CN116738540A - Method for presenting and using BIM data on mobile device through graphic interaction engine - Google Patents

Abstract

The present invention relates to the field of computers. The method for presenting and using BIM data on mobile equipment through a graphic interaction engine is characterized in that a recommended drawing list A for drawing skip link and a corresponding recommended weight vector Va are established based on drawing labels and BIM set rules; generating a recommended drawing list B for drawing skip link and a corresponding recommended weight vector Vb based on deep learning of user use data; after cutting and normalizing the recommended weight vector Va and the recommended weight vector Vb, the client orders the recommended weight vectors according to the order of the weights from large to small, so as to obtain a final recommended drawing list; and the user opens a drawing, and the client side provides the corresponding drawing link for the user according to the final recommended drawing list. According to the method, the link relation among the BIM three-dimensional model, the BIM two-dimensional drawing and the non-BIM three-dimensional model is established through optimization of jump links.

Description

Method for presenting and using BIM data on mobile device through graphic interaction engine

Technical Field

The present invention relates to the field of computers, and in particular, to a method for processing computer data.

Background

When a drawing is read or processed by a computer, the following problems exist in the prior art:

1. drawing (paper or electronic version) information of engineering design is stored in a two-dimensional form, the information reading efficiency is low, and the design information cannot be extracted or checked quickly; the two-dimensional drawing cannot accurately and comprehensively describe the three-dimensional design.

2. The non-BIM three-dimensional model and the two-dimensional drawing cannot be completely corresponding, and the three-dimensional design cannot be accurately described; and the drawings and the models cannot be directly overlapped on the same interface for comparison, so that the design is very mental and easy to make mistakes.

3. The existing BIM engineering software tool tends to use BIM models in a pure three-dimensional environment, information and interfaces are too complicated and inconvenient to use, so that a large number of BIM models are not fully utilized, and the design result actually used is still a two-dimensional drawing, so that a large amount of work is wasted.

4. On mobile equipment with lower hardware configuration, a large number of geometric shapes and vector characters cannot be smoothly displayed in a three-dimensional interface, and drawings and models of complex projects must be checked on high-end equipment.

5. The high-quality three-dimensional rendering chart is used in a two-dimensional grating image mode or a real-time rendering tool with single function, and cannot be synchronously overlapped with engineering drawings to be presented, and manual comparison is needed.

Disclosure of Invention

The present invention is directed to a method for presenting and using BIM data on a mobile device through a graphic interaction engine, so as to solve at least one technical problem described above.

The technical problems solved by the invention can be realized by adopting the following technical scheme:

the method for presenting and using BIM data on mobile equipment through a graphic interaction engine is characterized in that a recommended drawing list A for drawing skip link and a corresponding recommended weight vector Va are established based on drawing labels and BIM set rules; generating a recommended drawing list B for drawing skip link and a corresponding recommended weight vector Vb based on deep learning of user use data; after cutting and normalizing the recommended weight vector Va and the recommended weight vector Vb, the client orders the recommended weight vectors according to the order of the weights from large to small, so as to obtain a final recommended drawing list; and the user opens a drawing, and the client side provides the corresponding drawing link for the user according to the final recommended drawing list. According to the method, the link relation is established among the BIM three-dimensional model, the BIM two-dimensional drawing, the non-BIM three-dimensional model and the non-BIM two-dimensional drawing through optimization of jump links, and direct superposition is allowed on the same interface, so that the difficulty of reading and understanding the design is reduced.

Various data in the BIM model are converted into a data format capable of being transmitted through a network. So that the server can be uploaded, allowing the client to download to the client for use. Thereby balancing the efficiency of data transmission and improving the flexibility of data format version change.

The three-dimensional body is split into a plurality of small blocks and distributed into a plurality of frames. Thus greatly reducing waiting and jamming caused by the operation of the main thread task.

Rendering data using a plurality of graphics and interaction techniques, comprising: overlapping the drawings, presenting the drawings and the model simultaneously, and automatically filling the section. Therefore, smooth, attractive and unique graphic interaction effect is realized at the mobile terminal, the defects of the traditional paper drawing and the conventional electronic drawing are overcome, and the information of the design drawing and the model is accurately, intuitively and comprehensively expressed.

Drawings

Fig. 1 is a flow chart of the present invention.

Detailed Description

In order that the manner in which the invention is practiced, as well as the features and objects and functions thereof, will be readily understood and appreciated, the invention will be further described in connection with the accompanying drawings.

Referring to fig. 1, a method of presenting and using BIM data at a mobile device through a graphical interaction engine includes the steps of:

1. a weighted directed graph (class A link) linking each drawing is established based on a predetermined rule and used for pre-reading and drawing skip recommending functions. The specific operation is as follows:

a. and establishing a drawing label. There are two established methods:

i. automatically read from the BIM model. Metadata of drawing attributes of the BIM model exist in a BIM database and are directly read according to a preset rule.

OCR automatically recognizes the frame information to generate labels of drawing related attributes. Compared with the method for directly reading information according to rules, the method has lower requirement on consistency of the data sources by OCR, and is applied to some special drawings lacking metadata or BIM models with inconsistent drawing data rules.

b. Establishing a class A link during BIM data processing: and based on the drawing labels and the set rules, the direct jump link of the drawing is automatically established, so that the effect of saving a large amount of drawing turning and finding work when the drawing is seen is realized.

Types of class a links include:

i. upper and lower layers: a plan of adjacent floors, such as a first floor plan and a second floor plan, and a top floor plan and a roof plan.

Adjacent region: a plan view of adjacent planar areas, such as the south and north areas of a single building. Drawing a detailed diagram: and a small-scale drawing, such as a building plane, a stair detail drawing plane and a wall detail drawing, which are drawn from the large-scale drawing, are linked, and a combination plane and a subitem plane are linked.

Content derivatization: and a link is established between the derivative drawing of the same base drawing, such as a planning general drawing and a greening general drawing, a fire fighting general drawing and the like.

v. spatial relationship: and establishing a link according to a spatial relationship, such as establishing a link between a wall detail drawing, a door and window large sample and a corresponding building elevation.

Designating an association: and establishing drawing links according to the logic relationship, such as establishing links between the total drawing and a layer of plane of each sub-building drawing.

c. Generating a Weighted Directed Graph (Weighted-Directed Graph) through a class A link, and realizing the following steps:

i. and establishing a corresponding Node (Node) for each drawing, and storing Id information of the drawing in Node data.

Simply traversing all class-a links, creating corresponding edges (edges), each Edge containing data: the distance of the start node and the target node, the edge, the type of edge (corresponding to the link type).

Generating a recommendation drawing list A and a corresponding recommendation weight vector Va by using the links.

2. Personalized recommended links (class B links) generated based on deep learning of user usage data. The technology is realized:

a. user operation data are obtained from a client and then uploaded to the cloud, and the user operation data are converted into vectors at the cloud through an encoder

Xt, upload the cloud.

b. The cloud recommendation system uses an LSTM model as a core, and a prediction state vector Ht is deduced by combining a long-term operation history Ct-1 of a user and a current client state Xt, and Ct-1 is updated to be Ct. Ct is one of the input data for the next inference operation.

c. The current selectable recommended item of the client is converted into a vector through an encoder, the vector is connected with Ht in series, an ANN model is input, and a final recommended result is output. The recommendation result is a drawing list B and a corresponding recommendation weight vector Vb.

3. And opening a drawing by the user, and providing drawing skip recommendation by combining the class A link and the class B link by the client.

The technology can reduce a large amount of manual graph finding and turning work, and improves the efficiency and accuracy of graph reading of engineering designers. The technology is realized:

a. the client obtains the weight vectors Va and Vb according to the above method, and reorders the dimension values from large to small, respectively.

b. And cutting the tail dimension of the other vector according to the small dimension of the two vectors so as to make the dimensions equal.

c. Normalizing (normalizing) the weight vectors after clipping

d. The method comprises the steps of concatenating two weight vectors, sequencing the weight vectors from large to small, obtaining a final recommended drawing list from an AB list according to the sequence, and recommending a specified number of drawing links for selection by a user.

4. Storage and reading of data

a. Data format: the various data in the BIM model are converted into a data format capable of being transmitted by a network by using different serialization algorithms, and are uploaded to a server and downloaded to a client for use. Rational application of various data formats can balance the efficiency of data transmission and the degree of variability of the version of the data format.

i. The two-dimensional drawing data and metadata can freely define a data structure by using a MessagePack binary serialization algorithm, highly compress the data volume and realize high-speed storage and reading. And thirdly, the three-dimensional model data is subjected to high compression by using a GLTF binary serialization algorithm, and the three-dimensional mesh data is subjected to telling storage and reading.

And other drawing and model metadata with smaller data volume can freely define a data structure by using a JSON serialization algorithm.

b. And the data pre-reading function module is used for generating a pre-reading sequence list by using the class A link weighted directed graph, and automatically downloading data in the background while the user operates the data pre-reading function module, wherein the downloading sequence accords with the most probable browsing sequence of the user. The technology can reduce the situation that the user waits for the completion of the downloading.

The technology is realized:

i. resetting the distance Dn of all nodes to a maximum value

And ii, creating an Agent, and traversing the whole graph in depth first from a node corresponding to the drawing currently used by the user. The Agent initial distance value Da is zero.

Every time an Agent spans an edge, increasing the reciprocal of the weight of the edge to an Agent distance value Da, and assigning Da a distance value Dn to the node where the Agent is currently located

And iv, after the traversing is finished, ordering all the nodes according to the distance value Dn from small to large, and downloading drawing data from a server according to the order.

5. The figure generation of drawings and models uses coroutine techniques. The three-dimensional body is generated in a main thread, a Coroutine technology is used for automatically splitting the task generated by the body into a plurality of small blocks according to the designated workload of each frame, and the small blocks are distributed into a plurality of frames. The dedicated controller controls the time that all the coves Cheng Yunsuan of each frame can run, allocating the available run time according to the importance of the cove, the more important the cove run time is. The technology can greatly reduce waiting and blocking caused by the operation of the main thread task.

6. Use of a graphical interaction engine: rendering data using a plurality of graphics and interaction techniques, comprising: overlapping the drawings, presenting the drawings and the model simultaneously, and automatically filling the section. The technology can realize smooth, attractive and unique graphic interaction effect at the mobile terminal, overcomes the defects of the traditional paper drawing and the conventional electronic drawing, and accurately, intuitively and comprehensively expresses the information of the design drawing and the model.

a. Folding the drawing: and when browsing the drawing A, the user can superimpose and display the content of the drawing B. The content of the drawing B is displayed in a fixed range, the drawing B is not displayed outside the range, and the range can move along with the drawing A, and can also be a screen fixed range. The display of drawing A and the display of drawing B in the range are different from the normal display. The background operation is combined with the use of a graphics card shader (GPUshader), so that the effect can be achieved without generating additional geometric shapes and materials. The technology is realized:

i. when the geometric figure of the drawing is generated for the first time, a special material is created for each drawing by using a shader, and a drawing coordinate system attribute value of the material is initialized;

and ii, selecting the content of the overlapped drawing B when the user browses the drawing A. If the drawing is not loaded at this time

B, downloading and generating a drawing B;

calculating the angular point coordinates of the current window, shifting inwards by a specified distance, converting the screen pixel coordinates into three-dimensional coordinates of a plane where the drawing B is located through a camera projection matrix, and converting the three-dimensional coordinates into two angular points in the plane of the drawing B through a coordinate system of the drawing B, namely, taking a two-dimensional rectangular range in the plane of the drawing B as a superposition area.

And IV, activating the object corresponding to the drawing B. Setting the attribute related to the overlay in the materials of the drawing A and the drawing B:

1. the Boolean value of the superimposed graph is 1

2. The corner attribute of the overlapped area is the corner coordinate of the last step

3. Display arrangement inside and outside of overlapping region

And v, when the coloring device of the materials of the drawing A and the drawing B calculates each pixel point, firstly converting the three-dimensional coordinates of the pixel points into coordinates through the coordinate system of the drawing B to form two-dimensional coordinate points, and then judging whether the pixel points are positioned in the overlapped rectangular area. If yes, rendering the pixel point according to the visual setting of the overlay; if not, rendering the pixel point according to the visual setting of the non-overlapping graph. If the visual setting is not displayed, the rendering of the pixel is skipped.

b. The drawing and the model are presented simultaneously, and the model can be in a sectioning form and automatically fills the sectioning surface. High-quality smooth display and interaction of a large number of geometric shapes on a mobile device side are achieved by using a shape collision device (binder), a drawing call batch (Draw Call Batching), a real-time occlusion elimination device (Runtime Occlusion Culling), a shader (GPU loader) and a rendering map (Render Texture).

i. Draw call aggregate (Draw Call Batching): when the three-dimensional shapes are generated, a drawing call batch-combining function of the graphic engine is called, and scattered three-dimensional shapes distributed in a large number of components are recombined to form a plurality of combined meshes. And when each frame is rendered, sending the assembled mesh to the GPU for rendering.

ii, real-time shielding and eliminating: when the three-dimensional body is generated, a box collision device and a mesh collision device which wrap the three-dimensional body are generated at the same time. Before each frame is rendered, rays are emitted from the camera viewpoint, and it is calculated whether each box collider is visible, i.e., is occluded by other colliders. If so, then calculate whether the mesh collider is visible, and decide whether the mesh of the object needs to be rendered in the frame.

Drawing range collision rejection: the visibility of the three-dimensional model is initially invisible. And establishing a new box collision device in the three-dimensional range displayed in the drawing, detecting all three-dimensional models staggered with the collision device and setting the three-dimensional models to be visible.

Cutting the three-dimensional model: the internal structure of the complex model needs to be presented in a three-dimensional cut-away mode, and corresponds to a section view in a design drawing. The technology is realized:

1. creating materials by custom shaders when generating three-dimensional shapes

2. Writing the coordinate system matrix of the drawing into the video memory of the shader

3. When the shader renders each pixel, a drawing coordinate system matrix is read, the current pixel point is converted into world coordinates, and whether the current pixel point is positioned in front of a plane where a drawing is positioned is judged. If yes, skipping the rendering of the pixel point.

c. Automatic filling of the section: the three-dimensional mesh body has no volume information, and the inside of the body is shown as a cavity when being displayed in a sectioning way. In order to meet the habit of viewing by engineering project personnel, the function of automatically filling the cut surface is required. The technology is realized:

i. using a drawing sectioning range collider to find a three-dimensional shape displayed by sectioning

ii, obtaining each original object mesh and material from the combined mesh in batch processing

inputting the mesh and the materials into a plurality of shaders for rendering. Here, two rendering paths are split:

1. path A

a. Setting a shader A: closing depth rejection, opening back rejection, setting rendering queue to be transparent

b. The mesh and texture are input to shader A1.

c. Shader A renders a pixel each time, adding 1 to the pixel buffer

d. The mesh and texture are input to shader A2.

e. The shader A2 determines whether each pixel buffer is odd. If the rendering object is odd, the filling color rendering is cut, otherwise, the rendering is skipped, and the rendering object is a rendering map A

2. Path B

a. Setting a shader B: opening depth rejection, closing background rejection, and setting a rendering queue to be transparent

b. Judging whether the pixel is a back surface, if so, performing the rendering according to the cut filling color, otherwise, skipping the rendering, and the rendering target is a rendering map B.

And step 3, repeating the step 3 for each mesh.

And v, inputting the rendering maps A and B into a shader C, and merging pixels to obtain the rendering map C.

And (vi) superposing the rendering map C on the rendering result of the current frame to realize the filling effect of the cut surface. The foregoing has shown and described the basic principles and main features of the present invention and the advantages of the present invention. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, and that the above embodiments and descriptions are merely illustrative of the principles of the present invention, and various changes and modifications may be made without departing from the spirit and scope of the invention, which is defined in the appended claims. The scope of the invention is defined by the appended claims and equivalents thereof.

Claims (9)

1. The method for presenting and using BIM data on mobile equipment through a graphic interaction engine is characterized in that a recommended drawing list A for drawing skip A type link and a corresponding recommended weight vector Va are established based on drawing labels and BIM set rules; generating a recommended drawing list B for drawing skip B-type link and a corresponding recommended weight vector Vb based on deep learning of user use data; after cutting and normalizing the recommended weight vector Va and the recommended weight vector Vb, the client orders the recommended weight vectors according to the order of the weights from large to small, so as to obtain a final recommended drawing list; and the user opens a drawing, and the client side provides the corresponding drawing link for the user according to the final recommended drawing list.

2. The method for presenting and using BIM data on a mobile device through a graphic interaction engine according to claim 1, wherein the method for generating the recommended drawing list a and the corresponding recommended weight vector Va comprises the following steps:

step 1, establishing a corresponding node for each drawing, and storing Id information of the drawing in node data;

step 2, traversing all A-type links, and establishing corresponding edges, wherein each edge contains data: the distance between the initial node and the target node, the distance between the edges and the corresponding link type of the edges;

and 3, generating a recommended drawing list A and a corresponding recommended weight vector Va by using the links.

3. The method for presenting and using BIM data on a mobile device through a graphical interaction engine according to claim 1, wherein the method for generating the recommended drawing list B and the corresponding recommended weight vector Vb includes the steps of:

step 1, user operation data are obtained from a client and then uploaded to a cloud end, and the user operation data are converted into vectors Xt through an encoder in the cloud end and uploaded to the cloud end;

and 2, the cloud recommendation system adopts an LSTM model as a core, and a predicted state vector Ht is deduced by combining a long-term operation history Ct-1 of a user and a current client state Xt, and Ct-1 is updated to be Ct. Ct is one of the input data for the next inference operation;

and 3, converting the current selectable recommended item of the client into a vector through an encoder, connecting the vector with Ht in series, inputting an ANN model, and outputting a final recommended result, wherein the recommended result is a drawing list B and a corresponding recommended weight vector Vb.

4. The method of presenting and using BIM data on a mobile device through a graphical interaction engine according to claim 1, wherein the clipping of the recommended weight vector Va and the recommended weight vector Vb is performed by:

step 1, a recommended weight vector Va and a recommended weight vector Vb reorder dimension values according to the order from large to small respectively;

and 2, cutting the tail dimension of the other vector according to the small dimension of the two vectors to make the dimensions equal.

5. The method of presenting and using BIM data on a mobile device through a graphical interaction engine according to claim 1, wherein the various data in the BIM model are converted into a network transmittable data format using different serialization algorithms:

the two-dimensional drawing data and the metadata use a MessagePack binary serialization algorithm, the three-dimensional model data use a GLTF binary serialization algorithm, and other drawing and model metadata with smaller data size use a JSON serialization algorithm.

6. The method for presenting and using BIM data on a mobile device through a graphical interaction engine according to claim 1, wherein the tasks generated by the body are automatically split into a plurality of small blocks according to the specified workload per frame using a coroutine technique, and are distributed to a plurality of frames, the controller controls the time that all the coroutines Cheng Yunsuan of each frame can run, and the more important coroutines run longer according to the importance of the coroutines.

7. The method for presenting and using BIM data on a mobile device through a graphic interaction engine according to claim 1, wherein the steps of applying a plurality of graphics to overlap with the presentation drawing of the interaction technology are as follows:

step 1, when geometric shapes of drawings are generated for the first time, creating special materials for each drawing by using a shader, and initializing drawing coordinate system attribute values of the materials;

step 2, selecting the content of the overlapped drawing B when a user browses the drawing A, and downloading and generating the drawing B if the data of the drawing B is not loaded at the moment;

calculating corner coordinates of a current window, shifting inwards by a specified distance, converting the screen pixel coordinates into three-dimensional coordinates of a plane where a drawing B is located through a camera projection matrix, and converting the three-dimensional coordinates into two corner points in the plane of the drawing B through a coordinate system of the drawing B, namely, taking a two-dimensional rectangular range in the plane of the drawing B as a superposition area;

step 4, activating an object corresponding to the drawing B, and setting the attribute related to the overlay in the materials of the drawing A and the drawing B;

step 5, when the coloring device of the materials of the drawing A and the drawing B calculates each pixel point, firstly converting the three-dimensional coordinates of the pixel point into coordinates through the coordinate system of the drawing B to form two-dimensional coordinate points, and then judging whether the pixel points are positioned in the overlapped rectangular area or not: if yes, rendering the pixel point according to the visual setting of the overlay; if not, rendering the pixel point according to the visual setting of the non-overlapping graph, and if the visual setting is not displayed, skipping the rendering of the pixel.

8. The method for presenting and using BIM data on a mobile device through a graphic interaction engine according to claim 1, wherein the method for applying a plurality of graphic and interaction technologies to make a drawing and a model simultaneously present a presentation drawing to be superimposed comprises the following specific steps:

step 1, drawing call batch combination: when the three-dimensional shapes are generated, a drawing call batch-combining function of the graphic engine is called, and scattered three-dimensional shapes distributed in a large number of components are recombined to form a plurality of combined meshes. Sending the assembled mesh to the GPU for rendering when each frame is rendered;

step 2, real-time shielding and eliminating: when the three-dimensional body is generated, a box collision device and a mesh collision device which wrap the three-dimensional body are generated at the same time, rays are emitted from a camera viewpoint before each frame is rendered, whether each box collision device is visible or not is calculated, namely whether each box collision device is shielded by other collision devices or not is judged, if so, whether the mesh collision device is visible or not is calculated, and whether the mesh of the object needs to be rendered in the frame is determined;

step 3, drawing range collision rejection: the visibility of the three-dimensional model is initially invisible, a new box collider is established in the three-dimensional range displayed by the drawing, and all the three-dimensional models staggered with the collider are detected and set to be visible.

9. The method for presenting and using BIM data on a mobile device through a graphic interaction engine according to claim 1, wherein the method for automatically filling the cut surface by using a plurality of graphic and interaction technologies comprises the following steps:

step 1, using a drawing sectioning range collider to find a three-dimensional body displayed by sectioning;

step 2, obtaining each original object mesh and material from the combined mesh in the batch processing;

step 3, inputting the mesh and the materials into a plurality of shaders for rendering: setting a shader A: closing depth rejection, opening back rejection, and setting a rendering queue to be transparent; inputting the mesh and the material into a shader A1; the shader A renders pixels each time, and adds 1 in the pixel buffer; inputting the mesh and the material into a shader A2; the shader A2 determines whether each pixel buffer is odd. If the color is odd, the filling color is cut, otherwise, the rendering is skipped, and the rendering target is a rendering map A; setting a shader B: opening depth rejection, closing background rejection, and setting a rendering queue to be transparent; judging whether the pixel is a back surface, if so, performing split filling color rendering, otherwise, skipping rendering, and rendering the target as a rendering map B;

step 4, repeating the step 3 for each mesh;

step 5: inputting the rendering maps A and B into a shader C, and merging pixels to obtain a rendering map C;

step 6: and superposing the rendering map C on the rendering result of the current frame to realize the section filling effect.