first plug-in box capacity limit.

[0099] Furthermore, the fault prediction module 14 in the low-voltage power supply and distribution system bus duct wiring fault prediction system is further configured to:

[0100] After adding the load limits of the N devices and the newly added load limit, the load deviation between the summed result and the first load limit is calculated to obtain a first load deviation; after adding the overshoot currents of the N devices and the newly added overshoot current, the current deviation between the summed result and the first overcurrent capacity is calculated to obtain a first overcurrent deviation; if the first load deviation and the first overcurrent deviation are both positive values, the first load deviation and the first overcurrent deviation are normalized and added based on a preset weighting rule, and the first accessible safety factor is output.

[0101] Furthermore, the fault prediction module 14 in the low-voltage power supply and distribution system bus duct wiring fault prediction system is further configured to:

[0102] If the first load deviation and / or the first overcurrent deviation is a negative value, the first accessible safety factor is set to zero.

[0103] Furthermore, the fault prediction module 14 in the low-voltage power supply and distribution system bus duct wiring fault prediction system is further configured to:

[0104] The second historical power distribution operation data of the second alternative bus duct segment is divided according to the change node of the access plug-in box to obtain multiple power distribution operation sub-data, wherein H temperature measurement key nodes are arranged on the second alternative bus duct segment; the first power distribution operation sub-data is extracted from the multiple power distribution operation sub-data, wherein the first power distribution operation sub-data includes current-related time series features, W device load time series features of W connected devices, and H node temperature time series features of H temperature measurement key nodes; a random forest regression algorithm is adopted, with load-current as input feature and temperature data as label, and the current-related time series features, W device load time series features, and H node temperature time series features as training data to train and obtain a first sub-temperature prediction model; by analogy, after constructing multiple sub-temperature prediction models based on the multiple power distribution operation sub-data, the multiple sub-temperature prediction models are feature-fused by introducing an attention mechanism to output a second segmented twin model; by analogy, the K segmented twin models are constructed based on the K historical power distribution operation data.

[0105] Furthermore, the fault prediction module 14 in the low-voltage power supply and distribution system bus duct wiring fault prediction system is further configured to:

[0106] Extract new load limits and new overshoot currents from the load characteristics of the new equipment; after performing power supply load disturbance on the K segmented twin models, input the new load limits and new overshoot currents into the K segmented twin models to predict the key node temperatures, and obtain K node predicted temperature time series arrays; use a preset temperature threshold to perform overtemperature frequency statistics on the K node predicted temperature time series arrays, and obtain K first-dimensional safety factors; in the process of inputting the new load limits and new overshoot currents into the K segmented twin models to predict the key node temperatures, perform real-time load statistics to obtain K segmented load time series data; based on the load limit, perform overload frequency statistics on the K segmented load time series data to obtain K second-dimensional safety factors; normalize the K first-dimensional safety factors and the K second-dimensional safety factors, and output the K segmented safety factors.

[0107] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. Figure 1 The low-voltage power supply and distribution system bus duct wiring fault prediction method and specific examples in Example 1 are also applicable to the low-voltage power supply and distribution system bus duct wiring fault prediction system in this embodiment. Through the above detailed description of the low-voltage power supply and distribution system bus duct wiring fault prediction method, technical personnel in this field can clearly understand the low-voltage power supply and distribution system bus duct wiring fault prediction system in this embodiment, so for the sake of brevity of the specification, it will not be described in detail here.

[0108] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not limited to the embodiments shown herein, but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

[0109] Obviously, for those skilled in the art, several improvements and modifications can be made to the present application without departing from the principles of the present application, and these improvements and modifications also fall within the scope of protection of the present application.

Claims

1. A method for predicting bus duct wiring faults in a low-voltage power supply and distribution system, characterized in that: include: Construct bus duct topology structure based on bus duct wiring information in low voltage power distribution system; receiving a new device requirement, wherein the new device requirement includes new device location information and new device load characteristics; Performing a card-in node analysis on the bus duct topology structure according to the location information of the newly added equipment, and locating a plurality of candidate bus duct segments, wherein the candidate bus duct segments have position identifiers for card-in candidate nodes; After obtaining multiple segmented twin models based on the historical operation digital twin modeling of the multiple candidate bus duct segments, the load characteristics of the newly added equipment are used to drive the multiple segmented twin models to predict bus duct wiring faults and output multiple segmented safety factors; Locating a target bus duct segment in the multiple candidate bus duct segments according to the multiple segment safety factors; In the low-voltage power distribution system, the plug-in candidate node of the target bus duct segment is used as the target plug-in node to perform the bus duct plug-in construction of the newly added plug-in box.

2. The method for predicting bus duct wiring faults in a low-voltage power supply and distribution system according to claim 1, wherein: After obtaining multiple segmented twin models based on the historical operation digital twin modeling of the multiple candidate bus duct segments, the load characteristics of the newly added equipment are used to drive the multiple segmented twin models to predict bus duct wiring faults and output multiple segmented safety factors, including: Retrieving multiple safety plug-in association information of the multiple candidate bus duct segments from the design information of the low-voltage power supply and distribution system; Mapping and quantifying the accessibility of the plug-in boxes of the multiple candidate bus duct sections according to the multiple safety plug-in association information and the multiple groups of equipment electrical load characteristics of the multiple candidate bus duct sections to obtain multiple accessibility safety factors; Using the accessible safety threshold to traverse the multiple accessible safety factors to screen and obtain K candidate bus duct segments; According to the segment ID, call the K historical power distribution operation data of the K candidate busbar segments from the local operation log, and build K segment twin models based on the K historical power distribution operation data; The K segmented twin models are driven to run according to the load characteristics of the newly added equipment to predict bus duct wiring faults, and K segmented safety factors are quantified and output.

3. The method for predicting bus duct wiring faults in a low-voltage power supply and distribution system according to claim 2, wherein: Performing a node analysis in the bus duct topology structure based on the location information of the newly added equipment and locating multiple candidate bus duct segments includes: Calibrate the power consumption position in the low-voltage power supply and distribution system according to the location information of the newly added equipment, and perform a card-in node analysis on the bus duct topology structure according to the calibration result to obtain multiple card-in candidate nodes; The plurality of candidate bus duct segments are obtained by locating the segments to which the plurality of candidate nodes are attached in the bus duct topology structure; Performing position mapping and identification of the plurality of candidate nodes in the plurality of candidate bus duct sections; By calling local data, multiple groups of equipment electrical load characteristics of the multiple candidate bus duct segments are obtained.

4. The method for predicting bus duct wiring faults in a low-voltage power supply and distribution system according to claim 2, wherein: According to the segment ID, K historical power distribution operation data of the K candidate busbar segments are called from the local operation log, and K segment twin models are constructed based on the K historical power distribution operation data, including: Segment the second historical power distribution operation data of the second candidate bus duct segment according to the change node of the access plug-in box to obtain a plurality of power distribution operation sub-data, wherein H temperature measurement key nodes are arranged on the second candidate bus duct segment; Extracting first power distribution operation sub-data from the plurality of power distribution operation sub-data, wherein the first power distribution operation sub-data includes current-related time series characteristics, W device load time series characteristics of W connected devices, and H node temperature time series characteristics of H key temperature measurement nodes; A random forest regression algorithm is used, with load-current as the input feature and temperature data as the label. The current-related time series features, W device load time series features, and H node temperature time series features are used as training data to train and obtain the first sub-temperature prediction model. Similarly, after constructing multiple sub-temperature prediction models based on the multiple power distribution operation sub-data, the multiple sub-temperature prediction models are subjected to feature fusion by introducing an attention mechanism to output a second segmented twin model; Similarly, the K segmented twin models are constructed based on the K historical power distribution operation data.

5. The method for predicting bus duct wiring faults in a low-voltage power supply and distribution system according to claim 4, wherein: The K segmented twin models are driven by the load characteristics of the newly added equipment to predict bus duct wiring faults, and K segmented safety factors are quantitatively output, including: Extracting a new load limit and a new overshoot current from the load characteristics of the new equipment; After performing a power supply load disturbance on the K segmented twin models, the newly added load limit and the newly added overshoot current are input into the K segmented twin models to predict the key node temperatures, thereby obtaining a K node predicted temperature time series array; Using a preset temperature threshold, performing over-temperature frequency statistics on the K node predicted temperature time series arrays to obtain K first-dimensional safety factors; In the process of inputting the newly added load limit and the newly added overshoot current into the K segmented twin models to predict the key node temperature, real-time load statistics are performed to obtain K segmented load time series data; According to the load limit, overload frequency statistics are performed on the K segmented load time series data to obtain K second-dimensional safety factors; Normalize the K first-dimensional safety coefficients and the K second-dimensional safety coefficients, and output the K segmented safety coefficients.

6. The method for predicting bus duct wiring faults in a low-voltage power supply and distribution system according to claim 3, wherein: According to the plurality of safety plug-in association information and the plurality of groups of electrical load characteristics of the equipment in the plurality of candidate bus duct sections, mapping is performed to quantify the accessibility of the plug-in boxes in the plurality of candidate bus duct sections, to obtain a plurality of accessibility safety factors, including: Extracting the electrical load characteristics of the first group of devices to obtain N plug-in coordinates, N device load limits, and N device overshoot currents of N connected devices; Extracting the first plug box capacity limit, the first load limit and the first overcurrent capability from the first safety plug association information; Extracting a new load limit and a new overshoot current from the load characteristics of the new equipment; Calculate the capacity of the plug-in box according to the first plug-in candidate node of the first candidate bus duct segment and N plug-in coordinates to obtain a real-time plug-in capacity feature; If the real-time plug-in capacity characteristics meet the first plug-in box capacity limit, the plug-in box accessibility is quantified based on the N device load limits, N device overshoot currents, the newly added load limits, the newly added overshoot currents, the first load limit and the first overcurrent capacity, and a first accessibility safety factor is output.

7. The method for predicting bus duct wiring faults in a low-voltage power supply and distribution system according to claim 6, wherein: The accessibility of the plug-in box is quantified based on the N device load limits, the N device overshoot currents, the newly added load limit, the newly added overshoot current, the first load limit, and the first overcurrent capacity, and a first accessibility safety factor is output, including: After adding the N device load limits and the newly added load limit, calculating a load deviation between the sum and the first load limit to obtain a first load deviation; After adding the overshoot currents of the N devices and the newly added overshoot current, calculating a current deviation between the summed result and the first overcurrent capability to obtain a first overcurrent deviation; If both the first load deviation and the first overcurrent deviation are positive values, the first load deviation and the first overcurrent deviation are normalized and added based on a preset weighting rule to output the first accessible safety factor.

8. The method for predicting bus duct wiring faults in a low-voltage power supply and distribution system according to claim 7, wherein: If the first load deviation and / or the first overcurrent deviation is a negative value, the first accessible safety factor is set to zero.

9. The method for predicting bus duct wiring faults in a low-voltage power supply and distribution system according to claim 6, wherein: The plug-in box capacity is calculated based on the first plug-in candidate node of the first candidate bus duct segment and the N plug-in coordinates to obtain real-time plug-in capacity characteristics, including: Taking the first candidate bus duct segment as a spatial constraint, performing plug-in box density calculation according to the first plug-in candidate node and N plug-in coordinates to obtain a real-time plug-in density feature; Using the first candidate bus duct segment as a spatial constraint, solving the plug-in box spacing based on the first plug-in candidate node and N plug-in coordinates, and outputting a real-time minimum plug-in spacing feature, wherein the real-time plug-in density feature and the real-time minimum plug-in spacing feature constitute the real-time plug-in capacity feature; If the mapping of the real-time plug density feature and the real-time minimum plug spacing feature satisfies the first plug density restriction and the first plug spacing restriction in the first plug box capacity restriction, it is determined that the real-time plug capacity feature satisfies the first plug box capacity restriction.

10. A low-voltage power supply and distribution system bus duct wiring fault prediction system, characterized in that: Steps for implementing the method for predicting bus duct wiring faults in a low-voltage power supply and distribution system according to any one of claims 1 to 9, the low-voltage power supply and distribution system bus duct wiring fault prediction system comprising: A topology structure building module is used to build a bus duct topology structure according to bus duct wiring information in a low-voltage power distribution system; An equipment demand receiving module, configured to receive new equipment demand, wherein the new equipment demand includes new equipment location information and new equipment load characteristics; A card-in node analysis module is used to perform a card-in node analysis in the bus duct topology structure according to the location information of the newly added equipment, and locate multiple candidate bus duct segments, wherein the candidate bus duct segments have position identifiers of card-in candidate nodes; a fault prediction module, configured to, after obtaining multiple segmented twin models based on the historical operation digital twin modeling of the multiple candidate bus duct segments, use the load characteristics of the newly added equipment to drive the multiple segmented twin models to predict bus duct wiring faults and output multiple segmented safety factors; A segment positioning module is used to locate a target bus duct segment in the multiple candidate bus duct segments according to the multiple segment safety factors; The plug-in node determination module is used to, in the low-voltage power distribution system, use the plug-in candidate node of the target bus duct segment as the target plug-in node to perform the bus duct plug-in construction of the newly added plug-in box.

Images