An interactive factory management system based on 3D visualization technology

Through an interactive factory management system based on three-dimensional visualization technology, the problem that traditional factory monitoring systems cannot effectively express the logical relationship of physical space is solved, low threshold, efficient factory supervision and accurate abnormal positioning are achieved, and the efficiency of factory production management is improved. and accuracy.

CN117764541BActive Publication Date: 2024-12-03BEIJING UNITED MEDIA TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410195311.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-02-22
Publication Date
2024-12-03
Estimated Expiration
2044-02-22

AI Technical Summary

Technical Problem

Traditional factory monitoring systems cannot effectively express the physical spatial logical relationship between assets and equipment, resulting in high understanding thresholds, high learning costs, low efficiency in handling accidents, and the existing three-dimensional three-dimensional factory systems use Unity3D technology to have high equipment requirements and limitations. It is difficult to improve the efficiency and accuracy of factory visual supervision.

Method used

Adopting an interactive factory management system based on three-dimensional visualization technology, including cloud computing platform, factory data acquisition module, virtual scene establishment module and factory visualization module, the component bone model and appearance model are generated through the architectural data acquisition unit and appearance data acquisition unit to generate the component bone model and appearance model, and perform the Block mapping and pixel color rendering, establish space-time visualization space, judge production line exceptions and generate maintenance decisions.

Benefits of technology

It lowers the threshold for supervision learning, improves the efficiency of handling accidents, accurately locates production abnormalities, realizes timely maintenance and repairs, and improves the efficiency and accuracy of factory visual supervision.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117764541B_ABST
    Figure CN117764541B_ABST
Patent Text Reader

Abstract

The present invention discloses an interactive factory management system based on three-dimensional visualization technology, relates to the field of visualization management technology, and effectively improves the accuracy of factory visualization supervision; the present invention generates a component skeleton model and a component appearance model through the architecture image data and appearance image data of each component in the factory, maps the component skeleton model and the component appearance model of the component in blocks, and generates a block three-dimensional image model according to the mapping results of the two, maps the block three-dimensional image model on a three-dimensional plane coordinate system and performs pixel color rendering, splices the block pixel models that have completed pixel color rendering in turn to obtain a factory visualization model, establishes normal operating criteria for each block three-dimensional model, judges whether each block three-dimensional image model is operating normally according to the normal operating criteria, and judges whether production is abnormal according to the judgment result.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of visual management, in particular to an interactive factory management system based on three-dimensional visual technology. Background Art

[0002] With the development of the economy, the continuous advancement of factory informatization and automation construction, the deployment and application of various information systems and automation systems have made workshop management as a product production increasingly important and important, with increasingly higher requirements for collaborative work, and increasingly higher requirements for security, availability, and operation and maintenance management.

[0003] At the same time, the various production monitoring equipment purchased by the factory production workshop is provided by different manufacturers. This is the result of the continuous evolution of the factory's informatization, automation, and intelligent construction. Due to the lack of unified planning and other constraints; the traditional monitoring system cannot well express the physical space logical relationship between assets and assets, equipment and equipment, and the abstract data statistics are boring and difficult to understand, resulting in high barriers to understanding, high learning costs, and low efficiency in handling accidents.

[0004] Currently, Unity3D technology is relatively mature for the development of three-dimensional factory systems. However, using Unity3D technology to build a visual editor meets the needs of users to build different three-dimensional scenes. However, the system built with this technology requires downloading plug-ins or has high requirements for equipment, which has certain limitations.

[0005] Therefore, while improving the efficiency of factory visualization supervision, improving the accuracy of supervision is a difficulty in the existing technology. For this purpose, an interactive factory management system based on three-dimensional visualization technology is provided. Summary of the invention

[0006] In order to solve the above technical problems, the purpose of the present invention is to provide an interactive factory management system based on three-dimensional visualization technology.

[0007] In order to achieve the above object, the present invention provides the following technical solutions:

[0008] An interactive factory management system based on three-dimensional visualization technology, comprising a cloud computing platform, wherein the cloud computing platform is communicatively connected with a factory data acquisition module, a virtual scene establishment module and a factory visualization module;

[0009] The factory data acquisition module is provided with an architecture data acquisition unit, an appearance data acquisition unit and a scene positioning device;

[0010] The scene positioning device is composed of a positioning device, a wireless signal device and a system clock, wherein the system clock is provided with data collection instructions and time synchronization instructions;

[0011] The architecture data acquisition unit is provided with a plurality of perspective projection cameras for acquiring architecture image data of each component in the factory;

[0012] The appearance data acquisition unit is provided with a plurality of orthogonal projection cameras for acquiring appearance image data of each component in the factory;

[0013] The virtual scene building module is provided with a component skeleton model unit, a component appearance model unit and a factory visualization model unit;

[0014] The component skeleton model unit is used to establish a corresponding component skeleton model according to the architecture image data of the component;

[0015] The component appearance model unit is used to establish a corresponding component appearance model according to the appearance image data of the component;

[0016] The factory visualization model unit is used to map the component skeleton model and the component appearance model of the component in blocks, and generate a block three-dimensional image model according to the mapping results of the two, map the block three-dimensional image model on a three-dimensional plane coordinate system and perform pixel color rendering, and sequentially splice the block pixel models that have completed pixel color rendering, thereby obtaining a factory visualization model;

[0017] The factory visualization module is used to establish a spatiotemporal visualization space, and map the factory visualization model generated by the virtual scene establishment module into the spatiotemporal visualization space, set a separate one-dimensional time axis for each block three-dimensional image model in the factory visualization model, and divide it into several time segments, and then establish the normal operating criteria of the corresponding block three-dimensional model, judge whether each block three-dimensional image model is operating normally according to the normal operating criteria, and judge whether the production is abnormal according to the judgment result.

[0018] Furthermore, the setting process of the scene positioning device includes:

[0019] The scene positioning device is installed on each component in each factory, and the factory data acquisition module sets a number for each scene positioning device. When any production line starts working, the system clocks in all the scene positioning devices on the production line communicate with each other to complete time synchronization. When the time synchronization is completed, each scene positioning device sends a data acquisition instruction to the factory data acquisition module;

[0020] After receiving the data collection instruction, the factory data collection module sends the data collection instruction to the architecture data collection unit and the appearance data collection unit;

[0021] The architecture data acquisition unit and the appearance data acquisition unit are respectively provided with H perspective projection cameras and orthogonal projection cameras on each production line, where H is an even integer greater than 0.

[0022] Furthermore, the collection process of the architecture image data and the appearance image data includes:

[0023] When the architecture data acquisition unit and the appearance data acquisition unit receive the data acquisition instruction, they respectively send the data acquisition instruction to the perspective projection camera and the orthogonal projection camera of the corresponding production line;

[0024] According to the data collection instructions, the perspective projection camera and the orthogonal projection camera simultaneously determine the location of each scene positioning device on the production line, and then locate the location of the component to be collected;

[0025] Then the perspective projection camera and the orthogonal projection camera mark the components in the moving state as production components, mark the components in the non-moving state as equipment components, and mark the production components and equipment components with the numbers of their corresponding component positioning devices;

[0026] The data collection event is set at the starting position of each production component on the production line. When the production component is located at the starting position of the production line, the perspective projection camera and the orthogonal projection camera start to execute the data collection event;

[0027] The perspective projection camera and the orthogonal projection camera locate the real-time position of the production component, and then respectively collect the structural image data and appearance image data of the production component, as well as the structural image data and appearance image data of the equipment component at the current position of the production component.

[0028] Furthermore, the process of establishing the component skeleton model and the component appearance model includes:

[0029] When the component skeleton model unit and the component appearance model unit receive the first architecture image data and appearance image data of any data collection event, they set several unit time nodes t1, t2, ..., t c , c is a natural number greater than 0;

[0030] After the component skeleton model unit receives the architecture image data of each component on the production line, it establishes the component skeleton model of the corresponding component according to the architecture image data; and after the component appearance model unit receives the appearance image data of each component on the production line, it establishes the component appearance model of the corresponding component according to the appearance image data.

[0031] Whenever the component skeleton model unit receives a corresponding architecture image data, a component skeleton model is generated accordingly, and three component skeleton models are generated for three adjacent unit time nodes for overlapping mapping, and then the mapping result is used as the component skeleton model of the last time node among the above three unit time nodes, and the number and unit time node of the corresponding device component are marked;

[0032] The component skeleton model is established by the method of establishing the component appearance model and marking the corresponding numbers and unit time nodes.

[0033] Furthermore, the process of performing pixel color rendering on the block three-dimensional image model includes:

[0034] The factory visualization model matches the component skeleton model and the component appearance model with the number, and then divides the component skeleton model and the component appearance model with the same number into J blocks according to the matching results, matches them with each other according to the corresponding positions of each block, and performs overlapping mapping to obtain J block three-dimensional image models, so that each block three-dimensional image model contains part of the equipment components and part of the production components at the same time, where J is a natural number greater than 0;

[0035] Establish J three-dimensional plane coordinate systems, and cut each block three-dimensional image model into P two-dimensional image data, and then map each two-dimensional image data onto the same three-dimensional plane coordinate system, where P is a natural number greater than 0;

[0036] Counting the pixel value of each pixel on the two-dimensional image data, and arranging each two-dimensional image data according to its position in the corresponding block three-dimensional image model;

[0037] Then, the pixels on the two-dimensional image data at adjacent positions are mapped to each other, the pixel values ​​of the pixels at adjacent positions are added and averaged, and then reassigned to the corresponding two pixels;

[0038] By using the above method, pixel values ​​of pixels on the first two-dimensional image data and the second two-dimensional image data, the third two-dimensional image data and the fourth two-dimensional image data, ..., the P-1th two-dimensional image data and the Pth two-dimensional image data are averaged;

[0039] Merging two adjacent two-dimensional image data to obtain a partial area three-dimensional image model, dividing a number of spatial pixels in the local area three-dimensional image model, and calculating the pixel value of each spatial pixel, collecting the same operation as averaging the pixel values ​​of each two-dimensional image data, and averaging the pixel values ​​of the spatial pixels of each local area three-dimensional image model;

[0040] Repeat the above operation until the local area 3D graphic models are gradually merged into a local 3D image model, divide a number of spatial pixels in the local 3D image model, obtain the pixel value of each spatial pixel, and then render the pixel color of each spatial pixel using the pixel value as the color value.

[0041] Furthermore, after the pixel color rendering of all regional three-dimensional image models is completed, the three-dimensional image models of each region are spliced ​​in sequence according to the corresponding positions on the production line to obtain a factory visualization model. At the same time, the block three-dimensional image models corresponding to each production component and equipment component in the factory visualization model are marked with corresponding numbers and unit time nodes.

[0042] Furthermore, the process of establishing the normal operation criteria includes:

[0043] The factory visualization module establishes a spatiotemporal visualization space, and maps the factory visualization model from the virtual scene establishment module into the spatiotemporal visualization space;

[0044] According to the block 3D image models corresponding to each production component and equipment component in the factory visualization model with unit time nodes, a separate one-dimensional time axis is set for each block 3D image model, and the same number of time segments are divided on the one-dimensional time axis according to the number of unit time nodes of the data collection event;

[0045] According to the change status of the block three-dimensional image models of the production components and each equipment component in each time segment on the one-dimensional time axis of the factory visualization model corresponding to the three data acquisition events, the block three-dimensional image models in the same time segment are overlapped to obtain the motion trajectory of each spatial pixel in the three block three-dimensional image models in the corresponding time segment, and the three motion trajectories are mapped, and then three overlapping parts are selected as standard trajectory curves. If the overlapping parts of the various trajectory curves are not enough to form a complete standard trajectory curve, the trajectory curve with the most overlapping parts with other trajectory curves is selected as the reference trajectory curve;

[0046] Then, some curves in the reference trajectory curve are selected to connect the overlapping parts to obtain the standard trajectory curve, and all the standard trajectory curves are fitted to obtain the normal operating criteria for the production line to generate the corresponding type of production components.

[0047] Furthermore, the process of determining whether production is abnormal according to the normal operation criteria includes:

[0048] When subsequent data collection events generate corresponding real-time factory visualization models, the factory visualization module sends the real-time factory visualization models to the spatiotemporal visualization space, thereby matching the corresponding normal operation criteria;

[0049] Then, a real-time trajectory curve of each spatial pixel in the real-time factory visualization model is generated, and it is overlapped and mapped with the corresponding standard trajectory curve in the normal operation criteria, so as to obtain the proportion of the non-overlapping part between each real-time trajectory curve and its corresponding standard trajectory curve;

[0050] If the non-overlapping portion between the two accounts for more than or equal to 50%, the corresponding spatial pixel is judged to be abnormal, otherwise the corresponding spatial pixel is judged to be normal;

[0051] Count the number of abnormal spatial pixels. If the number of abnormal spatial pixels is greater than or equal to 1 / 3 of the total number of spatial pixels in the factory visualization model, it is determined that the corresponding production component or equipment component is abnormal. Then, a maintenance decision is generated based on the corresponding production component or equipment component and sent to the factory maintenance personnel. The factory maintenance personnel maintain the abnormal production component or equipment component based on the maintenance decision.

[0052] If the number of abnormal spatial pixels is less than 1 / 3 of the total number of spatial pixels in the factory visualization model, the current production is judged to be normal, and the above operation is repeated until the corresponding data collection event ends.

[0053] Compared with the prior art, the present invention has the following beneficial effects:

[0054] 1. The present invention simultaneously carries out factory data collection, factory visualization model establishment and factory production status supervision. When abnormalities occur in factory production, real-time supervision and deployment of the production line can be achieved, which effectively improves the efficiency of factory visualization supervision.

[0055] 2. Generate component skeleton models and component appearance models through the architecture image data and appearance image data of each component in the factory, map the component skeleton model and component appearance model of the component in blocks, and generate a block three-dimensional image model based on the mapping results of the two, map the block three-dimensional image model on the three-dimensional plane coordinate system and perform pixel color rendering, and splice the block pixel models that have completed pixel color rendering in turn to obtain a factory visualization model. Compared with traditional factory visualization supervision technology, the present invention has a low learning threshold for supervision and high efficiency in handling accidents. When production abnormalities occur, the site of the production abnormality can be accurately located, thereby achieving timely maintenance and repair. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0057] Figure 1 It is a schematic diagram of the present invention. DETAILED DESCRIPTION

[0058] To make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be described in detail below. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other implementation methods obtained by ordinary technicians in this field without creative work belong to the scope of protection of the present invention.

[0059] like Figure 1 As shown, an interactive factory management system based on three-dimensional visualization technology includes a cloud computing platform, wherein the cloud computing platform is communicatively connected with a factory data acquisition module, a virtual scene establishment module and a factory visualization module;

[0060] The factory data acquisition module is provided with an architecture data acquisition unit, an appearance data acquisition unit and a scene positioning device;

[0061] The factory data acquisition module sets a scene positioning device for each component in the factory, which is used to locate the acquisition object of the architecture data acquisition unit and the appearance data acquisition unit;

[0062] The scene positioning device is composed of a positioning device, a wireless signal device and a system clock, wherein the system clock is provided with data collection instructions and time synchronization instructions;

[0063] The architecture data acquisition unit is provided with a plurality of perspective projection cameras for acquiring architecture image data of each component in the factory;

[0064] The appearance data acquisition unit is provided with a plurality of orthogonal projection cameras for acquiring appearance image data of each component in the factory;

[0065] The virtual scene building module is provided with a component skeleton model unit, a component appearance model unit and a factory visualization model unit;

[0066] The component skeleton model unit is used to establish a corresponding component skeleton model according to the architecture image data of the component;

[0067] The component appearance model unit is used to establish a corresponding component appearance model according to the appearance image data of the component;

[0068] The factory visualization model unit is used to map the component skeleton model and the component appearance model of the component in blocks, and generate a block three-dimensional image model according to the mapping results of the two, map the block three-dimensional image model on a three-dimensional plane coordinate system and perform pixel color rendering, and sequentially splice the block pixel models that have completed pixel color rendering, thereby obtaining a factory visualization model;

[0069] The factory visualization module is used to establish a spatiotemporal visualization space, and map the factory visualization model generated by the virtual scene establishment module into the spatiotemporal visualization space, set a separate one-dimensional time axis for each block three-dimensional image model in the factory visualization model, and divide it into several time segments, and then establish the normal operating criteria of the corresponding block three-dimensional model, judge whether each block three-dimensional image model is operating normally according to the normal operating criteria, and judge whether the production is abnormal according to the judgment result.

[0070] Further, the working principle of the present invention is described below by way of examples:

[0071] The factory maintenance personnel install the scene positioning device on each component in each factory, and after the scene positioning device is installed, it communicates with the factory data acquisition module, and then the factory data acquisition module sets a number for each scene positioning device, where the numbers are S1, S2, ..., S n , where n is a natural number greater than 0;

[0072] It should be noted that the components may be various production equipment on a production line or processing components being produced, etc.;

[0073] When any production line in the factory starts working, the system clocks in all scene positioning devices on the production line communicate with each other to complete time synchronization. When the time synchronization is completed, each scene positioning device sends a data collection instruction to the factory data collection module;

[0074] After receiving the data collection instruction, the factory data collection module sends the data collection instruction to the architecture data collection unit and the appearance data collection unit;

[0075] The architecture data acquisition unit and the appearance data acquisition unit are respectively provided with H perspective projection cameras and orthogonal projection cameras on each production line, where H is an even integer greater than 0;

[0076] It should be noted that, on the production line, the perspective projection cameras are located at the head and tail of the production line, and the number of perspective projection cameras at the head and tail is the same, while the orthogonal projection cameras are located on both sides of the production line in a uniform cross distribution, that is, two adjacent orthogonal projection cameras are arranged diagonally on both sides of the production line;

[0077] Further, when the architecture data acquisition unit and the appearance data acquisition unit receive the data acquisition instruction, they respectively send the data acquisition instruction to the perspective projection camera and the orthogonal projection camera located at the corresponding production line;

[0078] According to the data collection instructions, the perspective projection camera and the orthogonal projection camera simultaneously determine the location of each scene positioning device on the production line, and then locate the location of the component to be collected;

[0079] Then the perspective projection camera and the orthogonal projection camera mark the components in the moving state as production components, mark the components in the non-moving state as equipment components, and mark the production components and equipment components with the numbers of their corresponding component positioning devices;

[0080] The data collection event is set at the starting position of each production component on the production line. When the production component is located at the starting position of the production line, the perspective projection camera and the orthogonal projection camera start to execute the data collection event;

[0081] The perspective projection camera and the orthogonal projection camera locate the real-time position of the production component, and then respectively collect the structural image data and the appearance image data of the production component, and collect the structural image data and the appearance image data of the equipment component at the current position of the production component;

[0082] Whenever the perspective projection camera and the orthogonal projection camera collect architecture image data and appearance image data, the corresponding component numbers are marked and sent to the architecture data collection unit and the appearance data collection unit until the data collection event ends.

[0083] Furthermore, whenever the factory data acquisition module receives the architecture image data and the appearance image data, the architecture image data and the appearance image data are sent to the virtual scene establishment module in real time;

[0084] Then, the virtual scene establishment module sends the architecture image data to the component skeleton model unit, and sends the appearance image data to the component appearance model unit. When the component skeleton model unit and the component appearance model unit receive the first architecture image data and appearance image data of any data collection event, the two set several unit time nodes t1, t2, ..., t c , c is a natural number greater than 0;

[0085] The process of establishing the component skeleton model and the component appearance model includes:

[0086] After the component skeleton model unit receives the architecture image data of each component on the production line, it establishes the component skeleton model of the corresponding component according to the architecture image data; and after the component appearance model unit receives the appearance image data of each component on the production line, it establishes the component appearance model of the corresponding component according to the appearance image data.

[0087] For device components, each time the component skeleton model unit receives a corresponding architecture image data, a component skeleton model is generated accordingly, and three component skeleton models are generated for three adjacent unit time nodes for overlapping mapping, and then the mapping result is used as the component skeleton model of the last time node among the above three unit time nodes, and the corresponding device component number and unit time node are marked;

[0088] For production components, a component skeleton model of the equipment component at each unit time node is established. In addition to marking the number of the corresponding production component, the number of the equipment component at the same unit time node is also marked;

[0089] By adopting the method of establishing component skeleton model, the component appearance model is established and the corresponding numbers and unit time nodes are marked;

[0090] Whenever the component skeleton model unit generates a component skeleton model or the component appearance model unit generates a component appearance model, it is automatically sent to the factory visualization model unit.

[0091] Furthermore, the factory visualization model matches the component skeleton model and the component appearance model with the number, and then divides the component skeleton model and the component appearance model with the same number into J blocks according to the matching result, matches each other according to the position corresponding to each block, and performs overlapping mapping to obtain J block three-dimensional image models, so that each block three-dimensional image model contains part of the equipment components and part of the production components at the same time, where J is a natural number greater than 0;

[0092] Establish J three-dimensional plane coordinate systems, and cut each block three-dimensional image model into P two-dimensional image data, and then map each two-dimensional image data onto the same three-dimensional plane coordinate system, where P is a natural number greater than 0;

[0093] Counting the pixel value of each pixel on the two-dimensional image data, and arranging each two-dimensional image data according to its position in the corresponding block three-dimensional image model;

[0094] Then, the pixels on the two-dimensional image data at adjacent positions are mapped to each other, the pixel values ​​of the pixels at adjacent positions are added and averaged, and then reassigned to the corresponding two pixels;

[0095] By using the above method, pixel values ​​of pixels on the first two-dimensional image data and the second two-dimensional image data, the third two-dimensional image data and the fourth two-dimensional image data, ..., the P-1th two-dimensional image data and the Pth two-dimensional image data are averaged;

[0096] Merging two adjacent two-dimensional image data to obtain a partial area three-dimensional image model, dividing a number of spatial pixels in the local area three-dimensional image model, and calculating the pixel value of each spatial pixel, and averaging the pixel values ​​of the spatial pixels of each local area three-dimensional image model by a similar operation to the pixel value averaging of each two-dimensional image data;

[0097] Repeat the above operation until each local area 3D graphic model is gradually merged into a local 3D image model, divide a number of spatial pixels in the local 3D image model, obtain the pixel value of each spatial pixel, and then perform pixel color rendering on each spatial pixel using the pixel value as the color value;

[0098] When the pixel color rendering of all the regional 3D image models is completed, the 3D image models of each region are sequentially spliced ​​according to the corresponding positions on the production line to obtain the factory visualization model. At the same time, the block 3D image models corresponding to each production component and equipment component in the factory visualization model are marked with corresponding numbers and unit time nodes;

[0099] It should be noted that, according to the continuous execution of the data collection event, the factory visualization model will be updated accordingly until the data collection event ends.

[0100] Furthermore, whenever the factory visualization model is updated, the virtual scene establishment module sends the factory visualization model to the factory visualization module in real time;

[0101] The factory visualization module establishes a spatiotemporal visualization space, and maps the factory visualization model from the virtual scene establishment module into the spatiotemporal visualization space;

[0102] According to the block 3D image models corresponding to each production component and equipment component in the factory visualization model with unit time nodes, a separate one-dimensional time axis is set for each block 3D image model, and the same number of time segments are divided on the one-dimensional time axis according to the number of unit time nodes of the data collection event;

[0103] The changes of the block 3D image models of the production components and each equipment component in each time period in the first three data collection events on the production line are used to establish the normal operation criteria for the production line to produce the corresponding types of production components;

[0104] The process of establishing the normal operation criteria includes:

[0105] According to the change status of the block three-dimensional image models of the production components and each equipment component in each time segment on the one-dimensional time axis of the factory visualization model corresponding to the three data acquisition events, the block three-dimensional image models in the same time segment are overlapped to obtain the motion trajectory of each spatial pixel in the three block three-dimensional image models in the corresponding time segment, and the three motion trajectories are mapped, and then three overlapping parts are selected as standard trajectory curves. If the overlapping parts of the various trajectory curves are not enough to form a complete standard trajectory curve, the trajectory curve with the most overlapping parts with other trajectory curves is selected as the reference trajectory curve;

[0106] Then, some curves in the reference trajectory curve are selected to connect the overlapping parts, so as to obtain the standard trajectory curve;

[0107] By adopting the above method, the standard trajectory curve of the three-dimensional image model of each block is obtained, and all the standard trajectory curves are fitted to obtain the normal operation criteria of the production line to generate the corresponding type of production components.

[0108] When subsequent data collection events generate corresponding real-time factory visualization models, the factory visualization module sends the real-time factory visualization models to the spatiotemporal visualization space, thereby matching the corresponding normal operation criteria;

[0109] Then, a real-time trajectory curve of each spatial pixel in the real-time factory visualization model is generated, and it is overlapped and mapped with the corresponding standard trajectory curve in the normal operation criteria, so as to obtain the proportion of the non-overlapping part between each real-time trajectory curve and its corresponding standard trajectory curve;

[0110] If the non-overlapping portion between the two accounts for more than or equal to 50%, the corresponding spatial pixel is judged to be abnormal, otherwise the corresponding spatial pixel is judged to be normal;

[0111] Count the number of abnormal spatial pixels. If the number of abnormal spatial pixels is greater than or equal to 1 / 3 of the total number of spatial pixels in the factory visualization model, it is determined that the corresponding production component or equipment component is abnormal. Then, a maintenance decision is generated based on the corresponding production component or equipment component and sent to the factory maintenance personnel. The factory maintenance personnel maintain the abnormal production component or equipment component based on the maintenance decision.

[0112] If the number of abnormal spatial pixels is less than 1 / 3 of the total number of spatial pixels in the factory visualization model, the current production is judged to be normal, and the above operation is repeated until the corresponding data collection event ends.

[0113] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. An interactive factory management system based on three-dimensional visualization technology, including a cloud computing platform, characterized in that: The cloud computing platform is communicatively connected with a factory data acquisition module, a virtual scene establishment module and a factory visualization module; The factory data acquisition module is provided with an architecture data acquisition unit, an appearance data acquisition unit and a scene positioning device; The scene positioning device is composed of a positioning device, a wireless signal device and a system clock, wherein the system clock is provided with data collection instructions and time synchronization instructions; The architecture data acquisition unit is provided with a plurality of perspective projection cameras for acquiring architecture image data of each component in the factory; The appearance data acquisition unit is provided with a plurality of orthogonal projection cameras for acquiring appearance image data of each component in the factory; The virtual scene building module is provided with a component skeleton model unit, a component appearance model unit and a factory visualization model unit; The component skeleton model unit is used to establish a corresponding component skeleton model according to the architecture image data of the component; The component appearance model unit is used to establish a corresponding component appearance model according to the appearance image data of the component; The factory visualization model unit is used to map the component skeleton model and the component appearance model of the component in blocks, and generate a block three-dimensional image model according to the mapping results of the two, map the block three-dimensional image model on a three-dimensional plane coordinate system and perform pixel color rendering, and sequentially splice the block pixel models that have completed pixel color rendering, thereby obtaining a factory visualization model; The factory visualization module is used to establish a spatiotemporal visualization space, and map the factory visualization model generated by the virtual scene establishment module into the spatiotemporal visualization space, set a separate one-dimensional time axis for each block three-dimensional image model in the factory visualization model, and divide it into several time segments, and then establish the normal operating criteria of the corresponding block three-dimensional model, judge whether each block three-dimensional image model is operating normally according to the normal operating criteria, and judge whether the production is abnormal according to the judgment result.

2. The interactive factory management system based on three-dimensional visualization technology according to claim 1, characterized in that: The setting process of the scene positioning device includes: The scene positioning device is installed on each component in each factory, and the factory data acquisition module sets a number for each scene positioning device. When any production line starts working, the system clocks in all the scene positioning devices on the production line communicate with each other to complete time synchronization. When the time synchronization is completed, each scene positioning device sends a data acquisition instruction to the factory data acquisition module; After receiving the data collection instruction, the factory data collection module sends the data collection instruction to the architecture data collection unit and the appearance data collection unit; The architecture data acquisition unit and the appearance data acquisition unit are respectively provided with H perspective projection cameras and orthogonal projection cameras on each production line, where H is an even integer greater than 0.

3. The interactive factory management system based on three-dimensional visualization technology according to claim 2, characterized in that: The acquisition process of the architecture image data and the appearance image data includes: When the architecture data acquisition unit and the appearance data acquisition unit receive the data acquisition instruction, they respectively send the data acquisition instruction to the perspective projection camera and the orthogonal projection camera of the corresponding production line; According to the data collection instructions, the perspective projection camera and the orthogonal projection camera simultaneously determine the location of each scene positioning device on the production line, and then locate the location of the component to be collected; Then the perspective projection camera and the orthogonal projection camera mark the components in the moving state as production components, mark the components in the non-moving state as equipment components, and mark the production components and equipment components with the numbers of their corresponding component positioning devices; The data collection event is set at the starting position of each production component on the production line. When the production component is located at the starting position of the production line, the perspective projection camera and the orthogonal projection camera start to execute the data collection event; The perspective projection camera and the orthogonal projection camera locate the real-time position of the production component, and then respectively collect the structural image data and appearance image data of the production component, as well as the structural image data and appearance image data of the equipment component at the current position of the production component.

4. The interactive factory management system based on three-dimensional visualization technology according to claim 3, characterized in that: The process of establishing the component skeleton model and the component appearance model includes: When the component skeleton model unit and the component appearance model unit receive the first architecture image data and appearance image data of any data collection event, they set several unit time nodes t1, t2, ..., t c , c is a natural number greater than 0; After the component skeleton model unit receives the architecture image data of each component on the production line, it establishes the component skeleton model of the corresponding component according to the architecture image data; and after the component appearance model unit receives the appearance image data of each component on the production line, it establishes the component appearance model of the corresponding component according to the appearance image data. Whenever the component skeleton model unit receives a corresponding architecture image data, a component skeleton model is generated accordingly, and the three component skeleton models generated at three adjacent unit time nodes are overlapped and mapped, and then the mapping result is used as the component skeleton model of the last time node among the above three unit time nodes, and the number and unit time node of the corresponding device component are marked; The component skeleton model is established by the method of establishing the component appearance model and marking the corresponding numbers and unit time nodes.

5. The interactive factory management system based on three-dimensional visualization technology according to claim 4, characterized in that: The process of performing pixel color rendering on the block three-dimensional image model includes: Step S11, the factory visualization model matches the component skeleton model and the component appearance model with the number, and then divides the component skeleton model and the component appearance model with the same number into J blocks according to the matching result, matches each other according to the position corresponding to each block, and performs overlapping mapping to obtain J block three-dimensional image models, so that each block three-dimensional image model contains part of the equipment components and part of the production components at the same time, wherein J is a natural number greater than 0; Step S12, establishing J three-dimensional plane coordinate systems, and cutting each block three-dimensional image model into P two-dimensional image data, and then mapping each two-dimensional image data onto the same three-dimensional plane coordinate system, where P is a natural number greater than 0; Step S13, counting the pixel value of each pixel on the two-dimensional image data, and arranging each two-dimensional image data according to its position in the corresponding block three-dimensional image model; Step S14, mapping pixels on the two-dimensional image data at adjacent positions to each other, adding pixel values ​​of pixels at adjacent positions to obtain an average value, and reassigning the average value to the corresponding two pixels; Step S15, through step S14, averaging the pixel values ​​of the pixels on the first two-dimensional image data and the second two-dimensional image data, the third two-dimensional image data and the fourth two-dimensional image data, ..., the P-1th two-dimensional image data and the Pth two-dimensional image data; Step S16: merging two adjacent two-dimensional image data to obtain a partial area three-dimensional image model, dividing a number of spatial pixels in the local area three-dimensional image model, and calculating the pixel value of each spatial pixel, collecting the same operation as averaging the pixel values ​​of each two-dimensional image data, and averaging the pixel values ​​of the spatial pixels of each local area three-dimensional image model; Step S17, repeat step S16 until the local area three-dimensional graphic models are gradually merged into a local three-dimensional image model, divide a number of spatial pixels in the local three-dimensional image model, and obtain the pixel value of each spatial pixel, and then render the pixel color of each spatial pixel using the pixel value as the color value.

6. The interactive factory management system based on three-dimensional visualization technology according to claim 5, characterized in that: When the pixel color rendering of all regional three-dimensional image models is completed, the three-dimensional image models of each region are spliced ​​in sequence according to the corresponding positions on the production line to obtain the factory visualization model. At the same time, the block three-dimensional image models corresponding to each production component and equipment component in the factory visualization model are marked with corresponding numbers and unit time nodes.

7. The interactive factory management system based on three-dimensional visualization technology according to claim 6, characterized in that: The process of establishing the normal operation criteria includes: The factory visualization module establishes a spatiotemporal visualization space, and maps the factory visualization model from the virtual scene establishment module into the spatiotemporal visualization space; According to the block 3D image models corresponding to each production component and equipment component in the factory visualization model with unit time nodes, a separate one-dimensional time axis is set for each block 3D image model, and the same number of time segments are divided on the one-dimensional time axis according to the number of unit time nodes of the data collection event; According to the change status of the block three-dimensional image models of the production components and each equipment component in each time segment on the one-dimensional time axis of the factory visualization model corresponding to the three data acquisition events, the block three-dimensional image models in the same time segment are overlapped to obtain the motion trajectory of each spatial pixel in the three block three-dimensional image models in the corresponding time segment, and the three motion trajectories are mapped to each other, and then three overlapping parts are selected as standard trajectory curves. If the overlapping parts of each trajectory curve are not enough to form a complete standard trajectory curve, the trajectory curve with the most overlapping parts with other trajectory curves is selected as the reference trajectory curve; Then, some curves in the reference trajectory curve are selected to connect the overlapping parts to obtain the standard trajectory curve, and all the standard trajectory curves are fitted to obtain the normal operating criteria for the production line to generate the corresponding type of production components.

8. The interactive factory management system based on three-dimensional visualization technology according to claim 7, characterized in that: The process of judging whether production is abnormal according to the normal operation criteria includes: Step S21, when subsequent data collection events generate corresponding real-time factory visualization models, the factory visualization module sends the real-time factory visualization model to the spatiotemporal visualization space, thereby matching the corresponding normal operation criteria; Step S22, further generating a real-time trajectory curve of each spatial pixel in the real-time factory visualization model, and overlapping and mapping it with the corresponding standard trajectory curve in the normal operation criteria, and further obtaining the proportion of non-overlapping parts between each real-time trajectory curve and its corresponding standard trajectory curve; Step S23: If the proportion of the non-overlapping portion between the two is greater than or equal to 50%, the corresponding spatial pixel is judged to be abnormal, otherwise the corresponding spatial pixel is judged to be normal; Step S24, counting the number of abnormal spatial pixels, if the number of abnormal spatial pixels is greater than or equal to 1 / 3 of the total number of spatial pixels in the factory visualization model, it is determined that the corresponding production component or equipment component is abnormal, and then a maintenance decision is generated according to the corresponding production component or equipment component and sent to the factory maintenance personnel, and the factory maintenance personnel maintains the abnormal production component or equipment component according to the maintenance decision; Step S25: If the number of abnormal spatial pixels is less than 1 / 3 of the total number of spatial pixels in the factory visualization model, it is determined that the current production is normal, and step S24 is repeated until the corresponding data collection event ends.