Intelligent production management system and management method based on 3D retrieval technology

Abstract

The present invention proposes an intelligent production management system and management method based on 3D retrieval technology, wherein the intelligent production management system includes a data preprocessing module, a data receiving module, a data storage module, a data processing module, a data matching module, a processing allocation module and a processing quotation module; based on 3D retrieval technology, the present invention first extracts features from 3D drawings through a neural network, and then matches the pre-stored historical processing parts quotation information and factory information to obtain a processing factory recommendation table and a processing quotation sheet. The present invention applies 3D retrieval technology to the intelligent management of production, which not only improves production efficiency, but also reduces dependence on people, which is conducive to the intelligent management of production.

Description

Technical Field

[0001] The present invention relates to the field of intelligent production management, and in particular to an intelligent production management method based on 3D retrieval technology. Background Art

[0002] The application of 3D models is becoming more and more extensive. 3D models are widely used in industrial product design, virtual reality, virtual humans, 3D games, education, film and television animation, etc. Therefore, there are trillions of 3D models, and a large number of 3D models are generated and disseminated every day. There is an urgent need to analyze, match and retrieve 3D models. It has become very important to study content-based 3D model analysis, matching and retrieval technology.

[0003] 3D search technology has long been mature and can be searched in the entire enterprise model database (PDM) according to geometric shapes. Engineers can directly find the results and call them directly into 3D CAD to improve design efficiency and avoid repeated design. In the field of industrial manufacturing, after obtaining the corresponding 3D model, technicians are often required to decompose the corresponding 3D model and select the appropriate processing method (self-processing or outsourcing) based on the data of the procurement department. At the same time, the corresponding technicians are also required to give a rough quotation for the model. This requires the technicians to complete it based on experience and a large amount of data, which has high requirements for technicians and requires a large amount of time. Summary of the invention

[0004] To solve the above problems, the present invention proposes an intelligent production management method based on 3D retrieval technology.

[0005] The main contents of the present invention include:

[0006] Intelligent production management system based on 3D retrieval technology, including:

[0007] Data receiving module, used to receive 3D drawings and production influencing parameters;

[0008] A data storage module, used to store the data received by the data receiving module and the information of historically processed parts; wherein the information of historically processed parts includes the quotation data of historically processed parts and the factory data of historically processed parts;

[0009] A data processing module, using a neural network, performs feature extraction on the data transmitted by the data receiving module to obtain a corresponding parts feature library;

[0010] A data matching module compares the part feature library obtained by the data processing module with the stored historically processed parts to obtain information of the historically processed parts with the highest matching degree;

[0011] A processing allocation module, wherein the data storage module respectively obtains the factory data of the historically processed parts with the highest matching degree output by the data matching module to form a processing factory recommendation table;

[0012] The processing quotation module uses the data storage module to obtain the quotation data of the historical processing parts with the highest matching degree output by the data matching module, and forms a processing quotation sheet according to a pre-designed quotation strategy.

[0013] Preferably, it further comprises a data preprocessing module, which is connected to the data receiving module and is used for unifying the formats of 3D drawings in different formats.

[0014] Preferably, the step of the data preprocessing module unifying the formats of 3D drawings of different formats comprises: converting the 3D drawings into STL format through the freeCAD command line tool.

[0015] Preferably, the step of using a neural network to extract features from the data transmitted by the data receiving module includes: first, evenly taking points from the surface information of the STL file to generate a PCL point cloud file; then extracting the VFH features of the PCL point cloud through the VFH viewpoint histogram feature algorithm and saving it as a feature file; then establishing a FLANN fast nearest neighbor search algorithm index based on the feature file, and finally finding the most similar target from the index through the FLANN fast nearest neighbor algorithm.

[0016] Preferably, the production influencing parameters include the processing accuracy of the parts, the materials of the parts and the number of parts processed.

[0017] The present invention also proposes an intelligent production management method based on 3D retrieval technology, comprising the following steps:

[0018] S1. Obtain the 3D drawings to be analyzed and the production influencing parameters;

[0019] S2. Using neural network, perform feature decomposition on 3D drawings to obtain a parts feature library;

[0020] S3. According to the set 3D matching algorithm, the part feature library is compared with the pre-stored information of the historically processed parts to obtain matching historically processed parts;

[0021] S4. Obtain the quotation information of the historical processed parts and the corresponding factory information that match S3, and obtain a processing factory recommendation table and a processing quotation sheet.

[0022] Preferably, after obtaining the 3D drawings to be analyzed, the method further includes a step of standardizing the format of the 3D drawings.

[0023] Preferably, the step of unifying the format of the 3D drawings is: converting the 3D drawings into STL format through the freeCAD command line tool.

[0024] Preferably, the steps of feature decomposing 3D drawings using a neural network include: first, evenly taking points from the surface information of the STL file to generate a PCL point cloud file; then extracting the VFH features of the PCL point cloud through a VFH viewpoint histogram feature algorithm and saving it as a feature file; then establishing a FLANN fast nearest neighbor search algorithm index based on the feature file, and finally finding the most similar target from the index through the FLANN fast nearest neighbor algorithm.

[0025] The beneficial effect of the present invention is that the present invention proposes an intelligent production management method based on 3D retrieval technology. Based on the 3D retrieval technology, the 3D drawings are firstly feature extracted by a neural network, and then matched with the pre-stored historical processing parts quotation information and factory information, so as to obtain a processing factory recommendation table and a processing quotation sheet. The present invention applies 3D retrieval technology to the intelligent management of production, which not only improves production efficiency, but also reduces dependence on people, and is conducive to the intelligent management of production. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is a functional block diagram of the present invention. DETAILED DESCRIPTION

[0027] The technical solution protected by the present invention is described in detail below with reference to the accompanying drawings.

[0028] See also Figure 1. The present invention proposes an intelligent production management system based on 3D retrieval technology, including a data preprocessing module, a data receiving module, a data storage module, a data processing module and a data matching module, and also includes a processing allocation module and a processing quotation module; wherein the data preprocessing module is connected to the data receiving module, and the data preprocessing module is used to unify the formats of 3D drawings of different formats; the data receiving module is used to receive the unified 3D drawings and production influencing parameters; and the data storage module is connected to the data receiving module, and is used to store the data received by the data receiving module and the information of historical processed parts; the data processing module is connected to the data receiving unit, and uses a neural network to process the data transmitted by the data receiving module. The feature extraction is performed on the data to obtain the corresponding part feature library; the data matching module is connected to the data storage module and the data processing module, and compares the part feature library obtained by the data processing module with the stored historical processing parts to obtain the information of the historical processing parts with the highest matching degree; the processing allocation module is connected to the data matching module, and obtains the factory data of the historical processing parts with the highest matching degree output by the data matching module from the data storage module to form a processing factory recommendation table; the processing quotation module is also connected to the data matching module, and obtains the quotation data of the historical processing parts with the highest matching degree output by the data matching module from the data storage module to form a processing quotation sheet according to the pre-designed quotation strategy.

[0029] In one embodiment, the information of the historically processed parts includes quotation data of the historically processed parts and factory data of the historically processed parts.

[0030] In one of the embodiments, the step of the data preprocessing module unifying the formats of 3D drawings of different formats includes: converting the 3D drawings into STL format through the freeCAD command line tool.

[0031] In one of the embodiments, a neural network is used to extract features from the data transmitted by the data receiving module, including the following steps: first, evenly taking points from the surface information of the STL file to generate a PCL point cloud file; then, the VFH features of the PCL point cloud are extracted by the VFH viewpoint histogram feature algorithm and saved as a feature file; then, a FLANN fast nearest neighbor search algorithm index is established based on the feature file, and finally, the most similar target is found from the index by the FLANN fast nearest neighbor algorithm.

[0032] Among them, production influencing parameters include part processing accuracy, part material and the number of parts processed.

[0033] The present invention also proposes an intelligent production management method based on 3D retrieval technology, comprising the following steps:

[0034] S1. Obtain the 3D drawings to be analyzed and the production influencing parameters;

[0035] S2. Using neural network, perform feature decomposition on 3D drawings to obtain a parts feature library;

[0036] S3. According to the set 3D matching algorithm, the part feature library is compared with the pre-stored information of the historically processed parts to obtain matching historically processed parts;

[0037] S4. Obtain the quotation information of the historical processed parts and the corresponding factory information that match S3, and obtain a processing factory recommendation table and a processing quotation sheet.

[0038] Preferably, after obtaining the 3D drawings to be analyzed, the method further includes a step of standardizing the format of the 3D drawings.

[0039] Preferably, the step of unifying the format of the 3D drawings is: converting the 3D drawings into STL format through the freeCAD command line tool.

[0040] Preferably, the steps of feature decomposing 3D drawings using a neural network include: first, evenly taking points from the surface information of the STL file to generate a PCL point cloud file; then extracting the VFH features of the PCL point cloud through a VFH viewpoint histogram feature algorithm and saving it as a feature file; then establishing a FLANN fast nearest neighbor search algorithm index based on the feature file, and finally finding the most similar target from the index through the FLANN fast nearest neighbor algorithm.

[0041] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. Intelligent production management system based on 3D retrieval technology, characterized by: include: Data receiving module, used to receive 3D drawings and production influencing parameters; A data storage module, used to store the data received by the data receiving module and the information of historically processed parts; wherein the information of historically processed parts includes the quotation data of historically processed parts and the factory data of historically processed parts; A data processing module, using a neural network, performs feature extraction on the data transmitted by the data receiving module to obtain a corresponding parts feature library; A data matching module compares the part feature library obtained by the data processing module with the stored historically processed parts to obtain information of the historically processed parts with the highest matching degree; A processing allocation module, wherein the data storage module respectively obtains the factory data of the historically processed parts with the highest matching degree output by the data matching module to form a processing factory recommendation table; A processing quotation module, in which the data storage module respectively obtains the quotation data of the historical processing parts with the highest matching degree output by the data matching module, and forms a processing quotation sheet according to a pre-designed quotation strategy; Production influencing parameters include part processing accuracy, part material and the number of parts processed.

2. The intelligent production management system based on 3D retrieval technology according to claim 1 is characterized in that: It also includes a data preprocessing module, which is connected to the data receiving module and is used to unify the formats of 3D drawings in different formats.

3. The intelligent production management system based on 3D retrieval technology according to claim 2 is characterized in that: The step of the data preprocessing module unifying the formats of 3D drawings of different formats includes: converting the 3D drawings into STL format through the freeCAD command line tool.

4. The intelligent production management system based on 3D retrieval technology according to claim 1 is characterized in that: The steps of extracting features from the data transmitted by the data receiving module by using a neural network include: first, evenly taking points from the surface information of the STL file to generate a PCL point cloud file; then extracting the VFH features of the PCL point cloud by using a VFH viewpoint histogram feature algorithm and saving them as a feature file; then establishing a FLANN fast nearest neighbor search algorithm index based on the feature file, and finally searching for the most similar target from the index by using the FLANN fast nearest neighbor algorithm.

5. An intelligent production management method based on 3D retrieval technology, characterized in that: The steps include: S1. Obtain the 3D drawings to be analyzed and the production influencing parameters; S2. Using neural network, perform feature decomposition on 3D drawings to obtain a parts feature library; S3. According to the set 3D matching algorithm, the part feature library is compared with the pre-stored information of the historically processed parts to obtain matching historically processed parts; S4. Obtain the quotation information of the historical processed parts and the corresponding factory information that match S3, and obtain a processing factory recommendation table and a processing quotation sheet.

6. The intelligent production management method based on 3D retrieval technology according to claim 5 is characterized in that: After obtaining the 3D drawings to be analyzed, the step of standardizing the format of the 3D drawings is also included.

7. The intelligent production management method based on 3D retrieval technology according to claim 6 is characterized in that: The steps to unify the format of 3D drawings are as follows: convert 3D drawings into STL format through the freeCAD command line tool.

8. The intelligent production management method based on 3D retrieval technology according to claim 5 is characterized in that: The steps of feature decomposition of 3D drawings using neural networks include: first, evenly taking points from the surface information of the STL file to generate a PCL point cloud file; then extracting the VFH features of the PCL point cloud through the VFH viewpoint histogram feature algorithm and saving it as a feature file; then establishing a FLANN fast nearest neighbor search algorithm index based on the feature file, and finally finding the most similar target from the index through the FLANN fast nearest neighbor algorithm.

Images