A device space planning method based on spatial graphical standards

Abstract

The present invention relates to a method for equipment space planning based on spatial graphical standards, and relates to the field of data processing systems or methods for administrative, commercial, financial, management, supervisory or predictive purposes. The method comprises: performing each learning operation on a feedforward neural network to obtain a feedforward neural network model; based on the three-dimensional values ​​of each spatial graphic corresponding to each side of the target equipment in the three-dimensional space and the three-dimensional coordinate values ​​of each point evenly spaced in the three-dimensional space corresponding to the reference internal space of the current cabinet, using the feedforward neural network model to intelligently predict whether the target equipment can be successfully placed in the current cabinet; through the present invention, in view of the technical problems of insufficient speed, efficiency and accuracy of equipment space planning, the feedforward neural network model can be used to intelligently predict whether the target equipment can be successfully placed in the current cabinet after multiple attempts based on multiple spatial graphical standard data, thereby solving the above technical problems.

Description

Technical Field

[0001] The present invention relates to the field of data processing systems or methods for administrative, commercial, financial, management, supervisory or forecasting purposes, and in particular to a device space planning method based on spatial graphical standards. Background Art

[0002] Spatial graphics are an important part of 3D geometry, which describe the shape, size and position of objects in 3D space. These graphics are composed of a series of points and lines, and the connection relationship between them is used to show the appearance and structure of objects. Spatial graphics provide people with an intuitive way to understand and represent 3D objects in the real world. Generally, by digitalizing each physical device based on spatial graphics standards, a reliable analysis of the relative layout and configuration relationship between multiple physical devices can be achieved, thereby helping multiple physical devices to carry out layout planning, effectively reducing space waste and improving the speed, efficiency and accuracy of equipment space planning.

[0003] For example, the invention patent with publication number CN111695174A proposes a BIM-based method for spatial planning of equipment layout in railway four-electric professional cabinets, the steps of which include: collecting cabinet layout specification requirement information; identifying and obtaining relevant information of the cabinet information model, and establishing a cabinet geometric information data architecture based on the railway four-electric BIM standard; according to the geometric information data architecture, formulating a spatial planning scheme for equipment layout in the cabinet to realize intelligent equipment layout. The present invention realizes the intelligent spatial layout of equipment in the cabinet in the railway four-electric engineering based on BIM, realizes the rational and effective use of cabinet space, improves the layout efficiency and the efficiency of creating the engineering data model, and at the same time, the created cabinet information model meets the electrical requirements of the equipment in the design period and the requirements of BIM information detection, evaluation, and certification in the later stage.

[0004] For example, the invention patent with publication number CN115906206A proposes a GIS-based spatial computing device and spatial planning analysis method, including a GIS spatial acquisition unit, a GIS central data processing terminal, a geographic data component acquisition unit, a spatial computing unit, a building area rapid calculation unit, a geographic environment identification unit and a GIS adjustment unit. The GIS spatial acquisition unit is bidirectionally connected to the GIS central data processing terminal, and the geographic data component acquisition unit is bidirectionally connected to the GIS central data processing terminal. The beneficial effects of the present invention are: by adding the GIS spatial acquisition unit, it is possible to locate the land position requiring image acquisition through satellites and drones, and accurately acquire and process the images, thereby improving the efficiency of spatial planning. By adding the geographic environment identification unit, it is possible to identify and mark the positions of hillsides, woodlands, cultivated land and rivers in the collected regional images.

[0005] It can be seen that the various technical solutions in the above-mentioned prior art are unable to pre-judge the accommodation relationship between two physical devices based on spatial graphical standards. For example, before placing a physical device into a current cabinet whose internal space is partially filled, it is impossible to effectively predict whether the physical device can be successfully placed into the current cabinet after multiple attempts. Therefore, it is necessary to actually execute these attempts, wasting a lot of manpower and time costs. More importantly, since it is impossible to immediately find the next cabinet for storage when it is predicted in advance that it will be difficult to place the device into the current cabinet, the speed, efficiency and accuracy of equipment space planning are seriously delayed. Summary of the invention

[0006] In order to solve the technical problems in the prior art, the present invention provides an equipment space planning method based on spatial graphical standards, which can use a customized structured feedforward neural network model to effectively predict whether the physical device can be successfully placed in the current cabinet after multiple attempts based on multiple targeted selected basic data before placing a physical device into a current cabinet whose partial internal space is filled. In this way, it is possible to determine whether the remaining space of the current cabinet can still successfully accommodate the target device after various attempts without actual operation, thereby realizing intelligent planning of equipment space based on spatial graphical standards and improving the speed, efficiency and accuracy of equipment space planning.

[0007] According to the present invention, a method for equipment space planning based on spatial graphical standards is provided, the method comprising: performing a numerical representation of the external structure of the target equipment according to the spatial graphical standards, so as to obtain the three-dimensional numerical values ​​of each spatial graphic corresponding to each side of the target equipment in the three-dimensional space; taking the maximum remaining internal space among the remaining internal spaces of the current cabinet whose partial internal space has been filled as the reference internal space, and obtaining the three-dimensional coordinate numerical values ​​corresponding to each point evenly spaced in the reference internal space in the three-dimensional space; performing a set number of learning operations on the feedforward neural network to obtain the feedforward neural network after completing each learning operation, and outputting the feedforward neural network after completing each learning operation as a feedforward neural network model; calculating the internal volume of the current cabinet in an empty state, the volume of the target equipment, and the volume of the target equipment in the three-dimensional space. The three-dimensional values ​​of each spatial graphic corresponding to each side of the reference internal space and the three-dimensional coordinate values ​​corresponding to each point evenly spaced in the three-dimensional space are input into the feedforward neural network model, and the feedforward neural network model is run to obtain a successful placement mark of whether the target device output by the feedforward neural network model can be successfully placed in the current cabinet whose partial internal space is filled; wherein, the external structure of the target device is numerically represented by spatial graphical standards to obtain the three-dimensional values ​​of each spatial graphic corresponding to each side of the target device in the three-dimensional space, including: after the external structure of the target device is numerically represented by spatial graphical standards, the three-dimensional values ​​of a single spatial graphic corresponding to each side of the target device in the three-dimensional space are the three-dimensional coordinate values ​​corresponding to each point evenly spaced in the side.

[0008] Preferably, performing a set number of learning operations on the feedforward neural network to obtain the feedforward neural network after completing each learning operation, and outputting the feedforward neural network after completing each learning operation as the feedforward neural network model includes: the value of the set number is positively correlated with the internal volume of the current cabinet in an empty state; wherein, after the spatial graphical standard numerical representation of the external structure of the target device is performed, the three-dimensional numerical value of the single spatial graphic corresponding to each side of the target device in the three-dimensional space is obtained as the three-dimensional coordinate numerical values ​​corresponding to each point evenly spaced in the side, including: for the target device, each side of the The three-dimensional coordinate system in which the single three-dimensional coordinate value corresponding to each point is located is a first three-dimensional coordinate system, and the centroid of the target device is used as the origin of the first three-dimensional coordinate system; wherein, the maximum remaining internal space among the remaining internal spaces of the current cabinet whose partial internal space has been filled is used as the reference internal space, and the three-dimensional coordinate values ​​corresponding to each point evenly spaced in the reference internal space are obtained, including: the three-dimensional coordinate system in which the single three-dimensional coordinate value corresponding to each point is located is a second three-dimensional coordinate system, and the centroid of the current cabinet is used as the origin of the second three-dimensional coordinate system.

[0009] Preferably, performing a set number of learning operations on the feedforward neural network to obtain a feedforward neural network after completing each learning operation, and outputting the feedforward neural network after completing each learning operation as a feedforward neural network model includes: in each learning operation performed on the feedforward neural network, using the successful placement mark corresponding to the attempt result of determining whether a certain device can be successfully placed in the current cabinet whose partially internal space is filled after completing the actual placement operation attempt as the output content of the feedforward neural network, using the internal volume of the current cabinet in an empty state, the volume of the certain device, the three-dimensional values ​​of each spatial graphic corresponding to each side of the certain device in the three-dimensional space, and the three-dimensional coordinate values ​​corresponding to each point evenly spaced in the three-dimensional space of the reference internal space before completing the actual placement operation attempt as the input content of the feedforward neural network, to complete this learning. Operation; wherein, in each learning operation performed on the feedforward neural network, a successful placement mark corresponding to the attempt result of determining whether a certain device can be successfully placed in the current cabinet whose partial internal space has been filled after the actual placement operation attempt is completed is used as the output content of the feedforward neural network, and the internal volume of the current cabinet in an empty state, the volume of the certain device, the three-dimensional values ​​of each spatial graphic corresponding to each side of the certain device in the three-dimensional space, and the three-dimensional coordinate values ​​corresponding to each point evenly spaced in the three-dimensional space of the reference internal space before the actual placement operation attempt is completed are used as the input content of the feedforward neural network, and completing this learning operation includes: the reference internal space before the actual placement operation attempt is the maximum remaining internal space among the remaining internal spaces of the current cabinet whose partial internal space has been filled before the actual placement operation attempt is completed.

[0010] Preferably, after the internal volume of the current cabinet in an empty state, the volume of the target device, the three-dimensional values ​​of each spatial graphic corresponding to each side of the target device in the three-dimensional space, and the three-dimensional coordinate values ​​corresponding to each point of the reference internal space evenly spaced in the three-dimensional space are input into the feedforward neural network model, and the feedforward neural network model is run to obtain the successful placement mark of whether the target device can be successfully placed in the current cabinet whose internal space is partially filled, the method also includes: after the successful placement mark of whether the target device can be successfully placed in the current cabinet whose internal space is partially filled, the device number corresponding to the target device and the device number corresponding to the current cabinet are entered into a network data packet, the obtained network data packet is wirelessly sent to a remote device space planning management server through a wireless communication link.

[0011] Preferably, after the internal volume of the current cabinet in an empty state, the volume of the target device, the three-dimensional values ​​of each spatial graphic corresponding to each side of the target device in the three-dimensional space, and the three-dimensional coordinate values ​​corresponding to each point of the reference internal space evenly spaced in the three-dimensional space are input into the feedforward neural network model, and the feedforward neural network model is run to obtain the successful placement mark of whether the target device can be successfully placed in the current cabinet whose partial internal space is filled, the method also includes: using an on-site display device to display in real time a reminder message corresponding to the successful placement mark of whether the target device can be successfully placed in the current cabinet whose partial internal space is filled.

[0012] Preferably, after the internal volume of the current cabinet in an empty state, the volume of the target device, the three-dimensional values ​​of each spatial graphic corresponding to each side of the target device in the three-dimensional space, and each three-dimensional coordinate value corresponding to each point evenly spaced in the reference internal space in the three-dimensional space are input into the feedforward neural network model, and the feedforward neural network model is run to obtain a successful placement mark output by the feedforward neural network model indicating whether the target device can be successfully placed into the current cabinet whose partially internal space is filled, the method further includes: when the successful placement mark of whether the target device can be successfully placed into the current cabinet whose partially internal space is filled indicates that the target device cannot be successfully placed into the current cabinet whose partially internal space is filled, executing an on-site alarm operation corresponding to the target device cannot be successfully placed into the current cabinet whose partially internal space is filled.

[0013] Preferably, after performing a set number of learning operations on the feedforward neural network to obtain the feedforward neural network after completing each learning operation, and outputting the feedforward neural network after completing each learning operation as a feedforward neural network model, the method also includes: completing model storage of the feedforward neural network model by storing various model parameters of the feedforward neural network model.

[0014] Preferably, the largest remaining internal space among the remaining internal spaces of the current cabinet whose partial internal space has been filled is used as the reference internal space, and the three-dimensional coordinate values ​​corresponding to the points of the reference internal space that are evenly spaced in the three-dimensional space are obtained, which also includes: the even spacing takes the first distance as the spacing distance; wherein, after the external structure of the target device is numerically represented by the spatial graphical standard, the three-dimensional value of the single spatial graphic corresponding to each side of the target device in the three-dimensional space is obtained as the three-dimensional coordinate values ​​corresponding to the points evenly spaced in the side, which also includes: the even spacing takes the second distance as the spacing distance, and the second distance is equal to the first distance.

[0015] Preferably, the internal volume of the current cabinet in an empty state, the volume of the target device, the three-dimensional values ​​of each spatial graphic corresponding to each side of the target device in the three-dimensional space, and each three-dimensional coordinate value corresponding to each point of the reference internal space evenly spaced in the three-dimensional space are input into the feedforward neural network model, and the feedforward neural network model is run to obtain a successful placement identifier of whether the target device output by the feedforward neural network model can be successfully placed in the current cabinet whose internal space is partially filled, including: when the obtained successful placement identifier is 0B10, it indicates that the target device output by the feedforward neural network model can be successfully placed in the current cabinet whose internal space is partially filled; Among them, the internal volume of the current cabinet in an empty state, the volume of the target device, the three-dimensional values ​​of each spatial graphic corresponding to each side of the target device in the three-dimensional space, and the three-dimensional coordinate values ​​corresponding to each point of the reference internal space evenly spaced in the three-dimensional space are input into the feedforward neural network model, and the feedforward neural network model is run to obtain a successful placement mark of whether the target device output by the feedforward neural network model can be successfully placed into the current cabinet whose partial internal space is filled. It also includes: when the obtained successful placement mark is 0B11, it indicates that the target device output by the feedforward neural network model cannot be successfully placed into the current cabinet whose partial internal space is filled.

[0016] Preferably, the positive correlation between the value of the set number and the internal volume of the current cabinet in an internal empty state includes: using a numerical conversion function to represent the numerical conversion relationship between the positive correlation between the value of the set number and the internal volume of the current cabinet in an internal empty state; wherein, using a numerical conversion function to represent the numerical conversion relationship between the positive correlation between the value of the set number and the internal volume of the current cabinet in an internal empty state includes: in the numerical conversion function, the internal volume of the current cabinet in an internal empty state is the input parameter of the numerical conversion function; wherein, using a numerical conversion function to represent the numerical conversion relationship between the positive correlation between the value of the set number and the internal volume of the current cabinet in an internal empty state also includes: in the numerical conversion function, the value of the set number corresponding to the internal volume of the current cabinet in an internal empty state is the output parameter of the numerical conversion function.

[0017] It can be seen that the present invention has at least the following outstanding substantial features: First: before each device is placed as a target device into the current cabinet whose partial internal space has been filled, the external structure of the target device is numerically represented according to the spatial graphical standard, and based on the obtained numerical representation and the spatial information of the maximum remaining internal space among the remaining internal spaces of the current cabinet whose partial internal space has been filled, an artificial intelligence mechanism is used to intelligently predict whether the target device can be successfully placed into the current cabinet whose partial internal space has been filled, thereby determining whether the remaining space of the current cabinet can still successfully accommodate the target device after various attempts without actual operation, thereby realizing intelligent planning of equipment space based on spatial graphical standards.

[0018] Second: The artificial intelligence mechanism adopted is based on a feedforward neural network model with a customized structure. The feedforward neural network model is a feedforward neural network after completing each learning operation, and the number of times the feedforward neural network completes the learning operation is positively correlated with the internal volume of the current cabinet when the cabinet is empty. In this way, feedforward neural network models with different structures are customized for different cabinets, ensuring the stability and effectiveness of the intelligent prediction results.

[0019] Third: the internal volume of the current cabinet when it is empty, the volume of the target device, the three-dimensional values ​​of each spatial graphic corresponding to each side of the target device in the three-dimensional space, and the three-dimensional coordinate values ​​corresponding to each point evenly spaced in the three-dimensional space of the reference internal space are used as multiple basic data for executing intelligent prediction of whether the target device can be successfully placed in the current cabinet whose internal space is partially filled. Among them, after the external structure of the target device is numerically represented by spatial graphical standards, the three-dimensional values ​​of a single spatial graphic corresponding to each side of the target device in the three-dimensional space are the three-dimensional coordinate values ​​corresponding to each point evenly spaced in the side. The full and comprehensive screening of the multiple basic data further ensures the stability and effectiveness of the intelligent prediction results.

[0020] Fourthly: in each learning operation performed on the feedforward neural network, a successful placement mark corresponding to the attempt result of determining whether a certain device can be successfully placed into the current cabinet whose partially internal space is filled after the actual placement operation is completed is used as the output content of the feedforward neural network, and the internal volume of the current cabinet in an empty state, the volume of the certain device, the three-dimensional values ​​of each spatial graphic corresponding to each side of the certain device in the three-dimensional space, and the three-dimensional coordinate values ​​corresponding to each point evenly spaced in the three-dimensional space of the reference internal space before the actual placement operation is completed are used as the input content of the feedforward neural network to complete this learning operation, thereby ensuring the learning effect of each learning operation of the feedforward neural network. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The embodiments of the present invention will be described below with reference to the accompanying drawings.

[0022] Figure 1 The technical flow chart of a device space planning method based on spatial graphical standards according to the present invention.

[0023] Figure 2 The present invention is a flowchart showing a method for equipment space planning based on spatial graphical standards according to a first embodiment of the present invention.

[0024] Figure 3 The figure is a flowchart showing the steps of a method for equipment space planning based on spatial graphical standards according to a second embodiment of the present invention.

[0025] Figure 4 The present invention is a flowchart showing a method for equipment space planning based on spatial graphical standards according to a third embodiment of the present invention.

[0026] Figure 5 The fourth embodiment of the present invention is a flowchart showing a method for equipment space planning based on spatial graphical standards.

[0027] Figure 6 The fifth embodiment of the present invention is a flowchart showing a method for equipment space planning based on spatial graphical standards. DETAILED DESCRIPTION

[0028] like Figure 1 As shown, a technical flow chart of a device space planning method based on spatial graphical standards according to the present invention is given.

[0029] The present invention is used in an equipment space planning scenario in which a target device is to be placed in a current cabinet whose internal space is partially filled. In the prior art, placement is completed by performing physical operations, that is, by continuously adjusting the placement angle of the target device to try whether the target device can be completely placed in the current cabinet whose internal space is partially filled. Obviously, the above-mentioned physical operations are complex, cumbersome and time-consuming.

[0030] like Figure 1As shown, the specific technical process of the present invention is as follows: Technical process one: for the equipment space planning scenario in which the target equipment is to be placed in the current cabinet whose partial internal space has been filled, the external structure of the target equipment is numerically represented in a spatial graphical standard to obtain the three-dimensional values ​​of each spatial graphic corresponding to each side of the target equipment in the three-dimensional space, and at the same time, the maximum remaining internal space in each remaining internal space of the current cabinet, that is, the three-dimensional coordinate values ​​corresponding to each point evenly spaced in the three-dimensional space of the reference internal space, are obtained, so as to obtain a plurality of basic data for predicting whether the target equipment can be successfully placed in the current cabinet; Figure 1 As shown, the appearance structure of the target device is numerically represented by a spatial graphical standard to obtain the three-dimensional values ​​of each spatial graphic corresponding to each side of the target device in the three-dimensional space, that is, to obtain basic data 1; at the same time, as Figure 1 As shown, the maximum remaining internal space among the remaining internal spaces of the current cabinet is obtained, that is, the three-dimensional coordinate values ​​corresponding to each point evenly spaced in the three-dimensional space of the reference internal space are respectively obtained, that is, basic data 2 is obtained; in addition, the basic data used to predict whether the target device can be successfully placed in the current cabinet also include the internal volume of the current cabinet when it is empty and the volume of the target device.

[0031] Technical process 2: Design a feedforward neural network model with a customized structure, wherein the feedforward neural network model is a feedforward neural network after completing each learning operation; specifically, the structural customization of the feedforward neural network model is mainly reflected in the following aspects: First, the number of times the feedforward neural network completes the learning operation is positively correlated with the internal volume of the current cabinet in an empty state, thereby customizing feedforward neural network models with different structures for different cabinets, thereby ensuring the stability and effectiveness of the intelligent prediction results; Second, in each learning operation performed on the feedforward neural network, the successful placement mark corresponding to the attempt result of determining whether a certain device can be successfully placed in the current cabinet whose partially internal space is filled after completing the actual placement operation attempt is used as the output content of the feedforward neural network, and the internal volume of the current cabinet in an empty state, the volume of the certain device, the three-dimensional values ​​of each spatial graphic corresponding to each side of the certain device in the three-dimensional space, and the three-dimensional coordinate values ​​corresponding to each point evenly spaced in the three-dimensional space of the reference internal space before completing the actual placement operation attempt are used as the input content of the feedforward neural network to complete this learning operation, thereby ensuring the learning effect of each learning operation of the feedforward neural network.

[0032] Technical process three: The feedforward neural network model with customized structure design of technical process two intelligently predicts whether the target device can be successfully placed into the current cabinet with partially filled internal space if the entity attempts at various angles based on the basic data targeted by technical process one; obviously, this is a technical difficulty and also the most outstanding substantive feature of the present invention. The present invention is not facing an empty cabinet, but a current cabinet with partially filled internal space. For target devices that may have different three-dimensional shapes, how to effectively evaluate whether the target device can be completely placed into the current cabinet with partially filled internal space without performing various angle placement attempts of the entity is the task to be completed by technical process three.

[0033] Technical process four: Based on the successful placement mark of whether the target device can be successfully placed in the current cabinet whose internal space is partially filled, the intelligent prediction result is remotely sent, displayed on site or alarmed in real time; this technical process is an auxiliary process of the present invention, which is an auxiliary and application of the main process composed of technical process one, technical process two and technical process three. In this way, facing the current cabinet with a complex internal environment with partial internal space filled, the artificial intelligence mechanism based on spatial graphical standard data can be used to predict in advance whether the target device can be completely placed in the current cabinet. If the advance prediction fails, there is no need to try to place the entity at various angles. The target device can be turned to other cabinets with remaining space to continue the intelligent prediction based on the artificial intelligence mechanism of spatial graphical standard data, thereby improving the speed, efficiency and accuracy of equipment space planning.

[0034] The key points of the present invention are: a feedforward neural network model with a customized structure introduced into the current cabinet in the face of a complex internal environment in which part of the internal space has been filled, data transformation of the three-dimensional solid shape based on spatial graphical standards to obtain the basic data required for intelligent prediction of whether the target equipment can be successfully placed in the current mechanism, and targeted design of each learning operation of the feedforward neural network to ensure the learning effect of a single learning operation.

[0035] Next, a method for equipment space planning based on spatial graphical standards of the present invention will be specifically described in the form of an embodiment.

[0036] First embodiment

[0037] Figure 2 The present invention is a flowchart showing a method for equipment space planning based on spatial graphical standards according to a first embodiment of the present invention.

[0038] like Figure 2As shown, the device space planning method based on the spatial graphical standard includes the following steps: Step S101: numerically represent the external structure of the target device according to the spatial graphical standard to obtain the three-dimensional numerical values ​​of each spatial graphic corresponding to each side of the target device in the three-dimensional space; illustratively, the numerically represent the external structure of the target device according to the spatial graphical standard to obtain the three-dimensional numerical values ​​of each spatial graphic corresponding to each side of the target device in the three-dimensional space includes: in the three-dimensional space, the three-dimensional data of the point is used to represent the horizontal coordinate value, the vertical coordinate value, and the vertical coordinate value in the three-dimensional coordinate system; accordingly, the three-dimensional of the spatial graphic The values ​​can be represented by the three-dimensional data corresponding to the points constituting the spatial graph; step S102: taking the largest remaining internal space among the remaining internal spaces of the current cabinet whose partial internal space has been filled as the reference internal space, and obtaining the three-dimensional coordinate values ​​corresponding to the points evenly spaced in the three-dimensional space of the reference internal space; for example, taking the largest remaining internal space among the remaining internal spaces of the current cabinet whose partial internal space has been filled as the reference internal space, and obtaining the three-dimensional coordinate values ​​corresponding to the points evenly spaced in the three-dimensional space of the reference internal space include: among the points evenly spaced in the three-dimensional space, the distance between two adjacent points The distance between them is fixed and is a fixed three-dimensional distance; step S103: performing a set number of learning operations on the feedforward neural network to obtain a feedforward neural network after completing each learning operation, and outputting the feedforward neural network after completing each learning operation as a feedforward neural network model; specifically, performing a set number of learning operations on the feedforward neural network to obtain a feedforward neural network after completing each learning operation, and outputting the feedforward neural network after completing each learning operation as a feedforward neural network model includes: selecting a numerical simulation mechanism to complete performing a set number of learning operations on the feedforward neural network to obtain a feedforward neural network after completing each learning operation, and outputting the feedforward neural network after completing each learning operation as a feedforward neural network model. The feedforward neural network after each learning operation is used as the simulation and test of the data processing process output by the feedforward neural network model; Step S104: the internal volume of the current cabinet in an empty state, the volume of the target device, the three-dimensional values ​​of the spatial graphics corresponding to the sides of the target device in the three-dimensional space, and the three-dimensional coordinate values ​​corresponding to the points of the reference internal space evenly spaced in the three-dimensional space are input into the feedforward neural network model, and the feedforward neural network model is run to obtain a successful placement mark of whether the target device output by the feedforward neural network model can be successfully placed in the current cabinet whose internal space is partially filled;For example, the internal volume of the current cabinet in an empty state, the volume of the target device, the three-dimensional values ​​of each spatial graphic corresponding to each side of the target device in the three-dimensional space, and the three-dimensional coordinate values ​​corresponding to each point of the reference internal space evenly spaced in the three-dimensional space are input into the feedforward neural network model, and the feedforward neural network model is run to obtain a successful placement identifier output by the feedforward neural network model to determine whether the target device can be successfully placed in the current cabinet whose internal space is partially filled. The successful placement identifier output by the feedforward neural network model has only two values ​​to choose from, and the two values ​​represent the target device. The device can be successfully placed in the current cabinet whose internal space is partially filled, and the target device cannot be successfully placed in the current cabinet whose internal space is partially filled; wherein, the external structure of the target device is numerically represented by a spatial graphical standard to obtain the three-dimensional values ​​of each spatial graphic corresponding to each side of the target device in the three-dimensional space, including: after the external structure of the target device is numerically represented by a spatial graphical standard, the three-dimensional values ​​of the single spatial graphic corresponding to each side of the target device in the three-dimensional space obtained are the three-dimensional coordinate values ​​corresponding to each point evenly spaced in the side; wherein, a set number of learning operations are performed on the feedforward neural network to obtain a complete The feedforward neural network after each learning operation is formed, and the feedforward neural network after each learning operation is output as a feedforward neural network model, including: the value of the set number is positively correlated with the internal volume of the current cabinet when the cabinet is in an empty state; for example, the value of the set number is positively correlated with the internal volume of the current cabinet when the cabinet is in an empty state, including: the internal volume of the current cabinet when the cabinet is in an empty state is 1 cubic meter, the value of the set number is 160, the internal volume of the current cabinet when the cabinet is in an empty state is 1.25 cubic meters, the value of the set number is 180, the internal volume of the current cabinet when the cabinet is in an empty state is 1.5 cubic meters. , the value of the set number is 200, the internal volume of the current cabinet in an empty state is 2.0 cubic meters, the value of the set number is 300, and so on; wherein, after the spatial graphical standard numerical representation of the external structure of the target device is performed, the three-dimensional numerical value of the single spatial graphic corresponding to each side of the target device in the three-dimensional space is obtained as the three-dimensional coordinate numerical values ​​corresponding to each point evenly spaced in the side, including: for the target device, the three-dimensional coordinate system in which the single three-dimensional coordinate numerical value corresponding to each point in each side is located is the first three-dimensional coordinate system, and the first three-dimensional coordinate system takes the centroid of the target device as the origin of the first three-dimensional coordinate system;Specifically, for the target device, the three-dimensional coordinate system in which the single three-dimensional coordinate value corresponding to each point in each side thereof is located is the first three-dimensional coordinate system, and the first three-dimensional coordinate system takes the centroid of the target device as the origin of the first three-dimensional coordinate system, including: since the directions of the three coordinate axes of the three-dimensional coordinate system are fixed, the origin of the three-dimensional coordinate system can directly determine the entire three-dimensional coordinate system, that is, the difference between different three-dimensional coordinate systems is only the position of the origin of the three-dimensional coordinate system; wherein, the maximum remaining internal space among the remaining internal spaces of the current cabinet whose partial internal space has been filled is taken as the reference internal space, and each point of the reference internal space evenly spaced in the three-dimensional space is obtained respectively The corresponding three-dimensional coordinate values ​​include: the three-dimensional coordinate system where the single three-dimensional coordinate value corresponding to each point is located is a second three-dimensional coordinate system, and the second three-dimensional coordinate system takes the centroid of the current cabinet as the origin of the second three-dimensional coordinate system; wherein, performing a set number of learning operations on the feedforward neural network to obtain the feedforward neural network after completing each learning operation, and outputting the feedforward neural network after completing each learning operation as a feedforward neural network model includes: in each learning operation performed on the feedforward neural network, completing the actual placement operation attempt to determine whether a certain device can be successfully placed in the current cabinet whose internal space is partially filled. The result of the attempt corresponds to the successful placement The input mark is used as the output content of the feedforward neural network, and the internal volume of the current cabinet in an empty state, the volume of the certain device, the three-dimensional values ​​of each space graphic corresponding to each side of the certain device in the three-dimensional space, and the three-dimensional coordinate values ​​corresponding to each point evenly spaced in the three-dimensional space of the reference internal space before the actual placement operation attempt is completed are used as the input content of the feedforward neural network to complete this learning operation; and wherein, in each learning operation performed on the feedforward neural network, the attempt result of determining whether a certain device can be successfully placed in the current cabinet whose internal space is partially filled after the actual placement operation attempt is completed. The corresponding successful placement mark is used as the output content of the feedforward neural network, and the internal volume of the current cabinet in an empty state, the volume of the certain device, the three-dimensional values ​​of each space figure corresponding to each side of the certain device in the three-dimensional space, and the three-dimensional coordinate values ​​corresponding to each point evenly spaced in the three-dimensional space of the reference internal space before the actual placement operation attempt is completed are used as the input content of the feedforward neural network. The completion of this learning operation includes: the reference internal space before the actual placement operation attempt is the maximum remaining internal space of each remaining internal space of the current cabinet whose partial internal space is filled before the actual placement operation attempt is completed. ;

[0039] Second embodiment

[0040] Figure 3The figure is a flowchart showing the steps of a method for equipment space planning based on spatial graphical standards according to a second embodiment of the present invention.

[0041] like Figure 3 As shown, compared with Figure 2 , after step S104, the method also includes: step S105: after the successful placement identifier of whether the target device can be successfully placed into the current cabinet whose partial internal space has been filled, the device number corresponding to the target device and the device number corresponding to the current cabinet are entered into the network data packet, the obtained network data packet is wirelessly sent to the remote device space planning management server through the wireless communication link; specifically, after the successful placement identifier of whether the target device can be successfully placed into the current cabinet whose partial internal space has been filled, the device number corresponding to the target device and the device number corresponding to the current cabinet are entered into the network data packet, the obtained network data packet is wirelessly sent to the remote device space planning management server through the wireless communication link, including: the remote device space planning management server is a big data server, a cloud computing server or a blockchain server.

[0042] Third embodiment

[0043] Figure 4 The present invention is a flowchart showing a method for equipment space planning based on spatial graphical standards according to a third embodiment of the present invention.

[0044] like Figure 4 As shown, compared with Figure 2 , after step S104, the method further includes: step S106: using an on-site display device to display in real time a reminder message corresponding to a successful placement mark of whether the target device can be successfully placed into the current cabinet whose partial internal space has been filled; specifically, using an on-site display device to display in real time a reminder message corresponding to a successful placement mark of whether the target device can be successfully placed into the current cabinet whose partial internal space has been filled includes: the on-site display device is an LED display matrix, an LCD display matrix or a liquid crystal display device.

[0045] Fourth embodiment

[0046] Figure 5 The fourth embodiment of the present invention is a flowchart showing a method for equipment space planning based on spatial graphical standards.

[0047] like Figure 5 As shown, compared with Figure 2, after step S104, the method also includes: step S107: when the successful placement mark of whether the target device can be successfully placed into the current cabinet whose partial internal space is filled indicates that the target device cannot be successfully placed into the current cabinet whose partial internal space is filled, execute the on-site alarm operation corresponding to the target device cannot be successfully placed into the current cabinet whose partial internal space is filled; specifically, when the successful placement mark of whether the target device can be successfully placed into the current cabinet whose partial internal space is filled indicates that the target device cannot be successfully placed into the current cabinet whose partial internal space is filled, execute the on-site alarm operation corresponding to the target device cannot be successfully placed into the current cabinet whose partial internal space is filled, including: can choose to use acoustic alarm equipment or optical alarm equipment to execute the on-site alarm operation corresponding to the target device cannot be successfully placed into the current cabinet whose partial internal space is filled.

[0048] Fifth embodiment

[0049] Figure 6 The fifth embodiment of the present invention is a flowchart showing a method for equipment space planning based on spatial graphical standards.

[0050] like Figure 6 As shown, compared with Figure 2 , after step S103, the method further includes: step S108: completing the model storage of the feedforward neural network model by storing various model parameters of the feedforward neural network model; for example, a TF storage device, a FLASH storage device or an MMC storage device can be selected to complete the model storage of the feedforward neural network model by storing various model parameters of the feedforward neural network model.

[0051] Next, various embodiments of the present invention will be further described.

[0052] In each of the above embodiments, optionally, in the equipment space planning method based on the spatial graphics standard: the maximum remaining internal space among the remaining internal spaces of the current cabinet whose partial internal space has been filled is used as the reference internal space, and the three-dimensional coordinate values ​​corresponding to each point evenly spaced in the three-dimensional space of the reference internal space are obtained, which also includes: the uniform spacing uses a first distance as the spacing distance; wherein, after the external structure of the target device is numerically represented by the spatial graphics standard, the three-dimensional value of the single spatial graphic corresponding to each side of the target device in the three-dimensional space is obtained. The three-dimensional coordinate values ​​corresponding to each point evenly spaced in the side also include: the uniform spacing uses a second distance as the spacing distance, and the second distance is equal to the first distance.

[0053] In each of the above embodiments, optionally, in the equipment space planning method based on the spatial graphical standard: the internal volume of the current cabinet in an empty state, the volume of the target equipment, the three-dimensional values ​​of each spatial graphic corresponding to each side of the target equipment in the three-dimensional space, and the three-dimensional coordinate values ​​corresponding to each point of the reference internal space evenly spaced in the three-dimensional space are input into the feedforward neural network model, and the feedforward neural network model is run to obtain a successful placement identifier of whether the target equipment output by the feedforward neural network model can be successfully placed in the current cabinet whose internal space is partially filled, including: when the obtained successful placement identifier is 0B10, the target equipment output by the feedforward neural network model can be successfully placed. Place it into the current cabinet whose internal space is partially filled; wherein, the internal volume of the current cabinet in an empty state, the volume of the target device, the three-dimensional values ​​of each spatial graphic corresponding to each side of the target device in the three-dimensional space, and the three-dimensional coordinate values ​​corresponding to each point of the reference internal space evenly spaced in the three-dimensional space are input into the feedforward neural network model, and the feedforward neural network model is run to obtain a successful placement identifier of whether the target device output by the feedforward neural network model can be successfully placed into the current cabinet whose internal space is partially filled. It also includes: when the obtained successful placement identifier is 0B11, it indicates that the target device output by the feedforward neural network model cannot be successfully placed into the current cabinet whose internal space is partially filled.

[0054] In each of the above embodiments, optionally, in the equipment space planning method based on the spatial graphical standard: the value of the set number is positively associated with the internal volume of the current cabinet in an internal empty state, including: using a numerical conversion function to represent the numerical conversion relationship of the positive association between the value of the set number and the internal volume of the current cabinet in an internal empty state; specifically, using a numerical conversion function to represent the numerical conversion relationship of the positive association between the value of the set number and the internal volume of the current cabinet in an internal empty state includes: the operation process of the numerical conversion function can be simulated by using the MATLAB toolbox; wherein , the numerical conversion relationship in which a numerical conversion function is used to represent the positive correlation between the value of the set number and the internal volume of the current cabinet in an internal empty state includes: in the numerical conversion function, the internal volume of the current cabinet in an internal empty state is an input parameter of the numerical conversion function; and wherein the numerical conversion relationship in which a numerical conversion function is used to represent the positive correlation between the value of the set number and the internal volume of the current cabinet in an internal empty state also includes: in the numerical conversion function, the value of the set number corresponding to the internal volume of the current cabinet in an internal empty state is an output parameter of the numerical conversion function.

[0055] In addition, in an equipment space planning method based on spatial graphical standards according to the present invention: the three-dimensional coordinate system in which the single three-dimensional coordinate value corresponding to each point is located is a second three-dimensional coordinate system, and the second three-dimensional coordinate system takes the centroid of the current cabinet as the origin of the second three-dimensional coordinate system, including: for the current cabinet, its centroid is the intersection of all hyperplanes that divide the current cabinet into two parts with equal moments in the three-dimensional space; for example, the MATLAB toolbox can be selected to complete the simulation and testing of the analytical process of the centroid of each physical device in the three-dimensional space, that is, the geometric center of each physical device in the three-dimensional space; and wherein, for the target device, the three-dimensional coordinate system in which the single three-dimensional coordinate value corresponding to each point in each side thereof is located is a first three-dimensional coordinate system, and the first three-dimensional coordinate system takes the centroid of the target device as the origin of the first three-dimensional coordinate system, including: for the target device, its centroid is the intersection of all hyperplanes that divide the target device into two parts with equal moments in the three-dimensional space.

[0056] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in the field can change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A device space planning method based on spatial graphical standards, characterized in that: The method comprises: The external structure of the target device is numerically represented in a spatial graphical standard to obtain three-dimensional numerical values ​​of each spatial graphic corresponding to each side of the target device in the three-dimensional space, wherein after the external structure of the target device is numerically represented in a spatial graphical standard, the three-dimensional numerical values ​​of a single spatial graphic corresponding to each side of the target device in the three-dimensional space obtained are the three-dimensional coordinate numerical values ​​corresponding to each point evenly spaced in the side; The largest remaining internal space among the remaining internal spaces of the current cabinet whose internal spaces have been partially filled is used as the reference internal space, and the three-dimensional coordinate values ​​corresponding to the points of the reference internal space evenly spaced in the three-dimensional space are obtained; Performing a set number of learning operations on the feedforward neural network to obtain a feedforward neural network after completing each learning operation, and outputting the feedforward neural network after completing each learning operation as a feedforward neural network model, wherein the value of the set number is positively correlated with the internal volume of the current cabinet in an empty state; The internal volume of the current cabinet when it is empty, the volume of the target device, the three-dimensional values ​​of each spatial graphic corresponding to each side of the target device in the three-dimensional space, and the three-dimensional coordinate values ​​corresponding to each point of the reference internal space evenly spaced in the three-dimensional space are input into the feedforward neural network model, and the feedforward neural network model is run to obtain a successful placement mark output by the feedforward neural network model indicating whether the target device can be successfully placed into the current cabinet whose partial internal space is filled.

2. The device space planning method based on spatial graphical standards according to claim 1, characterized in that: After the appearance structure of the target device is numerically represented in a spatial graphical standard, the three-dimensional values ​​of the single spatial graphics corresponding to each side of the target device in the three-dimensional space are obtained as the three-dimensional coordinate values ​​corresponding to each point evenly spaced in the side, including: for the target device, the three-dimensional coordinate system where the single three-dimensional coordinate values ​​corresponding to each point in each side is located is a first three-dimensional coordinate system, and the first three-dimensional coordinate system takes the centroid of the target device as the origin of the first three-dimensional coordinate system; Among them, the maximum remaining internal space among the remaining internal spaces of the current cabinet whose partial internal space has been filled is used as the reference internal space, and the three-dimensional coordinate values ​​corresponding to the points evenly spaced in the reference internal space are obtained, including: the three-dimensional coordinate system in which the single three-dimensional coordinate value corresponding to each point is located is the second three-dimensional coordinate system, and the second three-dimensional coordinate system takes the centroid of the current cabinet as the origin of the second three-dimensional coordinate system.

3. The device space planning method based on spatial graphical standards according to claim 2, characterized in that: Performing a set number of learning operations on the feedforward neural network to obtain the feedforward neural network after completing each learning operation, and outputting the feedforward neural network after completing each learning operation as a feedforward neural network model includes: in each learning operation performed on the feedforward neural network, using the successful placement mark corresponding to the result of the attempt to determine whether a certain device can be successfully placed in the current cabinet whose internal space is partially filled after completing the actual placement operation attempt as the output content of the feedforward neural network, using the internal volume of the current cabinet in an empty state, the volume of the certain device, the three-dimensional values ​​of each space graphic corresponding to each side of the certain device in the three-dimensional space, and the three-dimensional coordinate values ​​corresponding to each point evenly spaced in the three-dimensional space of the reference internal space before completing the actual placement operation attempt as the input content of the feedforward neural network to complete this learning operation; Among them, in each learning operation performed on the feedforward neural network, the successful placement mark corresponding to the attempt result of determining whether a certain device can be successfully placed in the current cabinet whose partial internal space is filled after the actual placement operation attempt is completed is used as the output content of the feedforward neural network, and the internal volume of the current cabinet in an empty state, the volume of the certain device, the three-dimensional values ​​of each spatial graphic corresponding to each side of the certain device in the three-dimensional space, and the three-dimensional coordinate values ​​corresponding to each point evenly spaced in the three-dimensional space of the reference internal space before the actual placement operation attempt is completed are used as the input content of the feedforward neural network. Completing this learning operation includes: the reference internal space before the actual placement operation attempt is the maximum remaining internal space among the remaining internal spaces of the current cabinet whose partial internal space is filled before the actual placement operation attempt is completed.

4. The device space planning method based on spatial graphical standards according to claim 3, characterized in that: After inputting the internal volume of the current cabinet in an empty state, the volume of the target device, the three-dimensional values ​​of each spatial graphic corresponding to each side of the target device in the three-dimensional space, and each three-dimensional coordinate value corresponding to each point of the reference internal space evenly spaced in the three-dimensional space into the feedforward neural network model, and running the feedforward neural network model to obtain a successful placement mark output by the feedforward neural network model indicating whether the target device can be successfully placed in the current cabinet whose internal space is partially filled, the method further includes: After the successful placement identifier of whether the target device can be successfully placed into the current cabinet whose internal space is partially filled, the device number corresponding to the target device and the device number corresponding to the current cabinet are entered into the network data packet, the obtained network data packet is wirelessly sent to the remote device space planning management server via a wireless communication link.

5. The device space planning method based on spatial graphical standards according to claim 3, characterized in that: After inputting the internal volume of the current cabinet in an empty state, the volume of the target device, the three-dimensional values ​​of each spatial graphic corresponding to each side of the target device in the three-dimensional space, and each three-dimensional coordinate value corresponding to each point of the reference internal space evenly spaced in the three-dimensional space into the feedforward neural network model, and running the feedforward neural network model to obtain a successful placement mark output by the feedforward neural network model indicating whether the target device can be successfully placed in the current cabinet whose internal space is partially filled, the method further includes: An on-site display device is used to display in real time reminder information corresponding to a successful placement mark of whether the target device can be successfully placed into the current cabinet whose internal space is partially filled.

6. The device space planning method based on spatial graphical standards according to claim 3, characterized in that: After inputting the internal volume of the current cabinet in an empty state, the volume of the target device, the three-dimensional values ​​of each spatial graphic corresponding to each side of the target device in the three-dimensional space, and each three-dimensional coordinate value corresponding to each point of the reference internal space evenly spaced in the three-dimensional space into the feedforward neural network model, and running the feedforward neural network model to obtain a successful placement mark output by the feedforward neural network model indicating whether the target device can be successfully placed in the current cabinet whose internal space is partially filled, the method further includes: When the successful placement indicator of whether the target device can be successfully placed into the current cabinet with partially filled internal space indicates that the target device cannot be successfully placed into the current cabinet with partially filled internal space, execute the on-site alarm operation corresponding to the target device cannot be successfully placed into the current cabinet with partially filled internal space.

7. The device space planning method based on spatial graphical standards according to claim 3, characterized in that: After performing a set number of learning operations on the feedforward neural network to obtain the feedforward neural network after completing each learning operation, and outputting the feedforward neural network after completing each learning operation as a feedforward neural network model, the method further includes: The model storage of the feedforward neural network model is completed by storing various model parameters of the feedforward neural network model.

8. The device space planning method based on spatial graphical standards according to any one of claims 3 to 7, characterized in that: Taking the largest remaining internal space of each remaining internal space of the current cabinet whose internal space has been partially filled as the reference internal space, obtaining the three-dimensional coordinate values ​​corresponding to each point of the reference internal space evenly spaced in the three-dimensional space also includes: the even spacing takes the first distance as the spacing distance; Among them, after the spatial graphical standard numerical representation of the external structure of the target device is performed, the three-dimensional numerical value of a single spatial graphic corresponding to each side of the target device in the three-dimensional space is obtained, and the three-dimensional coordinate values ​​corresponding to each point evenly spaced in the side are also included: the evenly spaced interval is a second distance, and the second distance is equal to the first distance.

9. The device space planning method based on spatial graphical standards according to any one of claims 3 to 7, characterized in that: Inputting the internal volume of the current cabinet in an empty state, the volume of the target device, the three-dimensional values ​​of each spatial graphic corresponding to each side of the target device in the three-dimensional space, and each three-dimensional coordinate value corresponding to each point of the reference internal space evenly spaced in the three-dimensional space into the feedforward neural network model, and running the feedforward neural network model to obtain a successful placement identifier of whether the target device output by the feedforward neural network model can be successfully placed in the current cabinet whose internal space is partially filled, including: when the obtained successful placement identifier is 0B10, it indicates that the target device output by the feedforward neural network model can be successfully placed in the current cabinet whose internal space is partially filled; Among them, the internal volume of the current cabinet in an empty state, the volume of the target device, the three-dimensional values ​​of each spatial graphic corresponding to each side of the target device in the three-dimensional space, and the three-dimensional coordinate values ​​corresponding to each point of the reference internal space evenly spaced in the three-dimensional space are input into the feedforward neural network model, and the feedforward neural network model is run to obtain a successful placement mark of whether the target device output by the feedforward neural network model can be successfully placed into the current cabinet whose partial internal space is filled. It also includes: when the obtained successful placement mark is 0B11, it indicates that the target device output by the feedforward neural network model cannot be successfully placed into the current cabinet whose partial internal space is filled.

10. The device space planning method based on spatial graphical standards according to any one of claims 3 to 7, characterized in that: The positive correlation between the value of the set number and the internal volume of the current cabinet in an empty state includes: using a numerical conversion function to represent the numerical conversion relationship between the value of the set number and the internal volume of the current cabinet in an empty state; Wherein, using a numerical conversion function to represent the numerical conversion relationship of the positive correlation between the value of the set number and the internal volume of the current cabinet in an internal empty state includes: in the numerical conversion function, the internal volume of the current cabinet in an internal empty state is an input parameter of the numerical conversion function; Among them, the numerical conversion relationship in which a numerical conversion function is used to represent the positive correlation between the value of the set quantity and the internal volume of the current cabinet when the cabinet is in an empty state also includes: in the numerical conversion function, the value of the set quantity corresponding to the internal volume of the current cabinet when the cabinet is in an empty state is the output parameter of the numerical conversion function.

Images