A main circuit physical cut-off control method of a distribution box under extreme weather

Abstract

The application provides a kind of main circuit physical cut-off control method of distribution box under extreme weather, comprising: obtaining three-dimensional space coordinate sequence collected by built-in inertia detection unit in distribution box, extracting gravity direction and the spatial posture of box bottom plate after smoothing processing;Analysis of the out-of-bound results of the included angle of the gravity direction to the set tilt threshold, obtain the tilt recognition result of the distribution box bottom plate, and extract the local minimum point of the water inlet edge, low edge and drain hole position of the rainwater flow along the bottom plate;Obtain the distribution coordinates of the preset transverse detection electrode in the physical space, convert it to the coordinate system with the gravity direction as the reference, calculate the Euclidean distance between each electrode and the low-lying water collection point;Evaluate the out-of-bound results of the spatial misplacement distance to the set distance threshold, and extract the overlap area of the transverse electrode spacing and the water collection path, obtain the risk of delay conduction of the dense electrode avoiding the low-lying water collection point.

Description

Technical Field

[0001] This invention relates to the field of information technology, and in particular to a method for physically disconnecting the main circuit of a distribution box under extreme weather conditions. Background Technology

[0002] The safety of distribution boxes under extreme weather conditions is vital to ensuring the stable operation of urban power networks. In scenarios such as heavy rain and flooding, timely physical disconnection of the main circuit is a crucial line of defense against leakage accidents and to protect personal safety. Current disconnection control mechanisms largely rely on fixed detector layouts and absolute water level standards. This approach often overlooks the spatial attitude shifts of the distribution box caused by external environmental factors such as foundation settlement after actual installation or long-term operation. This leads to a disconnect between the preset detection benchmark and the actual physical evolution of floodwater, making it difficult to respond accurately in the early stages of flooding. Conventionally, densely arranging detector electrodes at the bottom of the box increases the probability of contact between the probes and the water surface, thereby quickly establishing a continuous conductive path at extremely low water levels and triggering disconnection as early as possible. However, this dense layout does not consider the actual physical tilt of the box's bottom plate. When the bottom plate tilts, the floodwater entering the box will flow rapidly towards specific low-lying catchment points along the tilt direction. If the originally dense electrode array happens to miss this actual catchment path, floodwater will accumulate in large quantities in the low-lying areas where there are no electrodes. In the prior art, CN120968072A discloses a drainage method and device based on dynamic water level sensing and hierarchical linkage control. It uses multiple sets of conductive electrode pairs arranged at height intervals to form a water level sensing unit. Based on a preset water level threshold, it triggers the manhole cover to flip, the fence to rise and fall, and the blade to start and stop. The solution focuses on increasing the number of electrodes or setting the trigger threshold according to the absolute height. It mainly focuses on the response action when the water level reaches a certain elevation. It does not consider the impact of changes in the spatial posture of the base plate of the distribution box due to factors such as foundation settlement and external impact during actual installation or operation on the flow path and accumulation position of water. This results in the detection logic being disconnected from the actual physical process of water immersion. For example, in the event of a typhoon or severe convective weather-induced backflow of rainwater, the distribution box may tilt to the left and rear due to loose foundation, while the densely arranged detection electrodes remain concentrated in the front or right area. At this time, a large amount of rainwater will rush into the box under the combined action of strong wind and gravity, and flow along the tilted bottom plate to the low-lying area to the left and rear. Even if enough water has accumulated in the box to cause a short circuit, the water cannot contact the electrodes to form a conductive circuit, resulting in a serious delay in the physical disconnection of the main circuit. The equipment continues to operate energized in the high-risk water immersion environment under extreme weather conditions. Summary of the Invention

[0003] This invention provides a method for physically disconnecting the main circuit of a distribution box under extreme weather conditions, mainly including: The three-dimensional spatial coordinate sequence of the distribution box is obtained, and after smoothing, the gravity direction and the spatial attitude of the box bottom plate are extracted. When the angle between the gravity vector and the normal vector of the base plate in the spatial orientation of the distribution box exceeds the set tilt threshold, the tilt recognition result of the base plate of the distribution box is obtained, and the local lowest points of the water inlet side, low side and drainage hole position of the rainwater along the flow direction of the base plate are extracted. By combining the three-dimensional topology of the bottom of the distribution box, the water collection path formed by rainwater along the reinforcing ribs, cable holes, drainage holes and bottom edge of the box is identified, and the local lowest point where the longest residence time on the path is defined as the low-lying water collection point. Obtain the preset distribution coordinates of the lateral detection electrodes, transform them to a coordinate system based on the direction of gravity, and calculate the Euclidean distance between each electrode and the low-lying water collection point. When the distance between all electrodes and the low-lying water catchment point is greater than the set threshold, it is determined that the current electrode layout cannot cover the actual water catchment area, and the capacitance value of the low-lying water catchment point is continuously collected. The results of spatial misalignment distance exceeding the set distance threshold are evaluated, and the overlapping area between the lateral electrode spacing and the water collection path is extracted to obtain the delayed conduction risk of dense electrodes avoiding the low-lying water collection point. When the rate of change of the capacitance value exceeds a set threshold, a virtual conduction signal is generated. The virtual conduction signal is injected into the main circuit cutoff control logic to form a closed conduction judgment circuit. The consistency between the logic state of the circuit and the preset physical cutoff trigger condition is evaluated. If they are consistent, a physical cutoff control command is sent to the main control circuit.

[0004] Furthermore, the step of obtaining the three-dimensional spatial coordinate sequence of the distribution box, smoothing it, and then extracting the gravity direction and the spatial attitude of the box's bottom plate includes: A sequence of coordinate points composed of triaxial acceleration and triaxial angular velocity is obtained; the coordinate point sequence is timestamped according to the sampling time, and data points with abnormal sampling intervals and range overflow are removed; the coordinate point sequence is smoothed by Kalman filtering algorithm, and the gravity field projection in the box coordinate system is separated from the triaxial acceleration components to obtain the gravity vector. The orientation of the bottom plate is represented by the bottom plate normal vector, and the spatial attitude of the bottom plate is obtained by the angle relationship between the gravity vector and the normal vector.

[0005] Furthermore, when the angle between the gravity vector and the normal vector of the base plate in the spatial orientation of the distribution box exceeds a set tilt threshold, the tilt recognition result of the base plate is obtained, and the local lowest points of the inlet edge, lower edge, and drain hole positions of the rainwater flowing along the base plate are extracted, including: When the included angle is greater than the set tilt threshold, it is determined that the base plate has tilted beyond the boundary and the tilt direction of the over-boundary tilt is extracted to obtain the tilt identification result of the distribution box base plate; the horizontal projection component of the gravity vector on the base plate plane is used as the rainwater flow direction vector, and the water inlet edge is obtained by extending the flow direction vector in the reverse direction to the edge of the base plate, and the low edge of the base plate is obtained by extending it in the forward direction to the edge of the base plate. The location of the drainage hole is determined by combining the physical coordinates of the drainage hole in the base plate plane diagram; the relative elevation value is compared point by point in the neighborhood of each sampling point, and the location where the relative elevation value is less than other sampling points in the neighborhood is extracted to obtain the local minimum point.

[0006] Furthermore, the method combines the three-dimensional topological structure of the bottom of the distribution box to identify the water collection path formed by rainwater along the reinforcing ribs, cable holes, drainage holes, and the bottom edge of the box. The local lowest point on the path where the water stays for the longest time is defined as the low-lying water collection point, including: Obtain three-dimensional topological data of the bottom of the distribution box. The three-dimensional topological data includes the geometric position coordinates and height fluctuation information of the reinforcing ribs, cable holes, drainage holes, and the direction of the bottom edge of the box. Discrete sampling is performed on the bottom edge of the box with equal arc length. Together with the reinforcing rib, cable hole and drainage hole, they are used as topological nodes. The physical connectivity between adjacent nodes is used as topological edges to construct a topological node connectivity graph covering the bottom plate surface. Starting from the local lowest point on the water inlet edge, the path is gradually advanced along the topological edge by searching for adjacent nodes in the direction of elevation descent until the local lowest point at the drainage hole is reached, thus obtaining the node sequence of the water collection path. For all local minimum points on the water catchment path node sequence, the local minimum point with the longest retention time is extracted to obtain the low-lying water catchment point.

[0007] Furthermore, after obtaining the sequence of water catchment path nodes, it includes: For each topological node in the water catchment path node sequence, the node water flow rate is obtained based on the ratio of the elevation difference between adjacent nodes to the topological distance between nodes. The node residence time is obtained by dividing the volume of the depression area where the node is located in the three-dimensional topological structure data by the node water flow rate.

[0008] Furthermore, the step of obtaining the preset distribution coordinates of the lateral detection electrodes, converting them to a coordinate system based on the direction of gravity, and calculating the Euclidean distance between each electrode and the low-lying water collection point includes: Obtain the distribution coordinates of the lateral detection electrodes with reference to the original installation coordinate system of the box; construct a rotation matrix based on the attitude angle of the distribution box base plate, and perform coordinate transformation on the distribution coordinates to obtain the electrode coordinates in the gravity coordinate system with reference to the direction of gravity; for the electrode coordinates in the gravity coordinate system and the low-lying water collection point coordinates in the same gravity coordinate system, calculate the distance between each electrode and the low-lying water collection point one by one using the Euclidean distance formula.

[0009] Furthermore, when the distance between all electrodes and the low-lying catchment point is greater than a set threshold, it is determined that the current electrode layout cannot cover the actual catchment area, and the capacitance value of the low-lying catchment point is continuously collected, including: Obtain the set of Euclidean distances between each electrode and the low-lying water catchment point and set a distance threshold; when all distance values ​​in the set of Euclidean distances are greater than the distance threshold, it is determined that the physical layout of the current lateral detection electrode does not fall within the actual water catchment coverage area of ​​the low-lying water catchment point, and the distribution box is calibrated to be in a state of trigger delay risk; retrieve the corresponding capacitance sampling channel from the capacitance probes deployed at the low-lying water catchment point, and stitch the capacitance readings at each sampling time along the time axis to obtain the capacitance value time series of the low-lying water catchment point.

[0010] Furthermore, the evaluation of spatial misalignment distance exceeding a set distance threshold, extracting the overlapping area between the lateral electrode spacing and the water catchment path, and obtaining the delayed conduction risk of dense electrodes avoiding the low-lying water catchment point, includes: The minimum value in the set of Euclidean distances between each electrode and the low-lying water catchment point is taken as the spatial misalignment distance between the electrode array and the low-lying water catchment point. The difference between the spatial misalignment distance and the set distance threshold is taken as the degree of boundary violation. The line connecting adjacent electrodes in the electrode coordinates under the gravity coordinate system is taken as the covering line segment, and the line connecting adjacent topological nodes in the water catchment path node sequence is taken as the water catchment path segment. The part where the covering line segment and the water catchment path segment intersect geometrically is extracted to obtain the overlapping area of ​​the lateral electrode spacing and the water catchment path. When the degree of boundary violation is greater than the preset threshold and the area of ​​the overlapping area is less than the preset area threshold, it is determined that the dense electrode array deviates from the low-lying water catchment point.

[0011] Furthermore, when the rate of change of the capacitance value exceeds a set threshold, a virtual conduction signal is generated. This virtual conduction signal is injected into the main circuit cutoff control logic, forming a closed conduction judgment loop. The consistency between the loop's logic state and the preset physical cutoff trigger condition is evaluated. If they are consistent, a physical cutoff control command is sent to the main control loop, including: For the capacitance value time series of the low-lying water catchment point, the difference between the capacitance readings at adjacent sampling times is divided by the sampling period to obtain the rate of change of the capacitance value. When the rate of change of the capacitance value exceeds a set threshold, it is determined that water has accumulated at the low-lying water catchment point. This accumulation event is output as a virtual conduction signal in the form of a conduction level signal. The virtual conduction signal is connected to the main circuit cutoff control logic through the logic injection port and performs an OR operation with the physical conduction signal output by the lateral detection electrode. The disconnected conduction judgment link is filled to form a closed conduction judgment loop. The logic state of the closed conduction judgment loop is obtained and compared with the preset physical cutoff trigger condition. When the comparison result is consistent, a physical cutoff control command is sent to the main control loop.

[0012] Furthermore, the preset physical cutoff trigger condition is the trigger condition of the corresponding conduction level in the main circuit cutoff control logic.

[0013] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects: This invention discloses a method for physical disconnection control of the main circuit of a distribution box under extreme weather conditions. Addressing the problem that rainwater accumulates in low-lying corners due to tilting or structural defects at the bottom of the distribution box, and that the lateral detection electrodes cannot detect water accumulation in time due to spatial misalignment, thus causing delayed triggering of leakage protection, this invention uses an inertial detection unit combined with a Kalman filter algorithm to accurately identify the box's tilt posture and gravity direction. Then, by combining the bottom three-dimensional topological structure analysis, it identifies the actual water collection path formed by reinforcing ribs, cable holes, drainage holes, and edges, pinpointing the low-lying water collection point with the longest residence time. By calculating the Euclidean distance between the preset lateral electrode and this water collection point, it identifies coverage blind spots and trigger delay risks in the electrode layout. When spatial misalignment of the electrode is detected, resulting in ineffective coverage of the water collection area, the capacitance change rate of the low-lying water collection point is continuously monitored. Once it exceeds a set threshold, water accumulation is confirmed, and a virtual conduction signal is immediately generated and injected into the main circuit disconnection control logic, forming a closed judgment loop and triggering a physical disconnection command. This method effectively compensates for the spatial blind spots of traditional electrode layouts, achieving proactive safety protection for the distribution box under extreme weather conditions. Attached Figure Description

[0014] Figure 1 This is a flowchart of a method for physically cutting off the main circuit of a distribution box under extreme weather conditions, according to the present invention.

[0015] Figure 2 This is a schematic diagram of a physical disconnection control method for the main circuit of a distribution box under extreme weather conditions, according to the present invention.

[0016] Figure 3 This is another schematic diagram of a method for physically cutting off the main circuit of a distribution box under extreme weather conditions, according to the present invention. Detailed Implementation

[0017] To further understand the content of this invention, a detailed description of the invention is provided in conjunction with the accompanying drawings and embodiments. The specific embodiments described herein are for illustrative purposes only and are not intended to limit the invention. It should also be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.

[0018] like Figures 1-3 This embodiment of a method for physically disconnecting the main circuit of a distribution box under extreme weather conditions may specifically include: Step S101: Obtain the three-dimensional spatial coordinate sequence collected by the inertial detection unit built into the distribution box, and extract the gravity direction and spatial attitude of the box bottom plate after smoothing.

[0019] A sequence of coordinate points consisting of triaxial acceleration and triaxial angular velocity is obtained from the inertial detection unit built into the distribution box. The inertial detection unit is fixedly installed on the inner wall of the box's bottom plate. The acceleration reflects the instantaneous acceleration component of the box relative to the gravitational field, and the angular velocity reflects the rotational rate of the box around its own coordinate axes. The coordinate point sequence is timestamped according to the sampling time, and data points with abnormal sampling intervals and range overflows are removed to obtain the time-aligned original attitude observation sequence. A Kalman filter algorithm is used to smooth the environmental vibration and electromagnetic interference noise present in the original attitude observation sequence. Using triaxial acceleration and triaxial angular velocity as state variables and the measured values ​​of the inertial detection unit as observation variables, random noise components are suppressed through recursive prediction and observation updates of the state variables to obtain a smoothed attitude observation sequence. Based on the smoothed attitude observation sequence, the projection of the gravitational field in the box coordinate system is separated from the triaxial acceleration components to obtain the gravity vector. The orientation of the base plate is represented by the normal vector of the plane containing the base plate, and the direction of the normal vector is updated by combining the integral of the angular velocity over time. The attitude angle of the base plate plane relative to the horizontal plane is determined by the angle between the gravity vector and the normal vector, thus obtaining the spatial attitude of the distribution box base plate.

[0020] In one embodiment, the distribution box is an outdoor pole-mounted or floor-mounted low-voltage distribution box. The inertial detection unit is fixedly installed at the geometric center of the inner wall of the box's bottom plate using screws and vibration damping pads. The X and Y axes of its body coordinate system are aligned with the length and width directions of the bottom plate, and the Z axis points perpendicularly to the normal direction of the bottom plate. The inertial detection unit continuously outputs triaxial acceleration and triaxial angular velocity at a certain sampling frequency, forming a coordinate point sequence that changes over time.

[0021] Specifically, the acceleration reflects the instantaneous acceleration component of the enclosure relative to the gravitational field, including the gravitational projection component when stationary and the linear acceleration component when in motion; the angular velocity reflects the rotational rate of the enclosure about its own coordinate axis. During long-term outdoor operation of the distribution box, foundation settlement and external impacts will cause the enclosure's attitude to drift slowly, and this drift will be reflected in the above two types of data.

[0022] It should be noted that the acquired coordinate point sequence exhibits two types of anomalies: firstly, sampling interval anomalies, manifested as timestamp jumps between adjacent data points exceeding the sampling period; secondly, range overflow, manifested as acceleration or angular velocity values ​​reaching the sensor's upper or lower limits. The coordinate point sequence is aligned in ascending order by timestamp, and the aforementioned abnormal data points are removed to obtain the time-aligned original attitude observation sequence.

[0023] In one possible implementation, the state variables of the Kalman filter algorithm are selected as a 6-dimensional vector consisting of triaxial acceleration and triaxial angular velocity, denoted as x; the observations are selected as the 6-dimensional measured values ​​output by the inertial detection unit at the current sampling time, denoted as z. The Kalman filter algorithm operates in a two-step alternating manner.

[0024] Specifically, the recursive prediction stage predicts the prior value of the state quantity at the current moment based on the estimated state quantity value at the previous moment and the state transition relationship. The observation update stage weights and fuses the prior value with the measured observation z. The weighting coefficient is determined by the relative magnitude of the prediction error and the observation noise. The result of the weighted fusion is the posterior estimate of the state quantity at the current moment. The posterior estimates are spliced ​​together in the time dimension to form a smoothed attitude observation sequence. For example, for the three-axis acceleration components in the smoothed attitude observation sequence, when the distribution box is stationary or nearly stationary, the resultant vector of the three-axis components is approximately equal to the gravitational acceleration g, and its direction is the projection direction of the gravitational field in the box coordinate system. Preferably, a sliding time window is used to perform mean statistics on the three-axis acceleration to filter out residual weak vibration components and obtain a stable gravity vector, whose direction cosine is denoted as (gx, gy, gz), where gx, gy, and gz are the components of the gravity vector on the X, Y, and Z axes of the box coordinate system, respectively.

[0025] It is understandable that the plane containing the bottom plate of the box remains constant in the box coordinate system, and its normal vector initially points along the positive Z-axis. The cumulative rotation of the box around each axis is obtained by integrating the angular velocity over time, and this is used to synchronously update the direction of the normal vector. The updated normal vector is denoted as (nx, ny, nz). Since both the gravity vector and the normal vector have been normalized to unit vectors, their magnitude is 1. Furthermore, the attitude angle of the bottom plate is determined by the angle between the gravity vector and the normal vector, where the cosine of the angle is equal to... When the included angle is close to 0 degrees, it indicates that the normal direction of the base plate is consistent with the direction of gravity, and the base plate is in a horizontal state; when the included angle deviates from 0 degrees, it indicates that the base plate has tilted at the corresponding angle. The angle value corresponding to the included angle is the attitude angle of the distribution box base plate relative to the horizontal plane. This attitude angle, together with the gravity vector and the normal vector, constitutes the spatial attitude description of the distribution box base plate.

[0026] Step S102: Analyze the result of the gravity direction angle exceeding the set tilt threshold, obtain the tilt identification result of the distribution box bottom plate, and extract the local lowest point of the water inlet side, low side and drainage hole position of the rainwater along the bottom plate flow direction.

[0027] The attitude angle between the gravity vector and the base plate normal vector in the spatial attitude of the distribution box base plate is obtained, and a tilt threshold is set. If the attitude angle is greater than the tilt threshold, it is determined that the base plate has tilted beyond the limit, and the corresponding tilt direction is extracted to obtain the tilt recognition result of the distribution box base plate; if the attitude angle is less than or equal to the tilt threshold, it is determined that the base plate is in a horizontal state. According to the tilt direction beyond the limit, the horizontal projection component of the gravity vector on the base plate plane is taken as the rainwater flow direction vector. Extending along the reverse direction of the flow direction vector to the edge of the base plate, the water inlet edge of the box is obtained; extending along the positive direction of the flow direction vector to the edge of the base plate, the bottom edge of the base plate is obtained. Combining the pre-stored physical coordinates of the drainage holes in the base plate plan view, the location of the drainage holes is determined. For each sampling point at the water inlet edge, the bottom edge of the base plate, and the location of the drainage holes, the vertical height of the base plate plane along the direction of the flow direction vector is taken as the relative elevation value, with the smaller value indicating a lower position. The relative elevation values ​​are compared point by point within the neighborhood of each sampling point. The locations where the relative elevation value is less than other sampling points in the neighborhood are extracted to obtain the local lowest points of the water inlet side, low side and drainage hole locations along the direction of rainwater flow along the bottom plate.

[0028] In one embodiment, the base plate of the distribution box is exposed to the outdoors for extended periods. Loosening of the foundation, ground settlement, or external impacts can all cause gradual changes in the base plate's attitude. The attitude angle, the angle between the gravity vector and the base plate's normal vector, approaches 0 degrees when the base plate is horizontal and increases with the degree of tilt when the base plate is tilted. The preset tilt threshold typically ranges from 2 to 5 degrees, with the specific value depending on the size of the distribution box base plate and its sensitivity to water accumulation.

[0029] It should be noted that when the attitude angle is less than or equal to the tilt threshold, the bottom plate is in a horizontal state, and the rainwater is evenly distributed after entering the box, without needing to enter the subsequent water collection path identification process; when the attitude angle is greater than the tilt threshold, the tilt identification process is triggered.

[0030] Specifically, the process of extracting the over-boundary tilt direction is as follows: the gravity vector is projected onto the base plate plane to obtain a two-dimensional direction vector. The direction pointed to by this direction vector in the base plate plane coordinate system is the over-boundary tilt direction. For example, if the direction vector points to the left rear quadrant of the base plate coordinate system, it indicates that the distribution box has tilted to the left rear, which corresponds to the scenario in the background business where a typhoon causes the foundation to loosen.

[0031] In one possible implementation, the horizontal projection component of the gravity vector onto the base plane is directly used as the rainwater flow direction vector. Its physical meaning is that after rainwater enters the tank, it slides down the inclined direction of the base plate under the action of gravity; the sliding direction is the direction of the tangential component of gravity along the base plate plane. Let the gravity vector be g, the base plate normal vector be n, and the direction of the flow direction vector f be calculated as follows: ,in The scalar represents the projection of the gravity vector onto the normal vector. The component of the gravity vector along the normal vector is represented by g. Subtracting this component from g gives the tangential component in the plane of the base plate. Further, the boundary where the flow vector extends in the opposite direction to the edge of the base plate and intersects with the edge of the base plate frame is the water inlet edge. This water inlet edge is the starting edge where rainwater seeps in from the top cover of the tank and flows along the base plate. The boundary where the flow vector extends in the positive direction to the edge of the base plate and intersects with the edge of the base plate frame is the bottom edge of the base plate. This bottom edge is the physical boundary where rainwater finally collects inside the base plate.

[0032] Understandably, the location of the drain holes on the distribution box base plate is fixed in the base plate plan during the factory manufacturing stage. With the geometric center of the base plate as the origin, the physical coordinates of the drain holes are pre-stored in the local storage module as (x, y) two-dimensional coordinate values. In the tilt recognition process, the physical coordinates of the drain holes are directly read as the location of the drain holes.

[0033] For example, the relative elevation value is obtained as follows: multiple sampling points are taken at the water inlet, the lower edge, and the drainage hole. The vertical height of the bottom plate plane at the sampling point along the opposite direction of the flow vector is taken as the relative elevation value of that point. The smaller the value, the lower the position of the point on the inclined bottom plate, and the easier it is for water to accumulate. The relative elevation value is positively correlated with the magnitude of the attitude angle, that is, the more severe the inclination, the greater the elevation difference between the sampling points. Preferably, a neighborhood range with a preset radius is defined on the bottom plate plane with each sampling point as the center, and the relative elevation values ​​of each sampling point are compared point by point within this neighborhood range. If the relative elevation value of a certain sampling point is less than that of all other sampling points within the neighborhood range, then the sampling point is marked as the local minimum point; if there are multiple parallel minimum values ​​within the neighborhood range, then the sampling point closest to the drainage hole is taken as the local minimum point. The extraction result of the local minimum point is recorded in the form of a set of coordinate points at the water inlet, the lower edge, and the drainage hole.

[0034] Step S103: Based on the three-dimensional topology of the bottom of the distribution box, identify the water collection path formed by rainwater along the reinforcing ribs, cable holes, drainage holes and bottom edges of the box, and define the local lowest point on the path where the longest residence time is defined as the low-lying water collection point.

[0035] The three-dimensional topological structure data of the bottom of the distribution box is obtained. This data is derived from the box design drawings or 3D scanning reconstruction during the manufacturing stage and includes the geometric coordinates and height undulation information of the reinforcing ribs, cable holes, drainage holes, and the direction of the bottom edges. The bottom edges are discretely sampled with equal arc lengths and used together with the reinforcing ribs, cable holes, and drainage holes as topological nodes. The physical connectivity between adjacent nodes is used as topological edges to construct a topological node connectivity graph covering the bottom surface. Based on the topological node connectivity graph, the local lowest point on the water inlet edge is used as the starting point of the path. The path searches for adjacent nodes in the direction of elevation decrease along the topological edges and gradually advances until the local lowest point at the drainage hole is reached, thus obtaining the water collection path node sequence formed by rainwater along the reinforcing ribs, cable holes, drainage holes, and bottom edges. For each topological node in the water catchment path node sequence, the node flow rate is obtained based on the ratio of the elevation difference between adjacent nodes to the topological distance between nodes. The node residence time is obtained by dividing the volume of the depression area where the node is located in the three-dimensional topological structure data by the node flow rate. For all local lowest points in the water catchment path node sequence, the local lowest point with the longest residence time is extracted to obtain the low-lying water catchment point.

[0036] In one embodiment, the three-dimensional topological structure data of the bottom of the distribution box is obtained in two ways. First, it is read from the box design drawings at the factory stage, including the cross-sectional shape and orientation coordinates of the reinforcing ribs, the diameter and center coordinates of cable holes, the diameter and center coordinates of drainage holes, and the side length and orientation coordinates of the bottom edges. Second, after the distribution box is installed, a 3D scanner is used to scan and reconstruct the bottom of the box, obtaining point cloud data of the bottom plate surface in three-dimensional space. The aforementioned geometric elements are then extracted after surface fitting. The three-dimensional topological structure data obtained in both methods includes geometric position coordinates and height undulation information. The bottom edges are physically continuous line segments and cannot be directly used as discrete nodes in the graph structure construction.

[0037] Specifically, the bottom edges of the box are discretely sampled at equal arc lengths, meaning multiple sampling points are taken along the edge at preset arc length intervals. The arc length interval is generally between 1 / 50 and 1 / 100 of the bottom plate length, thus discretizing each edge into several sampling points. These sampling points, along with the fitting points taken along the ridge of the reinforcing rib, the center points of the cable holes, and the center points of the drainage holes, together constitute a set of topological nodes. Each topological node has its physical coordinates and corresponding relative elevation value.

[0038] In one possible implementation, the determination of topological edges is based on the physical connectivity between nodes. If two topological nodes are directly adjacent to each other on the base plate plane and there are no reinforcing ridges between them, a topological edge is established between the two topological nodes. If there are reinforcing ridges between two topological nodes, rainwater will flow along one side of the reinforcing ridge and cannot cross it, so no topological edge is established between the two topological nodes. The topological node connectivity graph uses topological nodes as vertices and topological edges as connections, completely covering the base plate surface. Further, the local lowest point on the water inlet edge is used as the path starting point. When there are multiple local lowest points on the water inlet edge, the drainage path to the drainage hole is calculated from each local lowest point as the starting point. The path length L and the cumulative height difference ΔH along the path are compared, where L represents the sum of the distances between adjacent topological nodes on the path, and ΔH represents the height difference between the starting point and the ending point. The path with the smallest L is selected first. If there are multiple paths with the same L, the path with the largest ΔH is selected as the dominant drainage path, and the starting point of this path is the actual path starting point. The starting point of the path corresponds to the physical position where rainwater first contacts the bottom plate after entering the container and begins to flow along the bottom plate.

[0039] It is understandable that rainwater flows physically following the law of sliding down the elevation direction under the influence of gravity. The process of searching for adjacent nodes along the topological edge in the direction of elevation decline is as follows: At the current node, traverse all adjacent nodes connected by its topological edge, obtain the relative elevation value of each adjacent node, and compare it with the relative elevation value of the current node; if the relative elevation value of an adjacent node is less than that of the current node, then that adjacent node is a downstream candidate node; among all downstream candidate nodes, select the adjacent node with the smallest relative elevation value as the next node, and move forward to that next node to continue the search. When there are multiple drainage holes, before the search begins, calculate the ratio of the topological distance L from the current starting point to each drainage hole to the elevation difference ΔH, where L is the cumulative length along the topological edge, and ΔH is the difference between the starting point elevation and the drainage hole elevation. Select the drainage hole with the smallest L divided by ΔH value as the target endpoint of the path. The search process continues until the local lowest point at the preset target drainage hole location is reached, resulting in a sequence of water collection path nodes formed by rainwater along the reinforcing rib, the cable hole, the drainage hole, and the bottom edge of the box.

[0040] For example, the water catchment path node sequence is represented by an ordered list of topological nodes, where the elevation difference between adjacent nodes in the sequence exhibits a monotonically decreasing trend, and each node in the sequence corresponds to a rainwater passage location on the base plate. Preferably, for each topological node in the water catchment path node sequence, the node flow rate v is obtained by dividing the elevation difference Δh between adjacent nodes by the topological distance L between the two nodes, i.e., v = Δh ÷ L, where Δh represents the difference in relative elevation values ​​between two adjacent nodes, and L represents the physical distance between two adjacent nodes along the topological edge on the base plate plane. This flow rate reflects the speed at which rainwater flows from the upstream node to the current node.

[0041] In one embodiment, the volume of the recessed area where the node is located is taken from the volume of the closed area enclosed by the node as the center and extending outwards to the rising elevation boundary in the three-dimensional topology data. Dividing this volume by the node's water flow rate yields the node's residence time, which reflects the physical time rainwater spends in the recessed area after entering it. For all local lowest points in the water catchment path node sequence, the node residence time of each local lowest point is compared, and the one with the largest value is extracted as the low-lying water catchment point. For example, when a distribution box tilts to the left rear due to loose foundation, rainwater sliding down the water catchment path passes through multiple local lowest points and finally converges in a recessed area in the left rear corner. This recessed area has a large volume and a low node water flow rate, resulting in the longest node residence time, and is therefore identified as the low-lying water catchment point.

[0042] Step S104: Obtain the preset distribution coordinates of the lateral detection electrodes in physical space, transform them into the coordinate system based on the direction of gravity, and calculate the Euclidean distance between each electrode and the low-lying water collection point.

[0043] The distribution coordinates of preset lateral detection electrodes within the distribution box in physical space are obtained, with the original installation coordinate system of the box as the reference. A rotation matrix is ​​constructed based on the attitude angle of the distribution box base plate, and the distribution coordinates are transformed to a coordinate system based on the direction of gravity, yielding the electrode coordinates in the gravity coordinate system. For the electrode coordinates in the gravity coordinate system and the coordinates of low-lying water collection points in the same gravity coordinate system, the Euclidean distance formula is used to calculate the distance between each electrode and the low-lying water collection point. The Euclidean distance characterizes the spatial interval between the electrode and the low-lying water collection point in the gravity coordinate system, thus obtaining the Euclidean distance between each electrode and the low-lying water collection point.

[0044] In one embodiment, the lateral detection electrodes refer to multiple metal probes arranged horizontally along the bottom plate of the distribution box. The coordinates of each electrode are fixed in the original installation coordinate system of the box at the factory. The original installation coordinate system of the box has the geometric center of the bottom plate as the origin, the length and width of the bottom plate as the X and Y axes, and the normal direction of the bottom plate as the Z axis. The distribution coordinates of each electrode are read from local storage and denoted as... , where i is the electrode number.

[0045] Specifically, after the distribution box tilts, the original installation coordinate system of the box shifts relative to the actual direction of gravity, causing the original coordinates to no longer directly reflect the true spatial position of the electrodes under the gravitational field. A three-dimensional rotation matrix R is constructed based on the attitude angles of the distribution box's base plate. These attitude angles include the pitch angle θ around the X-axis and the roll angle φ around the Y-axis. The rotation matrix R is constructed by first rotating it by an angle θ around the X-axis and then by an angle φ around the Y-axis. The matrix elements are composed of combinations of cosθ, sinθ, cosφ, and sinφ. Specifically, the element in the first row and first column of R is cosφ, the element in the first row and third column is sinφ, the element in the second row and first column is sinθ multiplied by sinφ, the element in the second row and second column is cosθ, the element in the second row and third column is negative sinθ multiplied by cosφ, the element in the third row and first column is negative cosθ multiplied by sinφ, the element in the third row and second column is sinθ, and the element in the third row and third column is cosθ multiplied by cosφ. The distribution coordinates are then multiplied sequentially by the rotation matrix R to perform a coordinate transformation, resulting in the electrode coordinates in the gravitational coordinate system. Furthermore, the low-lying water catchment point, after the aforementioned topology node processing, is now located in a gravity coordinate system, and its coordinates are denoted as... The Euclidean distance formula is applied to each electrode. Each electrode is calculated individually, where d represents the spatial interval between electrode i and the low-lying water catchment point, to obtain the set of Euclidean distances between each electrode and the low-lying water catchment point.

[0046] Step S105: If the distance between all electrodes and the low-lying water catchment point is greater than the set threshold, it is determined that the current electrode layout cannot effectively cover the real water catchment area and there is a risk of triggering delay. The capacitance value of the low-lying water catchment point is continuously collected.

[0047] A set of Euclidean distances between each electrode and the low-lying catchment point is obtained, and a distance threshold is set. Each distance value in the set of Euclidean distances is compared with the distance threshold. If all distance values ​​in the set of Euclidean distances are greater than the distance threshold, it is determined that the physical layout of the current lateral detection electrode does not fall within the actual catchment coverage area of ​​the low-lying catchment point, and a determination result of the electrode layout coverage blind zone is obtained. Based on the determination result of the electrode layout coverage blind zone, the distribution box is marked as being in a state of trigger delay risk. From the capacitance probes pre-deployed on the bottom plate of the distribution box at the location of the low-lying catchment point, the capacitance sampling channel corresponding to the spatial location of the low-lying catchment point is retrieved, and a capacitance sampling link pointing to the low-lying catchment point is established. The capacitance sampling link is used to continuously collect the dielectric constant change at the location of the low-lying catchment point, and the capacitance readings at each sampling time are sequentially spliced ​​along the time axis to form a time series of capacitance values ​​at the location of the low-lying catchment point, thus obtaining the capacitance value of the low-lying catchment point.

[0048] In one embodiment, the distance threshold refers to the maximum spatial radius in which the lateral detection electrode can effectively contact and capture accumulated water. Its value is determined by the smaller of the physical sensing radius of a single electrode and the diffusion radius of rainwater on the base plate surface. Preferably, the distance threshold ranges from 1 / 10 to 1 / 8 of the diagonal length of the base plate. In outdoor pole-mounted or floor-mounted low-voltage distribution box scenarios, this threshold can be stored locally and used as a preset threshold for judgment.

[0049] Specifically, each distance value in the Euclidean distance set is compared with the distance threshold one by one. If the distance value of a certain electrode is less than or equal to the distance threshold, it indicates that the physical arrangement of the electrode has covered the actual catchment area where the low-lying water catchment point is located; if all distance values ​​in the Euclidean distance set are greater than the distance threshold, it indicates that all electrodes fall outside the sensing range of the low-lying water catchment point.

[0050] It should be noted that the physical meaning of the determination result of the electrode layout covering the blind zone is as follows: after rainwater collects along the inclined bottom plate to the low-lying water collection point, even if the water level has reached the water volume that normally triggers the cut-off, all the transverse detection electrodes still cannot contact the water surface at that point and cannot form a conductive loop between the electrodes, thus causing a trigger delay.

[0051] It is understandable that the trigger delay risk state is marked in the local status register in the form of binary flag bits. The flag bit being set to 1 indicates that the current distribution box has entered the trigger delay risk state, which serves as a prerequisite for the subsequent activation of the capacitor sampling link.

[0052] In one possible implementation, the capacitive probe and the lateral detection electrode are two independent sensor devices. The capacitive probe has a higher distribution density than the lateral detection electrode, and its physical arrangement covers multiple potential low-lying areas on the bottom plate of the enclosure. During the manufacturing process, the capacitive probes are pre-deployed according to a gridded layout on the bottom plate surface, with each probe connected to the acquisition front end via an independent capacitive sampling channel. Further, using the physical coordinates of the low-lying water catchment point as a query index, the capacitive probe closest to the spatial location of the low-lying water catchment point is found in the capacitive probe deployment coordinate table. The corresponding capacitive sampling channel of this probe is activated, and the sampling channel is bound to the low-lying water catchment point, thus obtaining a capacitive sampling link pointing to the low-lying water catchment point.

[0053] For example, the physical principle of the dielectric constant change is as follows: In a dry state, the medium surrounding the capacitance probe is air, and its dielectric constant is approximately 1.0; when rainwater collects at the low-lying catchment point and submerges the capacitance probe, the surrounding medium transforms into water, and its dielectric constant rises to approximately 80. This jump in dielectric constant causes the equivalent capacitance value at both ends of the capacitance probe to jump from an initial dry value to a wet value, with the jump reaching tens of times. Preferably, the capacitance sampling link continuously collects capacitance readings at both ends of the capacitance probe according to a preset sampling period, the sampling period ranging from 50 milliseconds to 200 milliseconds. The capacitance readings at each sampling time are concatenated in ascending order of timestamps to form a capacitance value time series at the low-lying catchment point. This capacitance value time series is stored in the form of a two-dimensional array, with the first dimension representing the sampling time and the second dimension representing the corresponding capacitance reading.

[0054] Step S106: Evaluate the result of the spatial misalignment distance exceeding the set distance threshold, and extract the overlapping area between the lateral electrode spacing and the water collection path to obtain the delayed conduction risk of dense electrodes avoiding the low-lying water collection point.

[0055] Obtain the set of Euclidean distances between each electrode and the low-lying water catchment point, and take the minimum value as the spatial misalignment distance between the electrode array and the low-lying water catchment point. Compare the spatial misalignment distance with a set distance threshold, and use the difference between the two as the degree of deviation. The larger the degree of deviation, the farther the electrode array deviates from the low-lying water catchment point, thus obtaining the deviation result of the spatial misalignment distance from the set distance threshold. Based on the electrode coordinates in the gravity coordinate system, the line connecting adjacent electrode coordinate points is taken as the coverage line segment corresponding to the lateral electrode spacing. Extract the physical coordinates of each topological node from the water catchment path node sequence, and take the line connecting adjacent topological nodes as the water catchment path segment. Compare the intersection of the coverage line segment and the water catchment path segment on the base plate plane, and extract the geometrically intersecting part to obtain the overlapping area of ​​the lateral electrode spacing and the water catchment path. The degree of boundary violation is jointly determined by the area of ​​the overlapping region between the lateral electrode spacing and the water collection path. If the degree of boundary violation is greater than a preset threshold and the area of ​​the overlapping region is less than a preset area threshold, it is determined that the dense electrode array deviates from the low-lying water collection point, and the risk of delayed conduction due to the dense electrodes avoiding the low-lying water collection point is obtained.

[0056] In one embodiment, the spatial misalignment distance reflects the degree to which the entire lateral detection electrode array deviates from the low-lying water catchment point. The minimum value is extracted from the set of Euclidean distances between each electrode and the low-lying water catchment point. This minimum value corresponds to the distance from the electrode closest to the low-lying water catchment point to that point, and is used as the spatial misalignment distance between the electrode array and the low-lying water catchment point. This selection method ensures that even if a particular electrode is relatively close to the low-lying water catchment point, as long as this closest distance is still relatively large, it can be determined that the entire array is at risk of deviation.

[0057] Specifically, the degree of deviation is obtained by subtracting the spatial misalignment distance from the set distance threshold. A larger deviation value indicates that the electrode array is further away from the low-lying water collection point; a negative value indicates that at least one electrode has fallen into the sensing range. The set distance threshold has a value range consistent with the physical sensing radius of a single electrode and can be stored locally.

[0058] It should be noted that the lateral electrode spacing is not an isolated numerical interval, but rather the physical coverage area formed by connecting adjacent electrodes in a gravity coordinate system. Based on the electrode coordinates in the gravity coordinate system, adjacent electrode coordinate points are connected in pairwise order according to their electrode numbers. These connecting lines constitute the coverage segment between the electrode pair, and the set of coverage segments reflects the actual sensing distribution of the lateral detection electrode array on the base plate plane. Further, the physical coordinates of each topological node are extracted from the water collection path node sequence in node order, and the coordinate points between adjacent topological nodes are connected sequentially to form water collection path segments. The set of water collection path segments characterizes the actual trajectory of rainwater sliding from the inlet edge through the reinforcing rib, the cable hole, and the bottom edge of the tank to the low-lying water collection point.

[0059] In one possible implementation, the geometric intersection comparison process is as follows: for each line segment in the set of covered line segments, a two-dimensional planar line segment intersection determination is performed sequentially with all path segments in the set of water catchment path segments. The intersection determination uses line segment endpoint parameterization, calculating the intersection parameter t of the lines containing the two line segments. If the intersection parameter t falls within the parameter value range of both line segments, the two line segments are determined to intersect; if any parameter exceeds the value range, the two line segments are determined not to intersect. The intersection points corresponding to all pairs of intersecting line segments together constitute the overlapping area of ​​the transverse electrode spacing and the water catchment path on the bottom plate plane. This overlapping area forms a geometric envelope with the intersection point as the vertex and the local adjacent neighborhood as the radius.

[0060] Understandably, the area of ​​the overlapping region is calculated as follows: First, the base plate plane is divided into a regular grid with a side length of 0.1 meters. All intersection points of the covering line segment and the water collection path segment are traversed. An induction circle is formed by extending the induction radius of a single electrode outward from each intersection point, where the radius of the single electrode induction circle ranges from 0.15 meters to 0.25 meters. The number of grid units covered by all induction circles is counted, with each grid unit contributing 0.01 square meters of area. The total area of ​​the overlapping region is obtained by summing these values. A larger area indicates a more complete physical overlap between the covering line segment and the water collection path segment on the base plate plane; conversely, a smaller area indicates that the electrode array is more likely to physically cross the water collection path.

[0061] Preferably, the preset threshold value corresponding to the boundary crossing degree is approximately 1 / 20 to 1 / 15 of the diagonal length of the base plate; the preset area threshold value corresponding to the overlapping area is approximately 1 / 4 to 1 / 2 of the area of ​​a single electrode sensing circle. That is, when the electrode sensing radius is 0.2 meters, the area of ​​a single sensing circle is approximately 0.126 square meters, and the preset area threshold value is 0.03 square meters to 0.06 square meters. The boundary crossing degree and the overlapping area are jointly determined. If the boundary crossing degree is greater than the preset threshold and the overlapping area is less than the preset area threshold, it is determined that although the dense electrode array is densely arranged, its coverage segments almost completely deviate from the rainwater runoff path, and the nearest electrode is still too far from the low-lying water catchment point, resulting in a delayed conduction risk where the dense electrodes avoid the low-lying water catchment point. This delayed conduction risk is recorded as a flag bit in the local status register. Flag bit 1 indicates that the current distribution box is in a state of delayed conduction risk where the dense electrodes avoid the low-lying water catchment point.

[0062] Step S107: When the rate of change of the capacitance value exceeds a set threshold, a virtual conduction signal is generated and injected into the main circuit cutoff control logic to form a closed conduction judgment circuit. The consistency between the logic state of the circuit and the preset physical cutoff trigger condition is evaluated. If they are consistent, a physical cutoff control command is immediately sent to the main control circuit to execute the physical cutoff control of the main circuit of the distribution box.

[0063] For the capacitance value time series of a low-lying catchment point, the difference between capacitance readings at adjacent sampling times is divided by the sampling period to obtain the rate of change of the capacitance value. If the rate of change of the capacitance value exceeds a set threshold, it is determined that water has accumulated at the low-lying catchment point, and this accumulation event is output as a conduction level signal to obtain a virtual conduction signal. The virtual conduction signal is connected to the main circuit cutoff control logic through a preset logic injection port, and is ORed with the physical conduction signal output by the lateral detection electrode in the control logic. This causes the conduction judgment link that was originally broken due to the electrode avoiding the low-lying catchment point to be filled by the virtual conduction signal, forming a closed conduction judgment loop. The current logic state of the closed conduction judgment loop is obtained, and the logic state is compared with the preset physical cutoff trigger condition, which refers to the trigger condition of the corresponding conduction level in the main circuit cutoff control logic. If the comparison result is consistent, a physical cutoff control command is immediately sent to the main control circuit to execute the physical cutoff control of the distribution box main circuit.

[0064] In one embodiment, the calculation process for the rate of change of the capacitance value is as follows: Capacitance readings C1 and C2 corresponding to two adjacent sampling times t1 and t2 are read from the capacitance value time series of the low-lying catchment point, and... The rate of change between adjacent sampling times is represented by C1, which represents the capacitance reading at the previous time, and C2, which represents the capacitance reading at the current time. The unit of the rate of change is picofarads per second, reflecting the instantaneous rate of change of the dielectric constant around the capacitance probe.

[0065] Specifically, the threshold value is determined based on the maximum random variation rate of the capacitance reading under dry conditions caused by fluctuations in ambient humidity. Preferably, the threshold value ranges from 1% to 5% of the full scale of a single capacitance probe.

[0066] It should be noted that if the rate of change of the capacitance value exceeds the set threshold, it indicates that the dielectric constant at the low-lying water catchment point has increased dramatically. The corresponding physical meaning is that the rainwater has substantially submerged the capacitance probe, and a water accumulation event has occurred at the low-lying water catchment point.

[0067] It is understood that the virtual conduction signal is a logic signal that is equivalent in level form to the original physical conduction signal between the electrodes. Inside the main circuit cutoff controller of the distribution box, the original lateral detection electrode outputs a conduction level by forming a water conduction loop between the electrodes; when a water accumulation event occurs at the low-lying water collection point, the acquisition front end outputs a conduction level equivalent to the conduction level between the electrodes according to the water accumulation event, thus obtaining the virtual conduction signal.

[0068] In one possible implementation, the physical form of the main circuit cutoff control logic is a logic judgment unit fixed in the programmable logic device during configuration. This logic judgment unit includes an OR gate and a comparator. The logic injection port is a reserved auxiliary signal input pin of the programmable logic device. The virtual conduction signal is connected to one input of the OR gate via this pin, and the physical conduction signal output by the lateral detection electrode is connected to the other input of the OR gate. The logic level of the OR gate output changes to a high level when either of the two input signals is high. This fills in the gap in the conduction judgment link, which was previously at a low level due to limitations in the field installation conditions preventing the electrode from completely covering the entire area of ​​the low-lying water catchment point. The conduction judgment link changes from an open circuit to a closed circuit, resulting in a closed conduction judgment loop. Furthermore, the preset physical cutoff trigger condition is stored inside the comparator of the programmable logic device, corresponding to the high-level duration threshold required to trigger the physical cutoff action in the main circuit cutoff control logic.

[0069] For example, during the factory shipment of the main circuit disconnect controller, the trigger condition is configured such that the output of the OR gate remains at a high level for a duration exceeding 30 milliseconds, which is a debouncing time threshold.

[0070] For example, the current logic state of the closed conduction judgment circuit is obtained, and this logic state is represented by the level value of the OR gate output. The logic state is sent to the judgment comparator and compared with the preset physical cut-off trigger condition. If the comparison result is consistent, that is, the logic state satisfies the preset physical cut-off trigger condition, the comparator output is set high; if inconsistent, the comparator output remains low. Preferably, when the comparator output is set high, the judgment comparator immediately sends a physical cut-off control command to the main control circuit. The main control circuit includes a vacuum circuit breaker or contactor shunt trip coil located on the main circuit of the distribution box. The physical cut-off control command drives the shunt trip coil to energize and disconnect the main circuit contact bridge, thereby executing the physical cut-off control of the main circuit of the distribution box.

[0071] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for physically disconnecting the main circuit of a distribution box under extreme weather conditions, characterized in that, The method includes: The three-dimensional spatial coordinate sequence of the distribution box is obtained, and after smoothing, the gravity direction and the spatial attitude of the box bottom plate are extracted. When the angle between the gravity vector and the normal vector of the base plate in the spatial orientation of the distribution box exceeds the set tilt threshold, the tilt recognition result of the base plate of the distribution box is obtained, and the local lowest points of the water inlet side, low side and drainage hole position of the rainwater along the flow direction of the base plate are extracted. By combining the three-dimensional topology of the bottom of the distribution box, the water collection path formed by rainwater along the reinforcing ribs, cable holes, drainage holes and bottom edge of the box is identified, and the local lowest point where the longest residence time on the path is defined as the low-lying water collection point. Obtain the preset distribution coordinates of the lateral detection electrodes, transform them to a coordinate system based on the direction of gravity, and calculate the Euclidean distance between each electrode and the low-lying water collection point. When the distance between all electrodes and the low-lying water catchment point is greater than the set threshold, it is determined that the current electrode layout cannot cover the actual water catchment area, and the capacitance value of the low-lying water catchment point is continuously collected. The results of spatial misalignment distance exceeding the set distance threshold are evaluated, and the overlapping area between the lateral electrode spacing and the water collection path is extracted to obtain the delayed conduction risk of dense electrodes avoiding the low-lying water collection point. When the rate of change of the capacitance value exceeds a set threshold, a virtual conduction signal is generated. The virtual conduction signal is injected into the main circuit cutoff control logic to form a closed conduction judgment circuit. The consistency between the logic state of the circuit and the preset physical cutoff trigger condition is evaluated. If they are consistent, a physical cutoff control command is sent to the main control circuit.

2. The method for physical disconnection control of the main circuit of the distribution box under extreme weather conditions according to claim 1, characterized in that, The process of obtaining the three-dimensional spatial coordinate sequence of the distribution box, smoothing it, and then extracting the gravity direction and the spatial attitude of the box's bottom plate includes: A sequence of coordinate points composed of triaxial acceleration and triaxial angular velocity is obtained; the coordinate point sequence is timestamped according to the sampling time, and data points with abnormal sampling intervals and range overflow are removed; the coordinate point sequence is smoothed by Kalman filtering algorithm, and the gravity field projection in the box coordinate system is separated from the triaxial acceleration components to obtain the gravity vector. The orientation of the bottom plate is represented by the bottom plate normal vector, and the spatial attitude of the bottom plate is obtained by the angle relationship between the gravity vector and the normal vector.

3. The method for physical disconnection control of the main circuit of the distribution box under extreme weather conditions according to claim 2, characterized in that, When the angle between the gravity vector and the normal vector of the base plate in the spatial orientation of the distribution box exceeds a set tilt threshold, the tilt recognition result of the distribution box base plate is obtained, and the local lowest points of the water inlet edge, low edge and drainage hole positions along the rainwater flow direction of the base plate are extracted, including: When the included angle is greater than the set tilt threshold, it is determined that the base plate has tilted beyond the boundary and the tilt direction of the over-boundary tilt is extracted to obtain the tilt identification result of the distribution box base plate; the horizontal projection component of the gravity vector on the base plate plane is used as the rainwater flow direction vector, and the water inlet edge is obtained by extending the flow direction vector in the reverse direction to the edge of the base plate, and the low edge of the base plate is obtained by extending it in the forward direction to the edge of the base plate. The location of the drainage hole is determined by combining the physical coordinates of the drainage hole in the base plate plane diagram; the relative elevation value is compared point by point in the neighborhood of each sampling point, and the location where the relative elevation value is less than other sampling points in the neighborhood is extracted to obtain the local minimum point.

4. The method for physical disconnection control of the main circuit of the distribution box under extreme weather conditions according to claim 1, characterized in that, The method combines the three-dimensional topology of the bottom of the distribution box to identify the water collection path formed by rainwater along the reinforcing ribs, cable holes, drainage holes, and the bottom edge of the box. The local lowest point on the path where the water stays for the longest time is defined as the low-lying water collection point, including: Obtain three-dimensional topological data of the bottom of the distribution box. The three-dimensional topological data includes the geometric position coordinates and height fluctuation information of the reinforcing ribs, cable holes, drainage holes, and the direction of the bottom edge of the box. Discrete sampling is performed on the bottom edge of the box with equal arc length. Together with the reinforcing rib, cable hole and drainage hole, they are used as topological nodes. The physical connectivity between adjacent nodes is used as topological edges to construct a topological node connectivity graph covering the bottom plate surface. Starting from the local lowest point on the water inlet edge, the path is gradually advanced along the topological edge by searching for adjacent nodes in the direction of elevation descent until the local lowest point at the drainage hole is reached, thus obtaining the node sequence of the water collection path. For all local minimum points on the water catchment path node sequence, the local minimum point with the longest retention time is extracted to obtain the low-lying water catchment point.

5. The method for physical disconnection control of the main circuit of the distribution box under extreme weather conditions according to claim 4, characterized in that, After obtaining the sequence of water catchment path nodes, it includes: For each topological node in the water catchment path node sequence, the node water flow rate is obtained based on the ratio of the elevation difference between adjacent nodes to the topological distance between nodes. The node residence time is obtained by dividing the volume of the depression area where the node is located in the three-dimensional topological structure data by the node water flow rate.

6. The method for physical disconnection control of the main circuit of the distribution box under extreme weather conditions according to claim 1, characterized in that, The process of obtaining the preset distribution coordinates of the lateral detection electrodes, converting them to a coordinate system based on the direction of gravity, and calculating the Euclidean distance between each electrode and the low-lying water catchment point includes: Obtain the distribution coordinates of the lateral detection electrodes with reference to the original installation coordinate system of the box; construct a rotation matrix based on the attitude angle of the distribution box base plate, and perform coordinate transformation on the distribution coordinates to obtain the electrode coordinates in the gravity coordinate system with reference to the direction of gravity; for the electrode coordinates in the gravity coordinate system and the low-lying water collection point coordinates in the same gravity coordinate system, calculate the distance between each electrode and the low-lying water collection point one by one using the Euclidean distance formula.

7. The method for physical disconnection control of the main circuit of the distribution box under extreme weather conditions according to claim 1, characterized in that, When the distance between all electrodes and the low-lying catchment point is greater than a set threshold, it is determined that the current electrode layout cannot cover the actual catchment area, and the capacitance value of the low-lying catchment point is continuously collected, including: Obtain the set of Euclidean distances between each electrode and the low-lying water catchment point and set a distance threshold; when all distance values ​​in the set of Euclidean distances are greater than the distance threshold, it is determined that the physical layout of the current lateral detection electrode does not fall within the actual water catchment coverage area of ​​the low-lying water catchment point, and the distribution box is calibrated to be in a state of trigger delay risk; retrieve the corresponding capacitance sampling channel from the capacitance probes deployed at the low-lying water catchment point, and stitch the capacitance readings at each sampling time along the time axis to obtain the capacitance value time series of the low-lying water catchment point.

8. The method for physical disconnection control of the main circuit of the distribution box under extreme weather conditions according to claim 1, characterized in that, The evaluation of spatial misalignment distance exceeding a set distance threshold extracts the overlapping area between the lateral electrode spacing and the water catchment path, obtaining the risk of delayed conduction due to dense electrodes avoiding the low-lying water catchment point, including: The minimum value in the set of Euclidean distances between each electrode and the low-lying water catchment point is taken as the spatial misalignment distance between the electrode array and the low-lying water catchment point. The difference between the spatial misalignment distance and the set distance threshold is taken as the degree of boundary violation. The line connecting adjacent electrodes in the electrode coordinates under the gravity coordinate system is taken as the covering line segment, and the line connecting adjacent topological nodes in the water catchment path node sequence is taken as the water catchment path segment. The part where the covering line segment and the water catchment path segment intersect geometrically is extracted to obtain the overlapping area of ​​the lateral electrode spacing and the water catchment path. When the degree of boundary violation is greater than the preset threshold and the area of ​​the overlapping area is less than the preset area threshold, it is determined that the dense electrode array deviates from the low-lying water catchment point.

9. The method for physical disconnection control of the main circuit of a distribution box under extreme weather conditions according to claim 1, characterized in that, When the rate of change of the capacitance value exceeds a set threshold, a virtual conduction signal is generated. This virtual conduction signal is injected into the main circuit cutoff control logic, forming a closed conduction judgment loop. The loop's logic state is evaluated for consistency with the preset physical cutoff trigger condition. If they are consistent, a physical cutoff control command is sent to the main control loop, including: For the capacitance value time series of the low-lying water catchment point, the difference between the capacitance readings at adjacent sampling times is divided by the sampling period to obtain the rate of change of the capacitance value. When the rate of change of the capacitance value exceeds a set threshold, it is determined that water has accumulated at the low-lying water catchment point. This accumulation event is output as a virtual conduction signal in the form of a conduction level signal. The virtual conduction signal is connected to the main circuit cutoff control logic through the logic injection port and performs an OR operation with the physical conduction signal output by the lateral detection electrode. The disconnected conduction judgment link is filled to form a closed conduction judgment loop. The logic state of the closed conduction judgment loop is obtained and compared with the preset physical cutoff trigger condition. When the comparison result is consistent, a physical cutoff control command is sent to the main control loop.

10. The method for physical disconnection control of the main circuit of a distribution box under extreme weather conditions according to claim 9, characterized in that, The preset physical cutoff trigger condition is the trigger condition corresponding to the conduction level in the main circuit cutoff control logic.

Images