CN115502443A - Multi-station intelligent drilling and compounding integrated machine for fixed ball valve seat rings - Google Patents
Multi-station intelligent drilling and compounding integrated machine for fixed ball valve seat rings Download PDFInfo
- Publication number
- CN115502443A CN115502443A CN202210983370.6A CN202210983370A CN115502443A CN 115502443 A CN115502443 A CN 115502443A CN 202210983370 A CN202210983370 A CN 202210983370A CN 115502443 A CN115502443 A CN 115502443A
- Authority
- CN
- China
- Prior art keywords
- drilling
- feature maps
- feature
- workpiece
- ball valve
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23B—TURNING; BORING
- B23B41/00—Boring or drilling machines or devices specially adapted for particular work; Accessories specially adapted therefor
- B23B41/02—Boring or drilling machines or devices specially adapted for particular work; Accessories specially adapted therefor for boring deep holes; Trepanning, e.g. of gun or rifle barrels
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23B—TURNING; BORING
- B23B47/00—Constructional features of components specially designed for boring or drilling machines; Accessories therefor
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Drilling And Boring (AREA)
- Machine Tool Sensing Apparatuses (AREA)
Abstract
The application relates to the technical field of intelligent manufacturing, and specifically discloses a compound all-in-one of fixed ball valve disk seat circle intelligence drilling of multistation, and it includes the frame, including the frame main part, set up in the work piece fixed platform and the setting of frame main part the universal removal chassis of the bottom of frame main part. The intelligent drilling and compounding all-in-one machine for the multi-station fixed ball valve seat ring further comprises an electric indexing chuck arranged on the workpiece fixing platform; and a plurality of drilling devices mounted to the frame, wherein each of the drilling devices includes a drilling device having a drill bit and a servo motor for driving the drilling device. In particular, the plurality of drilling apparatuses can be manipulated to simultaneously drill a plurality of locations of a workpiece to be machined to improve drilling efficiency and drilling accuracy.
Description
Technical Field
The application relates to the technical field of intelligent manufacturing, and more specifically relates to a compound all-in-one of fixed ball valve disk seat circle intelligence drilling of multistation.
Background
In the valve manufacturing field, the ball valve is a kind of valve class that uses very extensively, in the ball valve manufacturing process, the ball valve disk seat is that machining precision and requirement are all higher, can design the grease injection structure generally in the ball valve design process, in this structure, the disk seat part need punch on excircle and plane, and the hole site of these two planes need correspond and communicate with each other, all be the manual work in the disk seat course of working at present and go the marking off, the drilling is gone to the processing through many times of clamping, the process is complicated loaded down with trivial details, machining efficiency is very low-efficient, and machining precision can not be controlled.
Therefore, an optimized drilling apparatus is desired.
Disclosure of Invention
The present application is proposed to solve the above-mentioned technical problems. The embodiment of the application provides a compound all-in-one of fixed ball valve seat circle intelligence drilling of multistation, and it includes the frame, including the frame main part, set up in the work piece fixed platform and the setting of frame main part the universal removal chassis of the bottom of frame main part. The intelligent drilling and compounding all-in-one machine for the multi-station fixed ball valve seat ring further comprises an electric indexing chuck arranged on the workpiece fixing platform; and a plurality of drilling apparatuses mounted to the frame, wherein each of the drilling apparatuses includes a drilling device having a drill bit and a servo motor for driving the drilling device. In particular, the plurality of drilling apparatuses can be manipulated to simultaneously drill a plurality of locations of a workpiece to be machined to improve drilling efficiency and drilling accuracy.
According to an aspect of the application, a compound all-in-one of fixed ball valve seat circle intelligence drilling of multistation is provided, and it includes:
the device comprises a rack, a positioning device and a control device, wherein the rack comprises a rack main body, a workpiece fixing platform arranged on the rack main body and a universal moving underframe arranged at the bottom of the rack main body;
the electric indexing chuck is arranged on the workpiece fixing platform; and
a plurality of drilling apparatuses mounted to the frame, wherein each of the drilling apparatuses includes a drilling device having a drill bit and a servo motor for driving the drilling device.
In the compound all-in-one of fixed ball valve disk seat intelligence drilling of above-mentioned multistation, the compound all-in-one of fixed ball valve disk seat intelligence drilling of multistation further includes:
the device comprises a formed workpiece image acquisition module, a processing module and a display module, wherein the formed workpiece image acquisition module is used for acquiring first to sixth images of a processed and formed workpiece, and the first to sixth images are six views of the processed and formed workpiece;
the forming image coding module is used for enabling the first image to the sixth image of the machined and formed workpiece to pass through a first convolution neural network using a spatial attention mechanism to obtain a first characteristic map to a sixth characteristic map;
a difference module, configured to calculate a difference between each of the first to sixth feature maps and a feature map of a corresponding view angle in a seventh to twelfth feature maps to obtain first to sixth difference feature maps, where the seventh to twelfth feature maps are seventh to twelfth images of a standard-shaped workpiece generated by the first convolution neural network using the spatial attention mechanism, and the seventh to twelfth images are six views of the standard-shaped workpiece;
the correction module is used for clustering the differential feature maps in the first to sixth differential feature maps according to positions to obtain corrected first to sixth differential feature maps; and
and the drilling quality evaluation module is used for arranging the corrected first to sixth differential characteristic maps into an input tensor and then obtaining a classification result through a classifier, wherein the classification result is used for indicating whether the quality of the machined workpiece meets a preset standard.
In the compound all-in-one of fixed ball valve seat circle intelligence drilling of above-mentioned multistation, the shaping image coding module is further used for: using each layer of the first convolutional neural network to respectively perform input data in forward transmission of layers:
performing convolution processing on the input data based on a convolution kernel to obtain a convolution characteristic diagram;
passing the convolution feature map through a spatial attention module to obtain a spatial attention score map;
multiplying the spatial attention score map and the convolution feature map according to position points to obtain a spatial attention feature map;
pooling the spatial attention feature map along a channel dimension to obtain a pooled feature map; and
performing nonlinear activation on the pooled feature map to obtain an activated feature map;
wherein the output of the last layer of the second convolutional neural network is seventh to twelfth feature maps.
In the compound all-in-one of above-mentioned fixed ball valve seat circle intelligence drilling of multistation, difference module further is used for: calculating differences between each feature map in the first to sixth feature maps and the feature map of the corresponding view angle in the seventh to twelfth feature maps by the following formula to obtain first to sixth difference feature maps;
wherein the formula is:
F n ＝F i θF i+6
wherein, F i Showing each of the first to sixth characteristic diagrams, F i+6 A feature map indicating a view angle of each of the seventh to twelfth feature maps corresponding to each of the first to sixth feature maps,
indicates making a difference by position, and F n Each of the first to sixth difference feature maps is shown.
In the compound all-in-one of fixed ball valve seat circle intelligence drilling of above-mentioned multistation, correction module further is used for: clustering each differential feature map in the first to sixth differential feature maps according to positions by the following formula to obtain corrected first to sixth differential feature maps;
wherein the formula is:
wherein, F n Each of the first to sixth differential feature maps is represented,
means for representing the feature values of all the positions in each feature map,. Indicates a vector dot product, and F n ' denotes each of the corrected first to sixth differential feature maps.
In the compound all-in-one of above-mentioned fixed ball valve seat circle intelligence drilling of multistation, drilling quality evaluation module further is used for: the classifier is used for arranging the corrected first to sixth differential feature maps into an input tensor and then obtaining a classification result through the classifier;
wherein the formula is:
softmax{(W n ，B n )：...：(W 1 ，B 1 )|Project(F)}
wherein Project (F) represents projecting the input tensor as a vector, W 1 To W n As a weight matrix for each fully connected layer, B 1 To B n A bias matrix representing the fully connected layers of each layer.
Compared with the prior art, the utility model provides a pair of compound all-in-one of fixed ball valve disk seat intelligence drilling of multistation, it includes the frame, including the frame main part, set up in the work piece fixed platform and the setting of frame main part the universal removal chassis of the bottom of frame main part. The intelligent drilling and compounding all-in-one machine for the multi-station fixed ball valve seat ring further comprises an electric indexing chuck arranged on the workpiece fixing platform; and a plurality of drilling devices mounted to the frame, wherein each of the drilling devices includes a drilling device having a drill bit and a servo motor for driving the drilling device. In particular, the plurality of drilling apparatuses can be manipulated to simultaneously drill a plurality of locations of a workpiece to be machined to improve drilling efficiency and drilling accuracy.
Drawings
The above and other objects, features and advantages of the present application will become more apparent by describing in more detail embodiments of the present application with reference to the attached drawings. The accompanying drawings are included to provide a further understanding of the embodiments of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the principles of the application. In the drawings, like reference numbers generally represent like parts or steps.
Fig. 1 illustrates a schematic plan structure diagram of a multi-station fixed ball valve seat ring intelligent drilling and compounding all-in-one machine according to an embodiment of the application.
Fig. 2 illustrates a schematic diagram of a PLC control board assembly in a multi-station fixed ball valve seat ring intelligent drilling and compounding all-in-one machine according to an embodiment of the present application.
Fig. 3A illustrates a cross-sectional view of a machined workpiece of a multi-station fixed ball valve seat ring intelligent drilling compound all-in-one machine according to an embodiment of the application.
Fig. 3B illustrates a perspective view of a machined workpiece of the multi-station fixed ball valve seat ring intelligent drilling composite all-in-one machine according to an embodiment of the application.
Fig. 4 illustrates an application scenario of the intelligent drilling and compounding all-in-one machine for the multi-station fixed ball valve seat ring according to the embodiment of the application.
Fig. 5 illustrates a block diagram view of a multi-station fixed ball valve seat ring intelligent drilling compound all-in-one machine according to an embodiment of the application.
In the figure: 1. a frame; 11. a rack main body; 12. a workpiece fixing platform 13 and a universal movable underframe; 2. an electric index chuck; 3. drilling equipment; 31. a drilling device; 32. a servo motor; 4. a PLC control panel; 5. and (5) processing the workpiece.
Detailed Description
Hereinafter, example embodiments according to the present application will be described in detail with reference to the accompanying drawings. It should be understood that the described embodiments are only some embodiments of the present application and not all embodiments of the present application, and that the present application is not limited by the example embodiments described herein.
Summary of the application
Fig. 1 illustrates a schematic plane structure diagram of a multi-station fixed ball valve seat ring intelligent drilling and compounding all-in-one machine according to an embodiment of the application. As shown in fig. 1, the intelligent drilling and compounding all-in-one machine for the multi-station fixed ball valve seat ring mainly comprises: a frame 1; an electrically indexed chuck 2 and a drilling apparatus 3. More specifically, as shown in fig. 1, the machine frame 1 includes a machine frame main body 11, a workpiece fixing platform 12 provided on the machine frame main body 11, and a universally movable base frame 13 provided on the bottom of the machine frame main body 11, wherein the electric index chuck 2 is provided on the workpiece fixing platform 12; the plurality of drilling devices 3 are mounted to the rack body 11, wherein each of the drilling devices 3 includes a drilling means 31 having a drill bit and a servo motor 32 for driving the drilling means 31.
In the embodiment of the application, the electric indexing chuck 2 can be set by 360 degrees, so that 360-degree indexing processing is realized, and the punching angle is accurate in positioning.
It is worth mentioning that the intelligent drilling and compounding all-in-one machine for the multi-station fixed ball valve seat ring further comprises a cooling and cleaning device, a PLC control panel 4, a numerical control editing panel and other components. All control signals are sent by the PLC control board 4, the servo motor 32 of the main shaft is started by a control program of the PLC control board 4 to drive the main shaft to operate, the main shaft feeding operation is controlled by a ball screw, and the operation positioning precision is high. The cooling and cleaning device is an automatic cooling device, the cooling and cleaning device synchronously operates while drilling is synchronously processed, and the cooling and cleaning device is closed after drilling is finished; the phenomenon that cooling liquid flies randomly in the machining process is solved without manual control.
Fig. 2 illustrates a schematic diagram of an assembly of a PLC control board 4 in the intelligent all-in-one drilling and compounding machine for a multi-station fixed ball valve seat ring according to the embodiment of the present application, and as shown in fig. 2, the PLC control board 4 includes an air switch, a PLC, a wiring port, a frequency converter and other components.
In a specific embodiment of the present application, fig. 3A illustrates a cross-sectional view of a machined workpiece of a multi-station fixed ball valve seat ring intelligent drilling composite all-in-one machine according to an embodiment of the present application. Fig. 3B illustrates a perspective view of a machined workpiece of a multi-station fixed ball valve seat ring intelligent drilling compound all-in-one machine according to an embodiment of the application. Fig. 3A and 3B show a machined workpiece 5, in which both the horizontal direction and the vertical direction of the machined workpiece 5 are drilled.
More specifically, in an embodiment of the present application, the universal mobile chassis integrates the cooling water tank system and the PLC control board 4 together by using the internal space of the chassis, so as to reduce the overall space of the device, and the PLC control board 4 is also protected, thereby preventing dust and other conductive objects from contacting and damaging the PLC system.
In conclusion, through the combined structure, the intelligent drilling and combining all-in-one machine for the multi-station fixed ball valve seat ring can realize multi-station 360-degree adjustable drilling, namely multi-position and different-direction synchronous drilling, can synchronously complete multi-direction hole site processing, and the three-station drilling equipment 3 can be adjusted at will without influencing each other, so that the intelligent drilling and combining all-in-one machine is suitable for any angle and height adjustment processing. Meanwhile, a workpiece to be machined is clamped through the electric indexing chuck 2, the electric indexing chuck 2 is automatically tightened, digital input is controlled through a numerical control panel, feeding and retracting parameters of the drilling device 31 are adjusted, and chip breaking operation can be achieved in the machining process to protect a drill bit; the feeding of the drilling device 31 is controlled by a ball screw, the feeding position precision is accurate, and the signal feedback is fast.
Further, although the multi-station 360-degree adjustable drilling machining can be performed by the multi-station fixed ball valve seat ring intelligent drilling composite all-in-one machine, in the drilling machining process, due to the problems of vibration of equipment, uneven installation of the equipment, control precision in the adjusting process and the like, the quality of a machined and formed workpiece cannot meet preset requirements. Therefore, it is desirable to perform quality inspection of machined workpieces to recheck the quality of the drilled holes before the workpieces are delivered to ensure that the delivered workpieces meet quality standards.
Correspondingly, in the technical scheme of this application, the comparison between the work piece of accessible machine-shaping and the standard shaping work piece determines whether the drilling processingquality of the compound all-in-one of fixed ball valve seat circle intelligence drilling of multistation satisfies predetermined standard.
Specifically, first to sixth images of a machined and formed workpiece can be acquired through a plurality of cameras arranged on a multi-station fixed ball valve seat ring intelligent drilling and compounding all-in-one machine, and the first to sixth images are six views of the machined and formed workpiece. In particular, in the technical scheme of the application, a convolutional neural network model with excellent performance in the field of image feature extraction is used for image feature extraction. And considering that the forming quality and the forming position of the drill hole of each forming surface should be considered preferentially when performing quality evaluation, in the technical scheme of the application, image feature extraction is performed on each of the first to sixth images of the machined and formed workpiece by using a first convolution neural network model with a spatial attention mechanism so as to obtain first to sixth feature maps with strengthened drill hole feature space.
Accordingly, in the technical solution of the present application, the sixth view of the standard formed workpiece is also encoded with the first convolutional neural network model with spatial attention mechanism to obtain the seventh to twelfth feature maps. Then, the difference between each of the first to sixth feature maps and the feature map of the corresponding view angle in the seventh to twelfth feature maps is calculated to obtain first to sixth difference feature maps. It should be understood that the first through sixth differential feature maps described herein represent feature representations in a high-dimensional feature space for the differences between the views from various perspectives of a machined as-machined workpiece and the views from various perspectives of a standard as-machined workpiece.
Since the first to sixth differential feature maps correspond to the visual features of six views of the workpiece, that is, to different view-angle phases constituting a complete view-angle period, in order to improve the classification capability of the first to sixth differential feature maps themselves for the classification probability, the feature-value expression of each feature map actually has a periodically expressed phase attribute, and the feature maps for phase perception of each feature map are aggregated by location, as follows:
the optimized phase perception characterization of the feature map introduces amplitude-phase class real value-virtual value characterization, and the feature values are subjected to position-based splicing expansion of the real value map based on the principle of an Euler formula to perform position-based aggregation, so that inductive bias possibly caused when a real value classification task without position attribute (linear projection is used as a feature vector when classification is performed in a classifier) is performed on the feature map is compensated in a multi-layer perception mode, and the classification capability of the first to sixth differential feature maps under the classification probability of the classifier is improved.
And further arranging the corrected first to sixth difference characteristic diagrams into an input tensor, and then obtaining a classification result through a classifier, wherein the classification result is used for indicating whether the quality of the machined and formed workpiece meets a preset standard.
Based on this, the application provides a compound all-in-one of fixed ball valve seat circle intelligence drilling of multistation further includes: the device comprises a formed workpiece image acquisition module, a processing module and a display module, wherein the formed workpiece image acquisition module is used for acquiring first to sixth images of a processed and formed workpiece, and the first to sixth images are six views of the processed and formed workpiece; the forming image coding module is used for enabling the first image to the sixth image of the machined and formed workpiece to pass through a first convolution neural network using a spatial attention mechanism to obtain a first characteristic map to a sixth characteristic map; a difference module, configured to calculate a difference between each of the first to sixth feature maps and a feature map of a corresponding view angle in a seventh to twelfth feature maps to obtain first to sixth difference feature maps, where the seventh to twelfth feature maps are seventh to twelfth images of a standard-shaped workpiece generated by the first convolution neural network using the spatial attention mechanism, and the seventh to twelfth images are six views of the standard-shaped workpiece; the correction module is used for clustering the differential feature maps in positions in the first to sixth differential feature maps respectively to obtain corrected first to sixth differential feature maps; and the drilling quality evaluation module is used for arranging the corrected first to sixth differential characteristic maps into an input tensor and then obtaining a classification result through a classifier, wherein the classification result is used for indicating whether the quality of the machined and formed workpiece meets a preset standard or not.
Fig. 4 illustrates an application scenario of the intelligent drilling and compounding all-in-one machine for the multi-station fixed ball valve seat ring according to the embodiment of the application. As shown in fig. 4, in this application scenario, first, six views of a machined workpiece are acquired by a plurality of cameras (e.g., C illustrated in fig. 1) disposed in a multi-station fixed ball valve seat ring intelligent drilling and compounding machine (e.g., M illustrated in fig. 1), and then six views of a standard workpiece are acquired by the plurality of cameras. Then, the acquired six views of the machined and formed workpiece and the six views of the standard and formed workpiece are input into a server (e.g., S illustrated in fig. 1) deployed with a detection algorithm of a multi-station fixed ball valve seat ring intelligent drilling and compositing machine, wherein the server can process the six views of the machined and formed workpiece and the six views of the standard and formed workpiece using the detection algorithm of the multi-station fixed ball valve seat ring intelligent drilling and compositing machine to generate a classification result indicating whether the quality of the machined and formed workpiece meets a predetermined standard.
Exemplary System
Fig. 5 illustrates a block diagram view of a multi-station fixed ball valve seat ring intelligent drilling compound all-in-one machine according to an embodiment of the application. As shown in fig. 5, the intelligent drilling and compounding all-in-one machine 100 for a multi-station fixed ball valve seat ring according to the embodiment of the present application further includes: the device comprises a formed workpiece image acquisition module 110, a processing module and a processing module, wherein the formed workpiece image acquisition module is used for acquiring first to sixth images of a processed and formed workpiece, and the first to sixth images are six views of the processed and formed workpiece; a forming image coding module 120, configured to pass the first to sixth images of the machined and formed workpiece through a first convolution neural network using a spatial attention mechanism to obtain first to sixth feature maps, respectively; a difference module 130, configured to calculate a difference between each of the first to sixth feature maps and a feature map of a corresponding view angle in a seventh to twelfth feature maps to obtain first to sixth difference feature maps, where the seventh to twelfth feature maps are seventh to twelfth images of a standard-shaped workpiece generated by the first convolution neural network using the spatial attention mechanism, and the seventh to twelfth images are six views of the standard-shaped workpiece; a correcting module 140, configured to perform position-based clustering on each of the first to sixth differential feature maps to obtain corrected first to sixth differential feature maps; and a drilling quality evaluation module 150, configured to arrange the corrected first to sixth differential feature maps into an input tensor and then pass through a classifier to obtain a classification result, where the classification result is used to indicate whether the quality of the machined and formed workpiece meets a predetermined standard.
In an embodiment of the present application, the formed workpiece image capturing module 110 is configured to obtain first to sixth images of a formed workpiece, where the first to sixth images are six views of the formed workpiece. It should be understood that in the technical scheme of this application, confirm whether the drilling processingquality of the compound all-in-one of multistation fixed ball valve seat circle intelligence drilling satisfies predetermined standard through the comparison between the work piece of machine-shaping and the standard shaping work piece. That is, the quality of the drilled hole is rechecked prior to the ex-warehouse of the workpiece based on the comparison between the machined and standard formed workpieces to ensure that the ex-warehouse workpiece meets the quality standards.
In a specific embodiment of the application, first to sixth images of a machined workpiece are acquired through a plurality of cameras deployed on a multi-station fixed ball valve seat ring intelligent drilling and compounding all-in-one machine, and the first to sixth images are six views of the machined workpiece.
In an embodiment of the present application, the formed image encoding module 120 is configured to pass the first to sixth images of the machined and formed workpiece through a first convolutional neural network using a spatial attention mechanism to obtain first to sixth feature maps, respectively. It should be understood that, in the technical solution of the present application, a convolutional neural network model having excellent performance in the field of image feature extraction is used for image feature extraction. And considering that the forming quality and the forming position of the drill hole of each forming surface should be considered preferentially when performing quality evaluation, in the technical scheme of the application, image feature extraction is performed on each of the first to sixth images of the machined and formed workpiece by using a first convolution neural network model with a spatial attention mechanism so as to obtain first to sixth feature maps with strengthened drill hole feature space.
In a specific embodiment of the present application, the formed image encoding module 120 is further configured to: using each layer of the first convolutional neural network to respectively perform input data in forward transmission of layers:
performing convolution processing on the input data based on a convolution kernel to obtain a convolution characteristic diagram; passing the convolved feature maps through a spatial attention module to obtain a spatial attention score map; multiplying the spatial attention score map and the convolution feature map according to position points to obtain a spatial attention feature map; pooling the spatial attention feature map along a channel dimension to obtain a pooled feature map; and performing nonlinear activation on the pooled feature map to obtain an activated feature map; wherein the output of the last layer of the second convolutional neural network is seventh to twelfth feature maps.
In an embodiment of the present application, the difference module 130 is configured to calculate a difference between each of the first to sixth feature maps and a feature map of a corresponding view angle in the seventh to twelfth feature maps to obtain first to sixth difference feature maps, where the seventh to twelfth feature maps are seventh to twelfth images of a standard-shaped workpiece generated by the first convolutional neural network using the spatial attention mechanism, and the seventh to twelfth images are six views of the standard-shaped workpiece. It should be understood that, considering that the present application detects the drilling quality of the intelligent drilling and compositing all-in-one machine for the multi-station fixed ball valve seat ring through comparison between a machined and formed workpiece and a standard and formed workpiece, first to sixth images of the standard and formed workpiece are also obtained, and six views of the standard and formed workpiece are also encoded by the first convolution neural network model with the spatial attention mechanism to obtain seventh to twelfth characteristic maps. Then, the difference between each of the first to sixth feature maps and the feature map of the corresponding view angle in the seventh to twelfth feature maps is calculated to obtain first to sixth difference feature maps. It should be understood that the first through sixth differential feature maps described herein represent feature representations in a high-dimensional feature space for the differences between the views from various perspectives of a machined as-machined workpiece and the views from various perspectives of a standard as-machined workpiece.
In one embodiment of the application, the standard formed workpiece is placed at the same position when the machined formed workpiece is shot, and the seventh to twelfth images of the standard formed workpiece are calibrated through a plurality of cameras which are arranged on the intelligent drilling and compounding all-in-one machine of the multi-station fixed ball valve seat ring, wherein the first to sixth images are six views of the standard formed workpiece.
In a specific embodiment of the present application, the difference module 130 is further configured to: calculating differences between each feature map in the first to sixth feature maps and the feature map of the corresponding view angle in the seventh to twelfth feature maps by the following formula to obtain first to sixth difference feature maps;
wherein the formula is:
wherein, F i Showing each of the first to sixth characteristic diagrams, F i+6 A feature map indicating a view angle of each of the seventh to twelfth feature maps corresponding to each of the first to sixth feature maps,
indicates making a difference by position, and F n To representEach of the first to sixth differential feature maps.
In this embodiment of the application, the correcting module 140 is configured to perform position-based clustering on each of the first to sixth differential feature maps to obtain the corrected first to sixth differential feature maps. It should be understood that, since the first to sixth differential feature maps correspond to the visual features of the six views of the workpiece, that is, to the different viewing angle phases constituting the complete viewing angle period, in order to improve the classification capability of the first to sixth differential feature maps themselves for the classification probability, the phase-aware feature maps of each feature map are aggregated by location for the phase attribute that the feature value expression of each feature map actually has a periodic expression.
In a specific embodiment of the present application, the correction module 140 is further configured to: clustering the differential feature maps in the first to sixth differential feature maps according to the position respectively by the following formula to obtain corrected first to sixth differential feature maps;
wherein the formula is:
wherein, F n Each of the first to sixth differential feature maps is represented,
means for representing the feature values of all the positions in each feature map,. Indicates a vector dot product, and F n ' denotes each of the corrected first to sixth differential feature maps.
The optimized phase perception characterization of the feature map introduces amplitude-phase class real value-virtual value characterization, and the feature values are subjected to position-based splicing expansion of the real value map based on the principle of an Euler formula to perform position-based aggregation, so that inductive bias possibly caused when a real value classification task without position attribute (linear projection is used as a feature vector when classification is performed in a classifier) is performed on the feature map is compensated in a multi-layer perception mode, and the classification capability of the first to sixth differential feature maps under the classification probability of the classifier is improved.
In this embodiment, the drilling quality evaluation module 150 is configured to arrange the corrected first to sixth differential feature maps into an input tensor and then pass through a classifier to obtain a classification result, where the classification result is used to indicate whether the quality of the machined workpiece meets a predetermined criterion.
In a specific embodiment of the present application, the borehole quality evaluation module 150 is further configured to: the corrected first to sixth differential feature maps are arranged into an input tensor and then are subjected to classifier to obtain a classification result;
wherein the formula is: softmax { (W) n ，B n )：...：(W 1 ，B 1 ) L Project (F) }, where Project (F) denotes the projection of the input tensor as a vector, W 1 To W n As a weight matrix for each fully connected layer, B 1 To B n A bias matrix representing the layers of the fully connected layer.
In one embodiment of the present application, the classifier includes at least one fully-connected layer and a Softmax classification function, where at least one fully-connected layer is used to fully-connect encode the input tensor to fully utilize the information of each position in the input tensor and obtain the classification eigenvector, that is, in this embodiment, the classifier takes at least one fully-connected layer as an encoder to project the input tensor as a one-dimensional eigenvector. Next, the categorizing feature vector is input to the Softmax categorizing function to calculate a Softmax function value of the categorizing feature vector, i.e., a probability value that the categorizing feature vector belongs to each categorizing label, and in the embodiment of the present application, the quality of the work-formed workpiece is a quality that satisfies a predetermined criterion (first label) and whether the quality of the work-formed workpiece satisfies the predetermined criterion (second label). And finally, taking the label corresponding to the larger probability value as the classification result.
In summary, according to the intelligent drilling and compositing all-in-one machine for the multi-station fixed ball valve seat ring, feature extraction and differential feature map calculation are respectively performed on six views of a machined workpiece and six views of a standard workpiece through a first convolution neural network using a spatial attention mechanism to obtain first to sixth differential feature maps, then, each of the first to sixth differential feature maps is clustered according to position to obtain first to sixth corrected differential feature maps, and the first to sixth corrected differential feature maps are arranged into a classification result which is input and then passes through a classifier to obtain whether quality of the machined workpiece meets a predetermined standard, so that quality detection is performed on the machined workpiece to recheck drilling quality before the workpiece is delivered out of a warehouse to ensure that the delivered workpiece can meet the quality standard.
The foregoing describes the general principles of the present application in conjunction with specific embodiments, however, it is noted that the advantages, effects, etc. mentioned in the present application are merely examples and are not limiting, and they should not be considered essential to the various embodiments of the present application. Furthermore, the foregoing disclosure of specific details is for the purpose of illustration and description and is not intended to be limiting, since the foregoing disclosure is not intended to be exhaustive or to limit the disclosure to the precise details disclosed.
The block diagrams of devices, apparatuses, devices, systems referred to in this application are only used as illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. These devices, apparatuses, devices, systems may be connected, arranged, configured in any manner, as will be appreciated by those skilled in the art. Words such as "including," "comprising," "having," and the like are open-ended words that mean "including, but not limited to," and are used interchangeably therewith. As used herein, the words "or" and "refer to, and are used interchangeably with, the word" and/or, "unless the context clearly dictates otherwise. The word "such as" is used herein to mean, and is used interchangeably with, the phrase "such as but not limited to".
It should also be noted that in the devices, apparatuses, and methods of the present application, the components or steps may be decomposed and/or recombined. These decompositions and/or recombinations are to be considered as equivalents of the present application.
The previous description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects without departing from the scope of the application. Thus, the present application is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
The foregoing description has been presented for purposes of illustration and description. Furthermore, the description is not intended to limit embodiments of the application to the form disclosed herein. While a number of example aspects and embodiments have been discussed above, those of skill in the art will recognize certain variations, modifications, alterations, additions and sub-combinations thereof.
Claims (6)
1. The utility model provides a compound all-in-one of fixed ball valve disk seat circle intelligence drilling of multistation which characterized in that includes:
the device comprises a rack, a clamping device and a clamping device, wherein the rack comprises a rack main body, a workpiece fixing platform arranged on the rack main body and a universal moving underframe arranged at the bottom of the rack main body;
the electric indexing chuck is arranged on the workpiece fixing platform; and
a plurality of drilling apparatuses mounted to the frame, wherein each of the drilling apparatuses includes a drilling device having a drill bit and a servo motor for driving the drilling device.
2. The intelligent multi-station fixed ball valve seat ring drilling and compounding all-in-one machine according to claim 1, further comprising:
the device comprises a formed workpiece image acquisition module, a processing module and a processing module, wherein the formed workpiece image acquisition module is used for acquiring first to sixth images of a processed and formed workpiece, and the first to sixth images are six views of the processed and formed workpiece;
the forming image coding module is used for enabling the first image to the sixth image of the machined and formed workpiece to pass through a first convolution neural network using a spatial attention mechanism to obtain a first characteristic map to a sixth characteristic map;
a difference module, configured to calculate a difference between each of the first to sixth feature maps and a feature map of a corresponding view angle in a seventh to twelfth feature maps to obtain first to sixth difference feature maps, where the seventh to twelfth feature maps are seventh to twelfth images of a standard-shaped workpiece generated by the first convolution neural network using the spatial attention mechanism, and the seventh to twelfth images are six views of the standard-shaped workpiece;
the correction module is used for clustering the differential feature maps in the first to sixth differential feature maps according to positions to obtain corrected first to sixth differential feature maps; and
and the drilling quality evaluation module is used for arranging the corrected first to sixth differential characteristic maps into an input tensor and then obtaining a classification result through a classifier, wherein the classification result is used for indicating whether the quality of the machined workpiece meets a preset standard.
3. The intelligent all-in-one machine of claim 2, wherein the molded image coding module is further configured to: using each layer of the first convolutional neural network to respectively perform input data in forward transmission of layers:
performing convolution processing on the input data based on a convolution kernel to obtain a convolution characteristic diagram;
passing the convolved feature maps through a spatial attention module to obtain a spatial attention score map;
multiplying the spatial attention score map and the convolution feature map according to position points to obtain a spatial attention feature map;
pooling the spatial attention feature map along a channel dimension to obtain a pooled feature map; and
performing nonlinear activation on the pooled feature map to obtain an activated feature map;
wherein the output of the last layer of the second convolutional neural network is seventh to twelfth feature maps.
4. The intelligent integrated machine for drilling and compounding multi-station fixed ball valve seat rings according to claim 3, wherein the difference module is further configured to: calculating differences between each feature map in the first to sixth feature maps and the feature map of the corresponding view angle in the seventh to twelfth feature maps by the following formula to obtain first to sixth difference feature maps;
wherein the formula is:
wherein, F i Showing each of the first to sixth characteristic diagrams, F i+6 A feature map indicating a view angle of each of the seventh to twelfth feature maps corresponding to each of the first to sixth feature maps,
indicates making a difference by position, and F n Each of the first to sixth differential feature maps is shown.
5. The intelligent drilling and compounding all-in-one machine of claim 4, wherein the correction module is further configured to: clustering each differential feature map in the first to sixth differential feature maps according to positions by the following formula to obtain corrected first to sixth differential feature maps;
wherein the formula is:
wherein, F n Each of the first to sixth differential feature maps is represented,
means for representing the feature values of all the positions in each feature map,. Indicates a vector dot product, and F n ' denotes each of the corrected first to sixth differential feature maps.
6. The intelligent multi-station fixed ball valve seat ring drilling and compounding all-in-one machine of claim 5, wherein the drilling quality evaluation module is further configured to: the classifier is used for arranging the corrected first to sixth differential feature maps into an input tensor and then obtaining a classification result through the classifier;
wherein the formula is:
softmax{(W n ,B n ):…:(W 1 ,B 1 )|Project(F)}
wherein Project (F) represents projecting the input tensor as a vector, W 1 To W n As a weight matrix for each fully connected layer, B 1 To B n A bias matrix representing the layers of the fully connected layer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210983370.6A CN115502443B (en) | 2022-08-16 | 2022-08-16 | Intelligent drilling and compounding integrated machine for multi-station fixed ball valve seat ring |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210983370.6A CN115502443B (en) | 2022-08-16 | 2022-08-16 | Intelligent drilling and compounding integrated machine for multi-station fixed ball valve seat ring |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115502443A true CN115502443A (en) | 2022-12-23 |
| CN115502443B CN115502443B (en) | 2023-05-12 |
Family
ID=84502645
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210983370.6A Active CN115502443B (en) | 2022-08-16 | 2022-08-16 | Intelligent drilling and compounding integrated machine for multi-station fixed ball valve seat ring |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115502443B (en) |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104772659A (en) * | 2015-04-20 | 2015-07-15 | 江苏胜德龙机电科技有限公司 | Numerically-controlled multi-station composite processing special machine tool for valve body of butterfly valve |
| CN107322306A (en) * | 2017-08-30 | 2017-11-07 | 重庆中天模具有限公司 | Rotating disc type indexes slide unit multistation drilling and tapping machine |
| CN109332928A (en) * | 2018-10-23 | 2019-02-15 | 江苏山扬智能装备有限公司 | Street lamp post robot welding system and welding method based on deep learning on-line checking |
| US20190122378A1 (en) * | 2017-04-17 | 2019-04-25 | The United States Of America, As Represented By The Secretary Of The Navy | Apparatuses and methods for machine vision systems including creation of a point cloud model and/or three dimensional model based on multiple images from different perspectives and combination of depth cues from camera motion and defocus with various applications including navigation systems, and pattern matching systems as well as estimating relative blur between images for use in depth from defocus or autofocusing applications |
| CN110340664A (en) * | 2019-07-19 | 2019-10-18 | 温州捷凯智能设备科技有限公司 | A kind of automatic multi-station valve drilling and tapping compounding machine |
| CN110666793A (en) * | 2019-09-11 | 2020-01-10 | 大连理工大学 | Method for realizing robot square part assembly based on deep reinforcement learning |
| EP3623882A1 (en) * | 2018-09-13 | 2020-03-18 | Siemens Aktiengesellschaft | Identifying type and alignment of a workpiece |
| CN112222450A (en) * | 2018-11-19 | 2021-01-15 | 莆田市荣兴机械有限公司 | Drilling method of workpiece multi-station synchronous drilling equipment |
| CN113226612A (en) * | 2018-11-22 | 2021-08-06 | 普雷茨特两合公司 | Identification of processing defects in laser processing systems by means of deep convolutional neural networks |
| CN114283117A (en) * | 2021-11-24 | 2022-04-05 | 广西大学 | Insulator defect detection method based on improved YOLOv3 convolutional neural network |
| CN114871486A (en) * | 2022-06-21 | 2022-08-09 | 大畏机床(江苏)有限公司 | Double housing planer and processing control method thereof |
-
2022
- 2022-08-16 CN CN202210983370.6A patent/CN115502443B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104772659A (en) * | 2015-04-20 | 2015-07-15 | 江苏胜德龙机电科技有限公司 | Numerically-controlled multi-station composite processing special machine tool for valve body of butterfly valve |
| US20190122378A1 (en) * | 2017-04-17 | 2019-04-25 | The United States Of America, As Represented By The Secretary Of The Navy | Apparatuses and methods for machine vision systems including creation of a point cloud model and/or three dimensional model based on multiple images from different perspectives and combination of depth cues from camera motion and defocus with various applications including navigation systems, and pattern matching systems as well as estimating relative blur between images for use in depth from defocus or autofocusing applications |
| CN107322306A (en) * | 2017-08-30 | 2017-11-07 | 重庆中天模具有限公司 | Rotating disc type indexes slide unit multistation drilling and tapping machine |
| EP3623882A1 (en) * | 2018-09-13 | 2020-03-18 | Siemens Aktiengesellschaft | Identifying type and alignment of a workpiece |
| CN109332928A (en) * | 2018-10-23 | 2019-02-15 | 江苏山扬智能装备有限公司 | Street lamp post robot welding system and welding method based on deep learning on-line checking |
| CN112222450A (en) * | 2018-11-19 | 2021-01-15 | 莆田市荣兴机械有限公司 | Drilling method of workpiece multi-station synchronous drilling equipment |
| CN113226612A (en) * | 2018-11-22 | 2021-08-06 | 普雷茨特两合公司 | Identification of processing defects in laser processing systems by means of deep convolutional neural networks |
| CN110340664A (en) * | 2019-07-19 | 2019-10-18 | 温州捷凯智能设备科技有限公司 | A kind of automatic multi-station valve drilling and tapping compounding machine |
| CN110666793A (en) * | 2019-09-11 | 2020-01-10 | 大连理工大学 | Method for realizing robot square part assembly based on deep reinforcement learning |
| CN114283117A (en) * | 2021-11-24 | 2022-04-05 | 广西大学 | Insulator defect detection method based on improved YOLOv3 convolutional neural network |
| CN114871486A (en) * | 2022-06-21 | 2022-08-09 | 大畏机床(江苏)有限公司 | Double housing planer and processing control method thereof |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115502443B (en) | 2023-05-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108312144B (en) | Robot automatic locking control system and method based on machine vision | |
| US11225018B2 (en) | 3D printing method and device with multi-axis mechanical system and visual surveillance | |
| CN111062990A (en) | Binocular vision positioning method for underwater robot target grabbing | |
| CN108932736A (en) | Two-dimensional laser radar Processing Method of Point-clouds and dynamic robot pose calibration method | |
| US7403648B2 (en) | Apparatus for generating three-dimensional model data | |
| CN111230862B (en) | Handheld workpiece deburring method and system based on visual recognition function | |
| CN1291822C (en) | System and method for generating cutting path automatically | |
| CN115239515B (en) | Precise intelligent processing and manufacturing system for mechanical parts and manufacturing method thereof | |
| CN103354770A (en) | Laser processing method and laser processing device | |
| CN114871486A (en) | Double housing planer and processing control method thereof | |
| CN107000091A (en) | The geometry control of electro-discharge machining cutter and best fit | |
| CN114571160A (en) | Offline curved surface weld extraction and attitude estimation method | |
| CN112697112A (en) | Method and device for measuring horizontal plane inclination angle of camera | |
| CN115502443A (en) | Multi-station intelligent drilling and compounding integrated machine for fixed ball valve seat rings | |
| CN107932502A (en) | A kind of SCARA method for planning track of robot based on binocular stereo vision | |
| Ben et al. | Research on visual orientation guidance of industrial robot based on cad model under binocular vision | |
| CN203125521U (en) | Three-dimensional (3D) binocular-vision industrial robot | |
| CN108460797B (en) | Method and device for calculating relative pose of depth camera and height of scene plane | |
| CN105824237A (en) | Line-laser-sensor-based adaptive deviation control method | |
| JPH08236995A (en) | Method of mounting chip | |
| CN108520533B (en) | Workpiece positioning-oriented multi-dimensional feature registration method | |
| CN113793313B (en) | High-precision tool setting method and device for machining full-surface micro-pit structure of thin-wall spherical shell type micro-component | |
| CN112312666B (en) | Circuit board screw driving method and system | |
| CN111331609B (en) | Method, device and system for acquiring preferred embodiment of robot | |
| CN111906770A (en) | Workpiece mounting method and system, computer readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |