A substation general plan arrangement scheme automatic generation method and system

Abstract

The present application relates to the technical field of device layout planning, in particular to a substation general layout scheme automatic generation method and system. The present application first divides the orthogonal discrete grid, constructs the initial grid occupation state matrix, reads various data of the device; further constructs the connection topology transfer matrix, obtains the topology node convergence degree of each device; further obtains the weighted center coordinates of the target device, executes the concentric square scanning, combines the latest grid occupation state matrix, obtains the target placement area; further updates the grid occupation state matrix; finally, after traversing all devices, based on the final grid occupation state matrix and site elevation reference data, the substation three-dimensional general layout scheme is generated by mapping, and the hard bus cannot be bent and the anti-shielding constraint is satisfied.

Description

Technical Field

[0001] This invention relates to the field of equipment layout planning technology, specifically to a method and system for automatically generating a substation general layout plan. Background Technology

[0002] In the design of substation transmission and transformation projects, the overall site layout is a core preliminary step, aiming to rationally arrange the spatial location of various electrical equipment within the site while meeting electrical safety regulations. With the development of computer-aided design, the industry has gradually introduced automated global search algorithms to assist manual equipment layout.

[0003] However, existing automated layout algorithms have significant limitations when dealing with the specific physical constraints of substations. On the one hand, equipment connections in the high-voltage areas of substations typically use rigid tubular busbars, which are completely inflexible, requiring that the pipelines between connected equipment remain straight and unobstructed. Existing conventional algorithms often treat all connections as equivalent flexible topological edges, lacking a global quantitative consideration of the degree of restriction on pipeline bending, which can easily lead to core hub equipment being pushed to the periphery of the site. On the other hand, conventional algorithms only consider avoiding the footprint of already placed equipment when placing equipment, without taking into account the physical straight passage corridor of the busbar. This can easily lead to subsequent equipment directly cutting off the connection path of previous equipment, resulting in equipment boundary or pipeline laying conflicts when reconstructing a 3D solid scene. Summary of the Invention

[0004] To address the technical problem that existing substation layout algorithms fail to consider the inflexibility constraints and anti-obstruction requirements of rigid busbars, resulting in layout schemes prone to equipment and pipeline overlap and interference, this invention aims to provide an automatic method and system for generating substation overall layout schemes. The specific technical solution adopted is as follows: A method for automatically generating a substation general layout scheme, the method comprising: Divide the two-dimensional design base map into orthogonal discrete grids and construct the initial grid occupancy state matrix; read the connection path constraint coefficients between the devices to be deployed, and obtain the safe occupation area and traversal index of each device; A connection topology transfer matrix is ​​constructed based on all the connection path constraint coefficients. Iteration is performed using a preset smoothing constant and a uniform distribution vector to obtain the topology node convergence degree of each device to be deployed. Based on the topology node convergence degree, the device's safe occupancy area, and the traversal index, the devices to be deployed are sequentially identified as target devices. Weighted center coordinates are obtained based on the grid coordinates of the calibrated associated devices of the target device and the corresponding connection path constraint coefficients. A concentric square scan is performed based on the weighted center coordinates of the target device, and the target placement area is obtained by combining the latest grid occupancy state matrix. Based on the target placement area, the occupancy refresh mechanism of the bus corridor is activated to update the grid occupancy status matrix; after traversing all devices, a three-dimensional overall layout scheme of the substation is generated based on the final grid occupancy status matrix and site elevation benchmark data.

[0005] Furthermore, the method for obtaining the convergence degree of the topology nodes includes: Using the connection topology transfer matrix, a preset smoothing constant, and an equal distribution vector, perform state superposition product iteration to obtain the updated node influence distribution vector; calculate the maximum value of the absolute value of the difference between the node influence distribution vectors of adjacent iterations; if the maximum value of the absolute value of the difference is less than the preset convergence tolerance constant, or the number of iterations reaches the preset upper limit of the number of iterations, terminate the iteration, and use the values ​​of each element in the steady-state node influence distribution vector as the topology node convergence degree of the corresponding device to be deployed.

[0006] Furthermore, the method for obtaining the weighted center coordinates includes: Assign a corresponding path constraint coefficient as a weight to the grid coordinates of all calibrated associated devices of the target device; calculate the weighted average of the weighted grid coordinates and round it to obtain the weighted center coordinates of the shortest connection of the target device, which is then used as the weighted center coordinates.

[0007] Furthermore, when there are no calibrated associated devices for the target device, the coordinates of the incoming and outgoing ports with the same voltage level as the target device are matched from the pre-acquired set of incoming and outgoing port coordinates. The average value of the matched incoming and outgoing port coordinates is then calculated and rounded down to be used as the weighted center coordinates.

[0008] Furthermore, the method for obtaining the target placement area includes: A search trajectory of concentric squares is generated by expanding outward from the weighted center coordinates as the origin, and the grid nodes on the search trajectory are extracted as candidate starting points. Based on the candidate starting points, combined with the grid length span and grid width span of the target device, candidate grid blocks are extracted. A virtual connection is constructed based on the candidate center of the target device and the geometric center of the corresponding calibrated associated device. With the grid connection where the virtual connection is located as the center, a preset half-width is expanded outward to both sides along the vertical normal direction to generate the projection grid area of ​​the straight line connection of the bus corridor. When all identifiers of the candidate grid block in the grid occupancy status matrix are found to be free identifiers, and all elements in the projected grid area of ​​the bus corridor straight line connection, except for the grid area occupied by the calibrated associated equipment body at the opposite end, have all identifiers in the grid occupancy status matrix as free identifiers, the candidate grid block is established as the target placement area.

[0009] Furthermore, the sorting method for the target devices includes: The devices to be deployed are sorted in descending order based on the convergence degree of the topology nodes. If there is a group of devices to be deployed with the same convergence degree of topology nodes in the first sorting result, the devices to be deployed in the group are sorted in descending order based on the safe space occupied by the devices. If there are subgroups of devices with the same safe space occupied by the devices in the second sorting result, then the devices to be arranged in the subgroup are sorted in ascending order for the third time using the traversal index. Based on the final sorting results, a device sorting table is generated, and the devices to be deployed are selected as target devices in sequence according to the order in the device sorting table.

[0010] Furthermore, the method for updating the grid occupancy state matrix includes: After establishing the target placement area, the elements in the grid occupancy status matrix that belong to the target placement area and the corresponding bus corridor straight line projection grid area are updated from idle identifiers to occupied identifiers to lock the equipment physical area and pipeline laying corridor location.

[0011] Furthermore, the method for obtaining the connection path constraint coefficient includes: Extract the voltage levels between the equipment to be arranged in the electrical main wiring topology table, call the pre-configured voltage level and non-bendability coefficient mapping table, and generate the connection path constraint coefficient.

[0012] Furthermore, the method also includes an out-of-bounds dead loop prevention truncation mechanism when no target placement area is generated: During the search trajectory of expanding outward to generate concentric squares, the current side length span value of the outward expansion is obtained in real time; if the side length span value is greater than the total number of diagonal grid spans of the site obtained in advance, it is determined that the target placement area cannot be generated, the current search operation is terminated and an abnormal site space alarm signal is output.

[0013] The present invention also proposes an automatic substation layout scheme generation system, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement any of the steps of the automatic substation layout scheme generation method described above.

[0014] The present invention has the following beneficial effects: This invention first divides the grid into orthogonal discrete grids, providing a unified discrete coordinate reference system and computational basis for the placement and collision detection of all subsequent equipment. An initial grid occupancy state matrix is ​​constructed, and various data from the equipment are read to lay the foundation for subsequent analysis. Next, a connection topology transfer matrix is ​​constructed, making the constraint degree of each connection a weight for the influence transmission between nodes in the topology network. The convergence degree of each topology node is obtained, directly quantifying the hub weight of each device in the station's inflexible bus network. Furthermore, the weighted center coordinates of the target equipment are obtained, quantifying the optimal theoretical physical reference point for the target equipment to approach the already deployed equipment. A concentric square scan is performed, and combined with the latest grid occupancy state matrix, the target placement area is obtained, finding a continuous blank area that satisfies both electrical connection constraints and is completely free of physical collisions. The grid occupancy state matrix is ​​further updated, forcing subsequent equipment to automatically avoid the planned high-voltage pipeline corridor in space. Finally, after traversing all equipment, a three-dimensional overall layout scheme for the substation is generated based on the final grid occupancy state matrix and site elevation benchmark data.

[0015] This invention obtains the convergence degree order of topological nodes through the connection path constraint coefficient, uses weighted center coordinates to concentrically scan to find the target placement area, refreshes the bus corridor occupancy and maps to generate a three-dimensional overall plane, and collaboratively satisfies the constraints of rigid busbars being non-bendable and preventing occlusion. Attached Figure Description

[0016] Figure 1 A flowchart illustrating an automatic method for generating a substation general layout scheme according to an embodiment of the present invention; Figure 2 This is a flowchart illustrating a method for obtaining a target placement area according to an embodiment of the present invention. Detailed Implementation

[0017] The following description, in conjunction with the accompanying drawings, details the specific scheme of the automatic generation method and system for substation general layout provided by the present invention.

[0018] Please see Figure 1 The document illustrates a flowchart of an automatic substation layout scheme generation method according to an embodiment of the present invention, specifically including: Step S1: Divide the orthogonal discrete grid according to the two-dimensional design base map and construct the initial grid occupancy state matrix; read the connection path constraint coefficient between the devices to be deployed, and obtain the safe occupation area and traversal index of each device.

[0019] Substation layout involves a large number of equipment coordinate arrangements and physical boundary avoidances. Continuous two-dimensional design base maps incur huge floating-point operation overhead when directly determining geometric overlap. First, the design base map is divided into orthogonal discrete grids with a fixed size resolution, and an initial grid occupancy state matrix is ​​constructed. The idle or occupied state of each grid is recorded in a Boolean two-dimensional matrix, providing a unified discrete coordinate reference system and calculation bearing base for the subsequent placement of all equipment and collision detection.

[0020] Since the degree of non-bending of connecting lines of different voltage levels in substations varies during installation, the connection path restriction coefficient between the equipment to be installed is also read to characterize the bending difficulty of the tubular busbar between the two equipment, and the equipment safety occupancy area and traversal index of each equipment are obtained to lay the foundation for subsequent analysis.

[0021] In a preferred embodiment of the present invention, the system first obtains a two-dimensional design base map of the substation site, and then divides the continuous two-dimensional design base map into discrete orthogonal grids with a pre-set size resolution (e.g., 0.5m × 0.5m precision commonly used in engineering design). Then, a global Boolean two-dimensional matrix that is completely consistent with the size ratio of the site is constructed in memory.

[0022] Next, the system uniformly sets all grid elements in the global Boolean two-dimensional matrix before the analysis begins to an idle identifier, which serves as the initial grid occupancy state matrix; in this example, the idle identifier is 0 and the occupancy identifier is 1. Based on this, the system scans the external polygonal boundary of the site, extracts the absolute grid coordinates of the external high-voltage inlet and low-voltage outlet, and reads the voltage level attributes of the inlet and outlet ports marked on the 2D design map. According to the read voltage level attributes, the system divides the extracted grid coordinates into a subset of high-voltage inlet port coordinates and a subset of low-voltage outlet port coordinates, which together constitute the inlet and outlet port coordinate set.

[0023] At the same time, the system reads the electrical main wiring topology netlist corresponding to the substation engineering drawings, finds all devices with direct electrical connections, and for each pair of devices to be arranged with a connection relationship, extracts the voltage level value marked on the connection line, calls the pre-configured voltage level and non-bendability coefficient mapping table, and generates the connection path restriction coefficient.

[0024] In this example, the mapping table can be configured as follows: for 500kV ultra-high voltage connection lines, due to their large tubing that is completely inflexible, the connection path restriction factor is configured as 1.0; for 220kV connection lines, the connection path restriction factor is configured as 0.8; and for 110kV connection lines, the connection path restriction factor is configured as 0.5.

[0025] For all node pairs in the electrical topology netlist that do not have a direct connection relationship, the connection path constraint coefficient is uniformly set to 0.

[0026] Furthermore, the system loads relevant design technical specifications documents for substation construction. The system extracts the fireproof distance constants and electrical insulation clearance constants required for different voltage levels and different insulation media from the specifications documents, converts the extracted constants into a key-value pair data dictionary, and stores it in memory.

[0027] To prevent short circuits caused by distances between energized parts being less than the safety insulation requirements, the system needs to convert the safety clearance into orthogonal outward expansion margins on the equipment boundaries. The process of obtaining the equipment safety footprint area and traversal index for each device includes: Based on the order in which each device to be deployed is extracted from the electrical main wiring topology netlist, each device to be deployed is assigned a consecutive natural number from 1 to N as the traversal index of the device. The system sequentially extracts each device i to be arranged from the traversal index, calls the 3D BIM device model library, extracts the maximum horizontal projection size of the device i to be arranged from the model library, and obtains the two-dimensional base length value and the two-dimensional base width value. Then, extract the insulation medium parameters of device i, look up the fireproof distance constant and the electrical insulation clearance constant in the data dictionary, and take the maximum value as the safety extension constant of device i.

[0028] Add twice the safety extension constant to the two-dimensional base length value and the two-dimensional base width value respectively to obtain the safety length value and the safety width value; use the product of the safety length value and the safety width value as the equipment safety occupancy area of ​​device i. Next, the safety length and safety width values ​​after adding the safety margin are divided by the side length of the grid size and rounded up to obtain the grid length span and grid width span corresponding to the equipment in the grid matrix, providing a basis for subsequent analysis.

[0029] It should be noted that, in other embodiments of the present invention, the implementer may adjust the specific precision of the grid size resolution, the specific coefficient ratio in the voltage level and non-bendability coefficient mapping table, and the specific spacing constant in the data dictionary according to the needs of the scenario. The specific values ​​of these parameters can be set and dynamically updated by those skilled in the art based on the actual land area of ​​different substations, power grid specifications, or underlying computing power requirements.

[0030] Step S2: Construct a connection topology transfer matrix based on the constraint coefficients of all connection paths, and iterate using a preset smoothing constant and uniform distribution vector to obtain the topology node convergence degree of each device to be deployed; based on the topology node convergence degree, the safe occupation area of ​​the device, and the traversal index, sequentially select the devices to be deployed as target devices; based on the grid coordinates of the calibrated associated devices of the target device and the corresponding connection path constraint coefficients, obtain the weighted center coordinates; perform a concentric square scan based on the weighted center coordinates of the target device, and combine the latest grid occupancy state matrix to obtain the target placement area.

[0031] Considering that the number of connections of a single device cannot reflect its true constraining position in the global electrical network, the constraint coefficients of the connection paths between all devices are constructed as a connection topology transfer matrix, so that the constraint degree of each connection is used as the influence transmission weight between nodes in the topology network. A smoothing constant and an equally distributed vector are introduced to perform iterative calculations on this matrix until convergence, so that the constraint influence of each device can be fully diffused and superimposed along all possible paths in the entire network. Finally, the topology node convergence degree of each device to be deployed is obtained, which directly quantifies the hub weight of each device in the station's inflexible bus network, providing a basis for subsequent device ranking.

[0032] Preferably, in one embodiment of the present invention, for a substation with a total of The system constructs a memory array for the devices to be deployed, with a maximum of 10 rows and 10 columns. The global two-dimensional matrix is ​​then used. The path constraint coefficient between device i and device j is then filled into the i-th row and j-th column of the matrix. The binary combination of all devices is traversed, and each position of the matrix is ​​filled one by one. Since there are no physical connections between devices (no self-loops), the elements on the main diagonal are 0. After the data is entered, the system performs row-by-row normalization on the matrix, specifically including: The system calculates the sum of all column element values ​​in the current row. If the sum is not 0, it divides each element value in the current row by the sum to normalize it; if the sum is 0, it assigns a value to each element value in the current row. ; After traversing all rows, the obtained matrix is ​​used as the connection topology transfer matrix. The system normalizes the connection path constraint coefficient representing the single-sided attribute into the state transfer ratio between nodes, thereby constructing the probability transition model required for subsequent iterations.

[0033] Furthermore, considering that the inflexible restraint effect of high-voltage trunk lines in substations can have long-range cascading effects along complex electrical networks, in this embodiment, the system introduces a smoothing term and performs a step-by-step matrix iterative product operation to find the network convergence focus of this line restraint effect in the entire graph, thereby identifying the key equipment that plays a crucial role in determining the overall line length. Specifically: The state superposition product iteration is performed using the connection topology transfer matrix, a preset smoothing constant, and an equal distribution vector to obtain the updated node influence distribution vector; the maximum value of the absolute difference between the node influence distribution vectors of adjacent iterations is calculated; if the maximum value of the absolute difference is less than the preset convergence tolerance constant, or the number of iterations reaches the preset upper limit of the number of iterations, the iteration is terminated, and the values ​​of each element in the steady-state node influence distribution vector are respectively used as the topology node convergence degree of the corresponding device to be deployed.

[0034] As an example, a uniformly distributed vector has a length of N and all values ​​are equal. vector As the initial distribution vector, the smoothing constant in this example Set the value to 0.85, and set the preset convergence tolerance constant. Take 0.0001; In the Rounds ( In the calculation of ), the system uses the following state superposition product formula to calculate the updated distribution vector. : ; The topology transfer matrix for the connection has dimensions of . The dimension of the distribution vector is ; This represents the incremental value resulting from the propagation and superposition of the previous state to the current network topology connections, reflecting the importance transfer brought about by the actual connections between devices; This represents a mathematical smoothing compensation term. The purpose of introducing this smoothing term is to complement the assignment of values ​​to isolated devices in the previous sub-step, preventing iterative operations from getting stuck in a dead end in a local dead-end network.

[0035] In the current number After the above formula for each round is calculated, the system extracts the current distribution vector sequentially. Compared with the previous round distribution vector For elements with the same position index, perform an absolute difference calculation. The system filters out the largest absolute difference value among all values; if it is less than 1, it is considered a maximum. Then terminate the iteration and... The element values ​​in the table are assigned to the corresponding devices, and are used as the topology node convergence degree of the corresponding devices to be deployed. The larger the value, the more concentrated the trunk line restraint carried by the device is, and the more significant the impact of its coordinate deviation on the total length of the global busbar is.

[0036] If the maximum value of the absolute value of the difference is greater than Then the currently generated The process continues in the next round of matrix product calculation until the termination condition is met.

[0037] It should be noted that, to prevent getting stuck in an abnormal infinite loop, a preset upper limit for the number of iterations is set; in this example, it is set to 1000. and The settings are determined by technicians based on the total number of nodes and accuracy requirements of the substation network topology. In other embodiments of the present invention, the implementer may adjust them as needed.

[0038] In substations, devices of the same voltage level (such as multiple sets of three-phase circuit breakers in the same circuit) may have completely symmetrical structural positions in the electrical network topology. This can lead to the system calculating identical topology node convergence values. If a group of devices with completely identical parameters is encountered, a computer lacking explicit instructions may output random out-of-order values, resulting in deployment failure. Therefore, the system uses topology node convergence, safe equipment footprint, and traversal index to ensure the generation of an absolutely unique queue, sequentially selecting the devices to be deployed as target devices.

[0039] As a preferred approach: perform the first descending sorting of the devices to be deployed based on the convergence degree of the topology nodes to ensure that the device with the highest topology hub status gets the earliest deployment right; Under the premise that the topological hub status is the same, the larger the area occupied by the equipment, the larger the area of ​​continuous blank grid area that needs to be found. When the site is still in a relatively complete state, it should be prioritized to be removed from the list and placed in the position. This can avoid the site becoming fragmented due to the gradual filling of other equipment, which would prevent large-area equipment from being unable to find a sufficiently large continuous blank area. Therefore, if there are groups of equipment to be placed with the same topological node convergence degree in the first sorting result, the equipment to be placed in the group will be sorted in descending order a second time based on the safe occupation area of ​​the equipment. If there are subgroups of devices with the same safe space occupied by the devices in the second sorting result, the devices in the subgroup are sorted in ascending order for the third time using the traversal index. Due to the uniqueness of the traversal index, it is ensured that there is still a definite and unchangeable sorting rule even when the two sets of main physical and topological parameters are exactly the same. This completely eliminates the system uncertainty caused by parallel items. Finally, a unique and fixed device sorting table is generated based on the final sorting result. According to the order in the device sorting table, the devices to be deployed are selected as target devices in sequence.

[0040] To minimize the total length of the non-bendable busbar by placing the newly placed target equipment as close as possible to its connected and already placed associated equipment, the system extracts the associated equipment that has a direct electrical connection to the target equipment and marks the calibrated associated equipment that has been calculated and located in previous layout rounds. Based on the grid coordinates of the calibrated associated equipment of the target equipment and the corresponding connection path constraint coefficient, the weighted center coordinates are obtained, and the optimal theoretical physical reference point for the target equipment to move closer to the already placed equipment is quantified, thus avoiding long-distance detours of pipelines from the source.

[0041] Considering that the weighted center coordinates are only a geometric point without physical dimensions and cannot directly accommodate equipment with actual floor space, a concentric square scan is performed based on the weighted center coordinates of the target equipment. Combined with the latest grid occupancy state matrix, the target placement area is obtained, and a continuous blank area that meets both electrical connection constraints and is completely free of physical collisions is found to ensure the geometric feasibility of equipment layout.

[0042] Preferably, in one embodiment of the present invention, there is a scenario of expansion or modification where new equipment is installed in the space of existing equipment. For equipment already installed and not moved in the scenario, before analyzing the installation location of the equipment to be deployed, the identifiers of the corresponding grid elements in the corresponding equipment entity grid area and bus corridor grid area in the grid occupancy state matrix are updated to occupancy identifiers, and the connection path constraint coefficient is obtained. At this time, the calibrated associated equipment includes pre-existing and immovable associated equipment, as well as associated equipment that has been sequentially placed in the target placement area in the current layout process.

[0043] Since the connection path constraint coefficient is larger, the requirements for the straightness and short distance of the connection line are more stringent. Therefore, the attraction strength of the corresponding associated equipment to the target equipment placement position should also be higher. So, the grid coordinates of all the calibrated associated equipment of the target equipment are assigned the corresponding connection path constraint coefficient as weight. The weighted average of the grid coordinates after weighting is calculated and rounded to obtain the weighted center coordinates of the shortest connection of the target device, which is then used as the weighted center coordinates.

[0044] In this weighted calculation process, the larger the weight value of the associated device coordinates, the greater its influence on the final calculation result. This makes the calculation result geometrically biased towards the associated devices with more rigid connection constraints and less tolerance for detours. The theoretical coordinates obtained after weighted averaging usually contain decimal values. Rounding these coordinates to the nearest grid node yields the shortest connection weighted center coordinates. These coordinates represent the optimal geometric reference position where the target device, after considering the different constraint strengths of all placed associated devices, can minimize the total weighted length of all busbars.

[0045] Meanwhile, considering that the target device may not have any calibrated associated devices, in this case, the coordinates of the incoming and outgoing ports with the same voltage level as the target device are matched from the pre-acquired set of incoming and outgoing port coordinates. The average value of the matched incoming and outgoing port coordinates is then calculated and rounded down to be used as the weighted center coordinates.

[0046] The coordinate rounding scheme specifically involves rounding the calculated x-coordinate and y-coordinate values ​​to the nearest integer grid index. If the matched set of entrance / exit port coordinates is empty, the obtained preset site geometric center grid coordinates are assigned as the weighted center coordinates.

[0047] Further, please refer to Figure 2 The flowchart illustrates a method for obtaining a target placement area according to an embodiment of the present invention, specifically including: Step S201: Generate a search trajectory of concentric squares by expanding outward from the weighted center coordinates as the origin, and extract the grid nodes on the search trajectory as candidate starting points.

[0048] Since the weighted center coordinates represent a specific grid cell in the grid occupancy state matrix, the area of ​​a single grid is usually much smaller than the safe occupancy area of ​​the actual device and cannot directly accommodate the physical device entity. Therefore, the weighted center coordinates are used as the origin (scanning origin), and a search trajectory of concentric squares is dynamically generated in all directions with the side length of a single discrete grid as a fixed step size. The grid nodes on the search trajectory are extracted as candidate starting points.

[0049] The search trajectory of a concentric square refers to a square-shaped loop composed of discrete grid nodes that expands layer by layer from the search origin outwards; the... The Chebyshev distance between all grid nodes on the layer search trajectory and the search origin is equal to ,in For integers greater than or equal to 0, the first... The side length of the concentric squares is Each grid unit represents a search trajectory, which is the grid nodes around the outermost ring of the square.

[0050] when At this time, it means that only the grid where the weighted center coordinates are located is used as a candidate starting point; if the target placement area is not determined through steps S202-S204, then the value is increased. This provides a new candidate starting point.

[0051] The concentric square expansion represents the process by which the target device searches for a place to fit in the outer ring at the cost of being closest to the optimal theoretical point when the optimal theoretical point is occupied, thus ensuring the compactness and optimization of the planar layout space.

[0052] Step S202: Based on the candidate starting point, and combined with the grid length span and grid width span of the target device, extract the candidate grid block.

[0053] Considering that physical devices need to exclusively occupy a space range in the discrete grid that strictly corresponds to the device's physical size plus a safety clearance, in this embodiment, when capturing each candidate starting point, the system uses the candidate starting point as the upper left corner coordinate reference point and extends simultaneously to the right and lower sides of the grid occupancy state matrix to extract a rectangular grid block with the same grid length span and grid width span as the target device, which is then used as a candidate grid block, quantifying the absolute physical boundary range that the target device needs to occupy at the candidate position.

[0054] Here, the coordinate system of the grid occupancy state matrix is ​​defined with the upper left corner of the two-dimensional plane as the origin, the positive X-axis pointing horizontally to the right, and the positive Y-axis pointing vertically downward. During this truncation process, if the corresponding grid coordinate value after extension exceeds the maximum number of rows or columns set by the grid occupancy state matrix, the truncation verification is skipped, and the corresponding candidate starting point is marked as unusable and discarded.

[0055] If a candidate grid block has an occupied identifier in the grid occupancy state matrix, the corresponding candidate starting point is marked as unavailable, and the analysis does not proceed to steps S203 and S204; if none of the candidate starting points proceed to the next step, the analysis returns to step S201 for further analysis. .

[0056] Step S203: Based on the candidate center of the target device and the geometric center of the corresponding calibrated associated device, construct a virtual connection. Taking the grid connection where the virtual connection is located as the center, expand outwards to both sides along the vertical normal direction by a preset half-width to generate the projection grid area of ​​the straight line connection of the bus corridor.

[0057] Considering that the addressing reference point (usually the upper left corner) and the actual electrical interface point (usually the equipment center) do not coincide in space, and the bus corridor has the actual physical width for maintenance and insulation, the candidate starting point is shifted to the right by half the grid length span and down by half the grid width span to obtain the candidate center. The upper left corner coordinates of the corresponding calibrated associated equipment are also shifted to obtain the geometric center. Virtual connections are constructed based on the candidate centers of the target equipment and the geometric centers of the corresponding calibrated associated equipment. Using the grid lines containing these virtual connections as centers, a preset half-width is extended outwards to both sides along the perpendicular normal direction to generate the projected grid area of ​​the bus corridor straight-line connections. Through the discretization mapping of the continuous geometric space, the electrical corridor range that the rigid high-voltage busbar must reach and cannot infringe upon in the two-dimensional discrete matrix is ​​quantified, avoiding misalignment of engineering entities caused by connection coordinate offsets.

[0058] Among them, the projected grid area of ​​the straight line connection of the bus corridor represents the comprehensive space requirement after superimposing the physical width of the rigid tubular busbars, electrical insulation safety clearance, and maintenance access. The implementer calibrates the preset half-width according to the phase-to-phase insulation clearance and maintenance access width parameters of the corresponding voltage level in the substation design code. In this embodiment, the preset half-width is 1 grid width.

[0059] It should be noted that when there are no calibrated associated devices for the target device, step S203 is skipped, and the projection grid area of ​​the straight line connection of the bus corridor is empty.

[0060] Step S204: When all identifiers of the candidate grid blocks in the grid occupancy status matrix are free identifiers, and all elements in the projected grid area of ​​the bus corridor straight line connection except for the grid area occupied by the calibrated associated equipment body at the opposite end have free identifiers in the grid occupancy status matrix, the candidate grid block is established as the target placement area.

[0061] Considering that substation design requires not only that the physical entities of the equipment do not overlap, but also that the laying path of the rigid tubular busbar cannot be blocked or interrupted by any obstacles, and considering that one of the reference points for generating the projection grid area of ​​the straight line connection of the busbar corridor is the geometric center of the calibrated associated equipment, which is located inside the grid area occupied by the calibrated associated equipment, when checking the occupancy status of the projection grid area of ​​the straight line connection of the busbar corridor, the grid area already registered and occupied by the calibrated associated equipment at the opposite end is ignored to avoid a conflict between the corridor verification and the occupancy of the associated equipment itself. Therefore, the identifiers in the (latest) grid occupancy status matrix are checked, and when the above conditions (conditions of step S204) are met, the candidate grid block is established as the target placement area.

[0062] It should be noted that, for the same layer, the Chebyshev distance between all grid nodes on the search trajectory and the search origin is equal. Therefore, any candidate position that meets the conditions on this layer is equivalent in terms of spatial cost. The natural traversal order (i.e., the order of extraction in clockwise or counterclockwise order) is used as the basis for decision when there are multiple solutions on the same layer, and the target placement area is determined according to the traversal order.

[0063] Considering the limited total area of ​​the substation site, this embodiment also includes a boundary-crossing anti-dead-loop cutoff mechanism when there are too many devices to be deployed or the site is severely fragmented due to the initial layout: During the search trajectory of expanding outward to generate concentric squares, the current number of layers being expanded outward is obtained in real time. If the number of floors If the span of the long side grid exceeds the total number of pre-acquired field grid spans, it is determined that the target placement area cannot be generated, the current search operation is terminated, and a field space anomaly alarm signal is output to improve the anomaly handling capability of the automated layout system.

[0064] Step S3: Based on the target placement area, activate the bus corridor occupancy refresh mechanism to update the grid occupancy status matrix; after traversing all devices, based on the final grid occupancy status matrix and site elevation benchmark data, map and generate the three-dimensional overall layout scheme of the substation.

[0065] Considering that conventional grid placement only considers the equipment itself, which can easily lead to the subsequent layout cutting off the planned straight pipelines, a bus corridor placement refresh mechanism is activated based on the target placement area. This updates the grid occupancy status matrix, marking the grids occupied by the equipment itself and the grids occupied by the pipeline passage corridor. This forces the subsequent equipment to be placed to automatically avoid the planned high-pressure pipeline laying corridor in space, completely blocking the risk of spatial overlap, interference, and deadlock.

[0066] Finally, in order to align the discrete Boolean anti-collision calculation results of the two-dimensional mesh matrix to the real-world engineering coordinate system, after traversing all equipment, based on the final mesh occupancy state matrix and site elevation benchmark data, the calculation from the underlying abstract data is transformed into an intuitive and compliant three-dimensional engineering view output, ultimately generating a three-dimensional general layout scheme for the substation with direct engineering implementation value.

[0067] Preferably, in one embodiment of the present invention, after the target placement area is established, the physical space occupied by the target placement area and the straight-line laying channel required for the high-voltage tubular busbar between the target placement area and the associated equipment should be exclusive in the subsequent layout process. Therefore, the elements in the grid occupancy status matrix belonging to the target placement area and the corresponding busbar corridor straight-line projection grid area are updated from idle identifiers to occupied identifiers to lock the equipment entity area and the pipeline laying corridor position. The busbar corridor occupancy refresh mechanism is as follows: in updating the grid occupancy status matrix, the elements in the busbar corridor straight-line projection grid area are updated.

[0068] Finally, the system receives the final grid occupancy status matrix, which contains the two-dimensional planar grid coordinates of all devices and the coordinate dataset (target placement area) corresponding to each device, and extracts the site elevation benchmark data; The system connects to a substation 3D computer-aided design platform (such as BIM engine-based 3D design software) via an external application programming interface (API). Based on the traversal index of each device, the system accurately retrieves the 3D model file of the physical appearance that matches the device from the built-in component library of the 3D-aided design platform.

[0069] Next, the system extracts the size resolution constants set when defining the mesh matrix in the early stage (e.g., (meters / grid). The system extracts the two-dimensional planar grid coordinates of each device and performs a multiplication operation with the size resolution constant. Using the product result, the system accurately converts the virtual grid coordinate values ​​into absolute length values ​​(in millimeters or meters, including absolute X-axis and absolute Y-axis coordinates) in the three-dimensional platform world coordinate system.

[0070] Subsequently, the system reads the default Z-axis elevation datum data (i.e., the absolute altitude or relative elevation of the site's ground plane) configured for the currently specified substation site within the 3D platform. The system then merges the converted (X,Y) plane coordinate data with the extracted Z-axis elevation datum data to form the complete 3D spatial insertion point coordinates (X,Y,Z).

[0071] Finally, the system injects the coordinates of the 3D insertion point into the reference insertion point attributes of each retrieved entity appearance model file. The system drives the rendering engine of the 3D computer-aided design platform to batch load all the 3D entity models with injected absolute spatial coordinates into the view space. After all components are assembled, the system directly generates and outputs the final 3D site layout scene file of the substation, which can be directly reviewed, applied, or delivered as drawings by designers.

[0072] It should be noted that constructing virtual connection projections during the target placement area acquisition phase is a preliminary interference detection before placement. Initiating the bus corridor occupancy refresh mechanism after establishing the target placement area is a post-placement state locking operation. These two steps, rather than a single operation, are executed to avoid data rollback due to search failures. During the concentric square scanning phase, there are numerous non-compliant candidate positions. If the state is written synchronously during detection, any contaminated grid markers must be revoked upon detection of collision interference. The step-by-step mechanism ensures the read-only attribute of the global grid occupancy state matrix during the optimization traversal process, performing a one-time write confirmation only after the final target placement area is established, eliminating the risk of dirty matrix data writes and wasted computational power.

[0073] An embodiment of the present invention also provides an automatic substation layout scheme generation system. The system includes a memory, a processor, and a computer program. The memory is used to store the corresponding computer program, and the processor is used to run the corresponding computer program. When the computer program runs in the processor, it can realize the automatic substation layout scheme generation method described in steps S1-S3.

[0074] In summary, to address the technical problem that existing substation layout algorithms fail to consider both the inflexibility constraint of rigid busbars and the need for anti-obstruction during installation, leading to frequent overlap and interference between equipment and pipeline spaces, this invention aims to provide an automatic method and system for generating substation overall layout schemes. This invention first divides the substation into orthogonal discrete grids, constructs an initial grid occupancy state matrix, and reads various equipment data; further, it constructs a connection topology transfer matrix to obtain the convergence degree of topology nodes for each equipment; further, it obtains the weighted center coordinates of the target equipment, performs a concentric square scan, and, combined with the latest grid occupancy state matrix, obtains the target placement area; further, it updates the grid occupancy state matrix; finally, after traversing all equipment, it maps and generates a three-dimensional overall layout scheme for the substation based on the final grid occupancy state matrix and site elevation benchmark data. This invention obtains the topology node convergence degree ordering through connection path constraint coefficients, uses weighted center coordinates for concentric scanning to find the target placement area, refreshes the busbar corridor occupancy, and then maps and generates a three-dimensional overall plan, collaboratively satisfying the inflexibility constraint and anti-obstruction requirement of rigid busbars.

Claims

1. A method for automatically generating a substation general layout scheme, characterized in that, The method includes: Divide the two-dimensional design base map into orthogonal discrete grids and construct the initial grid occupancy state matrix; read the connection path constraint coefficients between the devices to be deployed, and obtain the safe occupation area and traversal index of each device; A connection topology transfer matrix is ​​constructed based on all the connection path constraint coefficients. Iteration is performed using a preset smoothing constant and a uniform distribution vector to obtain the topology node convergence degree of each device to be deployed. Based on the topology node convergence degree, the device's safe occupancy area, and the traversal index, the devices to be deployed are sequentially identified as target devices. Weighted center coordinates are obtained based on the grid coordinates of the calibrated associated devices of the target device and the corresponding connection path constraint coefficients. A concentric square scan is performed based on the weighted center coordinates of the target device, and the target placement area is obtained by combining the latest grid occupancy state matrix. Based on the target placement area, the occupancy refresh mechanism of the bus corridor is activated to update the grid occupancy status matrix; after traversing all devices, a three-dimensional overall layout scheme of the substation is generated based on the final grid occupancy status matrix and site elevation benchmark data.

2. The method for automatically generating a substation general layout scheme according to claim 1, characterized in that, The method for obtaining the convergence degree of the topology nodes includes: Using the connection topology transfer matrix, a preset smoothing constant, and an equal distribution vector, perform state superposition product iteration to obtain the updated node influence distribution vector; calculate the maximum value of the absolute value of the difference between the node influence distribution vectors of adjacent iterations; if the maximum value of the absolute value of the difference is less than the preset convergence tolerance constant, or the number of iterations reaches the preset upper limit of the number of iterations, terminate the iteration, and use the values ​​of each element in the steady-state node influence distribution vector as the topology node convergence degree of the corresponding device to be deployed.

3. The method for automatically generating a substation general layout scheme according to claim 1, characterized in that, The method for obtaining the weighted center coordinates includes: Assign a corresponding path constraint coefficient as a weight to the grid coordinates of all calibrated associated devices of the target device; calculate the weighted average of the weighted grid coordinates and round it to obtain the weighted center coordinates of the shortest connection of the target device, which is then used as the weighted center coordinates.

4. The method for automatically generating a substation general layout scheme according to claim 1, characterized in that, When there are no calibrated associated devices for the target device, the coordinates of the incoming and outgoing ports with the same voltage level as the target device are matched from the pre-acquired set of incoming and outgoing port coordinates. The average value of the matched incoming and outgoing port coordinates is then calculated and rounded down to be used as the weighted center coordinates.

5. The method for automatically generating a substation general layout scheme according to claim 1, characterized in that, The method for obtaining the target placement area includes: A search trajectory of concentric squares is generated by expanding outward from the weighted center coordinates as the origin, and the grid nodes on the search trajectory are extracted as candidate starting points. Based on the candidate starting points, combined with the grid length span and grid width span of the target device, candidate grid blocks are extracted. A virtual connection is constructed based on the candidate center of the target device and the geometric center of the corresponding calibrated associated device. With the grid connection where the virtual connection is located as the center, a preset half-width is expanded outward to both sides along the vertical normal direction to generate the projection grid area of ​​the straight line connection of the bus corridor. When all identifiers of the candidate grid block in the grid occupancy status matrix are found to be free identifiers, and all elements in the projected grid area of ​​the bus corridor straight line connection, except for the grid area occupied by the calibrated associated equipment body at the opposite end, have all identifiers in the grid occupancy status matrix as free identifiers, the candidate grid block is established as the target placement area.

6. The method for automatically generating a substation general layout scheme according to claim 1, characterized in that, The sorting method for the target devices includes: The devices to be deployed are sorted in descending order based on the convergence degree of the topology nodes. If there is a group of devices to be deployed with the same convergence degree of topology nodes in the first sorting result, the devices to be deployed in the group are sorted in descending order based on the safe space occupied by the devices. If there are subgroups of devices with the same safe space occupied by the devices in the second sorting result, then the devices to be arranged in the subgroup are sorted in ascending order for the third time using the traversal index. Based on the final sorting results, a device sorting table is generated, and the devices to be deployed are selected as target devices in sequence according to the order in the device sorting table.

7. The method for automatically generating a substation general layout scheme according to claim 1, characterized in that, The method for updating the grid occupancy state matrix includes: After establishing the target placement area, the elements in the grid occupancy status matrix that belong to the target placement area and the corresponding bus corridor straight line projection grid area are updated from idle identifiers to occupied identifiers to lock the equipment physical area and pipeline laying corridor location.

8. The method for automatically generating a substation general layout scheme according to claim 1, characterized in that, The method for obtaining the connection path constraint coefficient includes: Extract the voltage levels between the equipment to be arranged in the electrical main wiring topology table, call the pre-configured voltage level and non-bendability coefficient mapping table, and generate the connection path constraint coefficient.

9. The method for automatically generating a substation general layout scheme according to claim 5, characterized in that, The method also includes an out-of-bounds anti-dead-loop truncation mechanism when no target placement area is generated: During the search trajectory of expanding outward to generate concentric squares, the current side length span value of the outward expansion is obtained in real time; if the side length span value is greater than the total number of diagonal grid spans of the site obtained in advance, it is determined that the target placement area cannot be generated, the current search operation is terminated and an abnormal site space alarm signal is output.

10. An automatic substation layout scheme generation system, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the automatic generation method for a substation general layout scheme as described in any one of claims 1 to 9.

Images