Automatic processing method and system for a pin

Abstract

The embodiment of the specification provides a pin shaft automatic processing method and system. The method is executed by a management platform, and the method comprises the following steps: acquiring feature information of a to-be-processed blank part, the feature information comprising two-dimensional feature information of the to-be-processed blank part, the two-dimensional feature information being related to image information of the to-be-processed blank part and / or image features extracted based on the image information; predicting an estimated quality of the to-be-processed blank part after being processed into a pin shaft based on the feature information and pin shaft processing process parameters; in response to the estimated quality satisfying a quality preset condition, transporting the to-be-processed blank part to a target station by a grabbing device; processing the to-be-processed blank part based on the pin shaft processing process parameters, comprising: performing cold heading process treatment on the to-be-processed blank part based on cold heading process parameters to obtain a semi-finished product; performing semi-finished product inspection in the process of the cold heading process treatment; performing thread rolling process treatment on the semi-finished product based on thread rolling process parameters to obtain a finished product.

Description

TECHNICAL FIELD

[0001] The present specification relates to the field of pin shaft processing, in particular to a pin shaft automatic processing method and system. BACKGROUND

[0002] Pin shaft processing is a common part processing. In the process of pin shaft processing, various equipment failure problems may exist, which may cause abnormal processing parts. However, due to the large number of processes and fast processing speed, the staff cannot timely find the abnormal conditions in the pin shaft processing process, which causes waste of raw materials. Moreover, even if the abnormality of the processing part is found, it is difficult to accurately determine the specific fault position and fault reason of the equipment based on the abnormal conditions of the processing part through manual judgment and historical experience.

[0003] Therefore, it is urgent to propose a pin shaft automatic processing method and system to timely and accurately control the quality of the pin shaft processing process, reasonably adjust and optimize the entire pin shaft processing process based on the quality of the processing part, improve the quality of the pin shaft processing, and ensure the efficient processing of the pin shaft. SUMMARY

[0004] One or more embodiments of the present specification provide a pin shaft automatic processing method, which is executed by a management platform, and the method comprises: obtaining feature information of a to-be-processed blank part, the feature information comprising two-dimensional feature information of the to-be-processed blank part, the two-dimensional feature information being related to image information of the to-be-processed blank part and / or image features extracted based on the image information; predicting an estimated quality of the to-be-processed blank part processed into a pin shaft based on the feature information and pin shaft processing process parameters; in response to the estimated quality satisfying a quality preset condition, transporting the to-be-processed blank part to a target station by a grabbing device; processing the to-be-processed blank part based on the pin shaft processing process parameters, the pin shaft processing process parameters comprising cold heading process parameters and threading process parameters, and the processing of the to-be-processed blank part based on the pin shaft processing process parameters comprising: performing cold heading process treatment on the to-be-processed blank part based on the cold heading process parameters to obtain a semi-finished product; performing semi-finished product inspection during the cold heading process treatment; and performing threading process treatment on the semi-finished product based on the threading process parameters to obtain a finished product.

[0005] One of the embodiments of the present specification provides a pin shaft automatic processing system, the system comprises a management platform, the system further comprises: an acquisition module configured to acquire feature information of a to-be-processed blank part, the feature information comprising two-dimensional feature information of the to-be-processed blank part, the two-dimensional feature information being related to image information of the to-be-processed blank part and / or image features extracted based on the image information; a prediction module configured to predict an estimated quality of the to-be-processed blank part processed into a pin shaft based on the feature information and pin shaft processing parameters; a grabbing module configured to transport the to-be-processed blank part to a target station by a grabbing device in response to the estimated quality satisfying a quality preset condition; and a processing module configured to process the to-be-processed blank part based on the pin shaft processing parameters, the pin shaft processing parameters comprising cold heading process parameters and threading process parameters, the processing module being further configured to: perform cold heading process treatment on the to-be-processed blank part based on the cold heading process parameters to obtain a semi-finished product; perform semi-finished product inspection during the cold heading process treatment; and perform threading process treatment on the semi-finished product based on the threading process parameters to obtain a finished product.

[0006] One or more embodiments of the present specification provide a pin shaft automatic processing device comprising a processor configured to execute a pin shaft automatic processing method.

[0007] One or more embodiments of the present specification provide a computer readable storage medium storing computer instructions, when a computer reads the computer instructions in the storage medium, the computer executes a pin shaft automatic processing method. BRIEF DESCRIPTION OF DRAWINGS

[0008] The present specification will be further illustrated in the form of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting, and in these embodiments, the same numbers represent the same structures, wherein:

[0009] Figure 1 is an application scenario schematic diagram of a pin shaft automatic processing system according to some embodiments of the present specification;

[0010] Figure 2 is an exemplary block diagram of a pin shaft automatic processing system according to some embodiments of the present specification;

[0011] Figure 3 is an exemplary flowchart of a pin shaft automatic processing method according to some embodiments of the present specification;

[0012] Figure 4 is an exemplary flowchart of determining three-dimensional feature information according to some embodiments of the present specification;

[0013] Figure 5 is an exemplary schematic diagram of a three-dimensional model generation model according to some embodiments of the present specification;

[0014] Figure 6 is an exemplary flowchart of performing semi-finished product inspection according to some embodiments of the present specification;

[0015] Figure 7 is an exemplary schematic diagram of pin shaft automatic machining process according to some embodiments of the present specification;

[0016] Figure 8 is an exemplary flowchart of determining pin shaft machining process optimization parameters according to some embodiments of the present specification. DETAILED DESCRIPTION

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present specification, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are only some examples or embodiments of the present specification, and for those skilled in the art, the present specification can also be applied to other similar scenarios without creative labor. Unless it is clear from the language context or otherwise stated, the same reference numbers in the drawings represent the same structure or operation.

[0018] It should be understood that the "system", "device", "unit" and / or "module" used herein is a method for distinguishing different components, elements, parts, portions or assemblies at different levels. However, if other words can achieve the same purpose, the words can be replaced by other expressions.

[0019] As shown in the specification and claims, unless the context clearly indicates otherwise, "one", "a", "an", and / or "the" do not necessarily mean singular, but can also include plural. Generally, the terms "include" and "contain" only indicate the inclusion of the steps and elements explicitly identified, and these steps and elements do not constitute an exclusive list, and the method or device can also include other steps or elements.

[0020] Flowcharts are used in the present specification to illustrate the operations performed by the system according to the embodiments of the present specification. It should be understood that the preceding or subsequent operations do not necessarily have to be performed in sequence. On the contrary, each step can be processed in reverse order or simultaneously. At the same time, other operations can also be added to these processes, or one or more steps of the operation can be removed from these processes.

[0021] A pin shaft is a kind of standardized fastener, which can be used for static fixed connection or relative motion with the connected part, mainly used for the hinge connection of two parts. The pin shaft is usually locked by a split pin, which is reliable in work and convenient to disassemble. The pin shaft can be automatically processed by a pin shaft automatic processing system. Before the processing starts, the management platform of the pin shaft automatic processing system can predict the quality of the pin shaft after the processing of the blank part, and the blank part meeting the quality condition is sent to each process (such as the cold heading process, the thread rolling process, etc.) in turn for processing, and the semi-finished product and the finished product are inspected, so as to adjust the pin shaft automatic processing process and the pin shaft processing parameters in time in combination with the quality of the semi-finished product and the finished product, to ensure the high quality of the pin shaft and the orderly progress of the pin shaft automatic processing process.

[0022] Figure 1 It is a schematic diagram of the application scene of the pin shaft automatic processing system according to some embodiments of the present specification.

[0023] In some embodiments, the application scene 100 of the pin shaft automatic processing system can include a management platform 110, a network 120 and a pin shaft processing system 130.

[0024] The management platform 110 refers to a platform that collects all data of the pin shaft automatic processing process and executes the pin shaft automatic processing method. In some embodiments, the management platform 110 can include a terminal device 111, a processor 112 and a storage device 113.

[0025] The terminal device 111 can refer to one or more terminal devices or software used by the user. In some embodiments, the terminal device 111 can be a mobile terminal device. For example, a tablet computer, a laptop computer, etc. In some embodiments, the terminal device 111 can be a fixed terminal device. For example, the terminal device 111 can be directly installed on the processor 112, becoming a part of the processor 112. In some embodiments, the terminal device 111 can include a signal transmitter and a signal receiver, which are configured to communicate with the pin shaft processing system 130 through the network 120 to obtain the related information of the pin shaft processing.

[0026] In some embodiments, the terminal device 111 can receive the request information of the user and send the request information to the processor 112 via the network 120. For example, the terminal device 111 can receive the request information of the user requiring to send the related information of the pin shaft processing, and send the request information to the processor 112 via the network 120. The terminal device 111 can also receive the information from the processor 112 via the network 120. For example, the terminal device 111 can receive the related information of the pin shaft processing system 130 from the processor 112, and the determined one or more related information can be displayed on the terminal device 111.

[0027] The processor 112 can be configured to process data and / or information from at least one component of the application scenario 100 or an external data source (e.g., a cloud data center). The processor 112 can be connected to the terminal device 111, the storage device 113, and / or the pin shaft machining system 130 via the network 120 to access and / or receive data and information. For example, the processor 112 can receive relevant information of the pin shaft machining system 130 (e.g., image information of a blank to be machined, etc.) via the network 120.

[0028] In some embodiments, the processor 112 can be a single processor or a group of processors. The group of processors can be centralized or distributed (e.g., the processor 112 can be a distributed system), and can be dedicated or simultaneously provided by other devices or systems. In some embodiments, the processor 112 can be locally connected to the network 120 or remotely connected to the network 120. In some embodiments, the processor 112 can be implemented on a cloud platform. By way of example only, the cloud platform can include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an on-premises cloud, a multi-layer cloud, etc., or any combination thereof.

[0029] The storage device 113 can be configured to store data and / or instructions. The data can include data related to the pin shaft machining system 130, etc. In some embodiments, the storage device 113 can store data and / or instructions used by the processor 112 to perform or use to complete the exemplary methods described in this specification. For example, the storage device 113 can store relevant information of the pin shaft machining system 130. For another example, the storage device 113 can store one or more machine learning models. In some embodiments, the storage device 113 can be part of the processor 112.

[0030] In some embodiments, the storage device 113 can include a mass storage, a removable storage, a volatile read-write memory, a read-only memory (ROM), etc., or any combination thereof. In some embodiments, the storage device 113 can be implemented on a cloud platform. In some embodiments, the storage device 113 can be connected to the network 120 to communicate with one or more components of the application scenario 100 (e.g., the processor 112, the pin shaft machining system 130).

[0031] The network 120 can facilitate the exchange of information and / or data. In some embodiments, one or more components in the application scenario 100 (e.g., the processor 112, the storage device 113, the pin shaft machining system 130) can send information and / or data to another component in the application scenario 100 via the network 120. The network 120 can include a local area network (LAN), a wide area network (WAN), a wired network, a wireless network, etc., or any combination thereof. In some embodiments, the network 120 can be any one or more of a wired network or a wireless network. In some embodiments, the network 120 can include one or more network access points. For example, the network 120 can include wired or wireless network access points, such as base stations and / or network switching points, through which one or more components of the application scenario 100 can connect to the network 120 to exchange data and / or information.

[0032] The pin shaft machining system 130 refers to a system that completes a pin shaft machining task based on the execution instruction of the management platform. In some embodiments, the pin shaft machining system 130 can include machines or devices that perform multiple machining processes. For example, cold heading machines 131, threading machines 132, etc. In some embodiments, there can be one or more machines or devices that perform the same machining process. For example, the pin shaft machining system 130 can include multiple cold heading machines 131.

[0033] The cold heading machine 131 refers to a device that performs cold heading process treatment on a workpiece to be machined and a process semi-finished product. Cold heading process treatment refers to a machining method that utilizes plastic deformation of metal under the action of external force and uses a mold to redistribute and transfer the volume of the metal body, thereby forming the required part. In some embodiments, cold heading process treatment can include multiple cold heading sub-process treatments, and the workpiece to be machined can undergo multiple cold heading sub-process treatments to obtain a final semi-finished product.

[0034] The threading machine 132 refers to a device that performs threading process treatment on the outer surface of the final semi-finished product after cold heading process treatment. Threading process treatment refers to a machining method that uses extrusion to cause plastic deformation of the workpiece by fixing one die plate and moving the workpiece with the other movable die plate, thereby forming the required thread. In some embodiments, the semi-finished product can undergo threading machining treatment to obtain a finished product.

[0035] In some embodiments, the pin shaft machining system 130 can also include other auxiliary devices. For example, the pin shaft machining system 130 can also include an image acquisition device for acquiring an image of the workpiece to be machined; for another example, the pin shaft machining system 130 can also include a quality inspection device for performing semi-finished product inspection.

[0036] In some embodiments, the pin shaft machining system 130 can comprise a signal transmitter and a signal receiver configured to communicate with the management platform 110 through the network 120 to obtain execution instructions of pin shaft machining or upload relevant information (e.g., image information of the blank to be machined, etc.) of pin shaft machining.

[0037] For more details about the blank to be machined, the semi-finished product, the finished product, the image information of the blank to be machined, see Figure 3 and the related description thereof; for more details about the semi-finished product inspection, see Figure 6 and the related description thereof.

[0038] It should be noted that the application scenarios are provided only for illustrative purposes and are not intended to limit the scope of the present specification. Various modifications or changes can be made according to the description of the present specification for those of ordinary skill in the art. For example, the application scenarios can also include a database. For another example, the application scenarios can be implemented on other devices to achieve similar or different functions. However, the changes and modifications will not depart from the scope of the present specification.

[0039] Figure 2 is an exemplary block diagram of a pin shaft automated machining system according to some embodiments of the present specification. As Figure 2 indicated, in some embodiments, the pin shaft automated machining system 200 can comprise an acquisition module 210, a prediction module 220, a grabbing module 230, and a machining module 240.

[0040] The acquisition module 210 can be configured to acquire feature information of the blank to be machined, the feature information comprising two-dimensional feature information of the blank to be machined, the two-dimensional feature information being related to image information of the blank to be machined and / or image features of the image information. For more details about the blank to be machined, the feature information and the way of acquiring the same, the two-dimensional feature information and the way of acquiring the same, the image information and the way of acquiring the same, the image features and the way of acquiring the same, see Figure 3 and the related description thereof.

[0041] In some embodiments, the acquisition module 210 can also be configured to acquire sampling data of the blank to be machined; generate a three-dimensional model of the blank to be machined based on the sampling data; and determine three-dimensional feature information of the blank to be machined based on the three-dimensional model. For more details about the three-dimensional feature information and the way of determining the same, the sampling data and the way of acquiring the same, the three-dimensional model and the way of generating the same, see Figure 4 , Figure 5 and the related description thereof.

[0042] The prediction module 220 can be configured to predict an estimated quality of the blank to be machined after being machined into a pin shaft based on the feature information and pin shaft machining process parameters. For more details about the pin shaft machining process parameters, the estimated quality and the way of predicting the same, see Figure 3and related descriptions thereof.

[0043] The grabbing module 230 can be configured to, in response to the estimated quality satisfying the quality preset condition, transport the to-be-processed blank to the target station by the grabbing device. For more details of the quality preset condition and the target station, see Figure 3 and related descriptions thereof.

[0044] The processing module 240 can be configured to process the to-be-processed blank based on the pin shaft processing parameters, the pin shaft processing parameters including cold heading process parameters and threading process parameters, and the processing of the to-be-processed blank based on the pin shaft processing parameters including: performing cold heading process on the to-be-processed blank based on the cold heading process parameters to obtain a semi-finished product; performing semi-finished product inspection during the cold heading process; performing threading process on the semi-finished product based on the threading process parameters to obtain a finished product. For more details of the cold heading process parameters, the threading process parameters, the semi-finished product, and the finished product, see Figure 3 and related descriptions thereof.

[0045] In some embodiments, the processing module 240 can also be configured to perform semi-finished product inspection based on the semi-finished product obtained after each to-be-processed blank passes through each cold heading sub-process to obtain a semi-finished product inspection result; and in response to the semi-finished product inspection result satisfying the inspection value preset condition and the continuity preset condition, stop the pin shaft automatic processing. For more details of the cold heading sub-process, the semi-finished product inspection result, the manner of performing semi-finished product inspection, the inspection value preset condition and the manner of determining the same, and the continuity preset condition and the manner of determining the same, see Figure 6 , Figure 7 and related descriptions thereof.

[0046] In some embodiments, the pin shaft automatic processing system 200 can further include an inspection module 250 and an optimization module 260.

[0047] The inspection module 250 can be configured to perform finished product inspection based on the finished product to obtain a finished product inspection result, the finished product inspection result including an abnormality rate and an abnormality type distribution of the finished product. For more details of the manner of performing finished product inspection, the finished product inspection result, the abnormality rate, and the abnormality type distribution, see Figure 8 and related descriptions thereof.

[0048] The optimization module 260 can be configured to determine pin shaft processing optimization parameters based on the finished product inspection result. For more details of the pin shaft processing optimization parameters and the manner of determining the same, see Figure 8 and related descriptions thereof.

[0049] It should be noted that the above description of the system and its modules is for the convenience of description only, and cannot limit the scope of the embodiments. It can be understood that, for those skilled in the art, after understanding the principle of the system, any combination of the modules or connection of the modules to other modules can be made without departing from the principle. For example, the grabbing module 230 and the processing module 240 can be integrated into one module. For another example, the modules can share one storage device, and the modules can also have their own storage devices. Such variations are within the scope of protection of the present specification.

[0050] Figure 3 is an exemplary flowchart of a pin shaft automated processing method according to some embodiments of the present specification. In some embodiments, the flow 300 can be executed by the management platform 110. As shown in Figure 3 the flow 300 includes the following steps:

[0051] Step 310: Obtain feature information of a to-be-processed blank part, the feature information including two-dimensional feature information of the to-be-processed blank part, the two-dimensional feature information being related to image information of the to-be-processed blank part and / or image features of the image information. In some embodiments, the step 310 can be executed by the obtaining module 210.

[0052] The to-be-processed blank part refers to a to-be-processed pin shaft forging blank part that has not been processed by a processing procedure. For example, Figure 7 the to-be-processed blank part shown in

[0053] The feature information refers to information reflecting the features of the to-be-processed blank part. For example, size information of the to-be-processed blank part, material information of the to-be-processed blank part, surface crack information of the to-be-processed blank part, etc.

[0054] In some embodiments, the feature information includes two-dimensional feature information of the to-be-processed blank part.

[0055] The two-dimensional feature information refers to feature information contained in a two-dimensional image of the to-be-processed blank part. For example, color feature information of the to-be-processed blank part, shape feature information of the to-be-processed blank part, edge feature information of the to-be-processed blank part, etc.

[0056] In some embodiments, the two-dimensional feature information is related to image information of the to-be-processed blank part and / or image features of the image information.

[0057] The image information refers to information contained in an image of the to-be-processed blank part. For example, color information of the to-be-processed blank part, shape information of the to-be-processed blank part, etc.

[0058] The image feature refers to a feature extracted based on image information of the to-be-processed blank. For example, a feature of a surface crack in the image of the to-be-processed blank.

[0059] In some embodiments, the acquisition module 210 can acquire the feature information of the to-be-processed blank through various feasible information acquisition methods.

[0060] In some embodiments, the acquisition module 210 can acquire an image of the to-be-processed blank through an image acquisition device (such as a camera, an infrared imager, etc.), extract image information and image features in the image of the to-be-processed blank through an image analysis method, and obtain two-dimensional feature information of the to-be-processed blank based on the image information and the image features.

[0061] In some embodiments, the feature information further includes three-dimensional feature information of the to-be-processed blank. In some embodiments, the acquisition module 210 can acquire sampling data of the to-be-processed blank, generate a three-dimensional model of the to-be-processed blank based on the sampling data, and determine three-dimensional feature information of the to-be-processed blank based on the three-dimensional model. For more information about the sampling data and its acquisition method, the three-dimensional model and its generation method, and the three-dimensional feature information and its determination method, see Figure 4 and related descriptions.

[0062] Step 320: predicting an estimated quality of the to-be-processed blank processed into a pin shaft based on the feature information and the pin shaft processing parameters. In some embodiments, step 320 can be performed by the prediction module 220.

[0063] The pin shaft processing parameter refers to a series of basic data or indexes of a process for completing the pin shaft processing. The pin shaft processing parameter can include a cold heading number (such as a number of cold heading sub-processes, etc.), die design data of each cold heading sub-process, threading process parameters (such as a rotation speed, etc.), and other external parameters (such as temperature, pressure, time, stirring speed, line moving speed, extrusion speed, etc. during the pin shaft processing).

[0064] The estimated quality refers to a quality of the to-be-processed blank processed into the pin shaft predicted in advance. The estimated quality can be represented by a score, for example, the estimated quality can be represented by a score of 1-10, and the higher the score, the higher the estimated quality.

[0065] In some embodiments, the prediction module 220 can process the feature information of the to-be-processed blank and the pin shaft processing parameters through modeling or various feasible data information processing methods, and predict the estimated quality of the to-be-processed blank processed into the pin shaft.

[0066] In some embodiments, the prediction module 220 can process the feature information of the to-be-processed blank and the pin shaft processing parameters through a quality prediction model to predict the estimated quality of the to-be-processed blank processed into a pin shaft.

[0067] The quality prediction model refers to a model for predicting the estimated quality of the to-be-processed blank processed into a pin shaft. In some embodiments, the quality prediction model can be various feasible neural network models. For example, a convolutional neural network (CNN), a deep neural network (DNN), or the like, or a combination thereof.

[0068] The input of the quality prediction model can include the feature information of the to-be-processed blank and the pin shaft processing parameters. For example, the input feature information of the to-be-processed blank can be the size information of the to-be-processed blank, the material information of the to-be-processed blank, and the like; for another example, the input pin shaft processing parameters can be the temperature in the pin shaft processing process, the stirring speed, and the like.

[0069] The output of the quality prediction model can include the estimated quality of the to-be-processed blank processed into a pin shaft. For example, the output estimated quality of the to-be-processed blank processed into a pin shaft can be a score corresponding to the estimated quality, such as 5 points.

[0070] In some embodiments, the quality prediction model can be trained based on historical data. In some embodiments, the quality prediction model can be trained based on a plurality of training samples and labels.

[0071] In some embodiments, the first training sample for training the quality prediction model includes the feature information of the sample to-be-processed blank and the sample pin shaft processing parameters corresponding to the sample to-be-processed blank, and the first label corresponding to the first training sample is the actual quality of the sample to-be-processed blank processed into a pin shaft. The first training sample can be obtained based on historical data, and the first label can be determined by manual annotation or automatic annotation. The above description is only an example and is not limited, and the label of the training data can be obtained by various ways.

[0072] During training, the first training sample is input into the quality prediction model, a loss function is constructed based on the output of the initial quality prediction model and the first label, the parameters of the initial quality prediction model are updated through the loss function, until the trained initial quality prediction model meets a preset condition, and a trained quality prediction model is obtained, wherein the preset condition can be that the loss function is less than a threshold, converges, or the training period reaches a threshold.

[0073] At step 330, in response to the estimated quality satisfying the quality preset condition, the to-be-processed blank is transported to a target work station by the grabbing device. In some embodiments, step 330 can be performed by the grabbing module 230.

[0074] The quality preset condition refers to a condition preset in advance for judging whether the to-be-processed blank can be processed based on the estimated quality. For example, the quality preset condition can be that the score corresponding to the estimated quality is greater than a threshold value (e.g., 8 points).

[0075] The target work station refers to a position where the to-be-processed blank is placed before the pin shaft processing. For example, a certain position on the conveying belt.

[0076] At step 340, the to-be-processed blank is processed based on the pin shaft processing parameters, which include cold heading process parameters and threading process parameters. In some embodiments, step 340 can be performed by the processing module 240.

[0077] The cold heading process parameters refer to a series of basic data or indicators of the process for completing the cold heading processing. For example, the cold heading process parameters can include the number of upsetting times, the pipeline moving speed, the extrusion speed, the lifting speed, etc.

[0078] The threading process parameters refer to a series of basic data or indicators of the process for completing the threading processing. For example, the threading process parameters can include the rotating speed of the threading machine, etc.

[0079] In some embodiments, the processing module 240 can process the to-be-processed blank based on the pin shaft processing parameters based on the execution instructions of the management platform 110. In some embodiments, the processing procedure of the to-be-processed blank by the processing module 240 can include a cold heading procedure and a threading procedure.

[0080] In some embodiments, the processing of the to-be-processed blank by the processing module 240 based on the pin shaft processing parameters can include sub-steps 341-342.

[0081] At sub-step 341, the to-be-processed blank is subjected to cold heading processing based on the cold heading process parameters to obtain a semi-finished product; and a semi-finished product inspection is performed during the cold heading processing.

[0082] For more information about the cold heading processing, see Figure 1 and the related description.

[0083] The semi-finished product refers to a processed product obtained after the to-be-processed blank is subjected to the cold heading processing. The semi-finished product includes a process semi-finished product and a final semi-finished product, the process semi-finished product refers to a processed product obtained after the to-be-processed blank is subjected to each cold heading sub-procedure, and the final semi-finished product refers to a processed product obtained after the to-be-processed blank is subjected to all cold heading sub-procedures.

[0084] As shown in Figure 7 the pin shaft machining process can include a cold heading process and a threading process, wherein the cold heading process includes three cold heading sub-processes. In the pin shaft automatic machining process, the to-be-machined blank is processed by the cold heading sub-process 1 to obtain a semi-finished product 1; the semi-finished product 1 is processed by the cold heading sub-process 2 to obtain a semi-finished product 2; the semi-finished product 2 is processed by the cold heading sub-process 3 to obtain a semi-finished product 3; and the semi-finished product 3 is processed by the threading process to obtain a finished product. Among them, the semi-finished product 1 and the semi-finished product 2 are process semi-finished products, and the semi-finished product 3 is a final semi-finished product. Each cold heading sub-process corresponds to a deformation degree, for example, the cold heading sub-process 1 corresponds to a deformation degree 1, and the cold heading sub-process 2 corresponds to a deformation degree 2. For more information about the deformation degree, see Figure 6 and related descriptions.

[0085] Semi-finished product inspection refers to inspecting the machining effect of the semi-finished product. For example, the shape, weight, texture, and other elements of the semi-finished product are inspected.

[0086] In some embodiments, the machining module 240 can inspect the semi-finished product based on various machining product inspection methods (such as sampling inspection, etc.).

[0087] In some embodiments, the machining module 240 can sample the final semi-finished product, compare the sample semi-finished product with the historical qualified semi-finished product in terms of various elements, and achieve semi-finished product inspection. For example, the image information of the sample semi-finished product is obtained by the image acquisition device, the historical image information of the historical qualified semi-finished product is obtained from the storage device 113, and the image information of the sample semi-finished product and the historical image information of the historical qualified semi-finished product are processed by the image analysis method to achieve semi-finished product inspection; for another example, the weight of the sample semi-finished product is obtained by the weight sensor, the historical weight of the historical qualified semi-finished product is obtained from the storage device 113, and the semi-finished product inspection is achieved by directly comparing the weight of the sample semi-finished product with the historical weight of the historical qualified semi-finished product.

[0088] In some embodiments, the machining module 240 can perform semi-finished product inspection based on the semi-finished product obtained after each to-be-machined blank is processed by each cold heading sub-process, and obtain a semi-finished product inspection result. For more information about the semi-finished product inspection result and the semi-finished product obtained after each cold heading sub-process, see Figure 6 and related descriptions.

[0089] In sub-step 342, the semi-finished product is processed by the threading process based on the threading process parameters, and a finished product is obtained.

[0090] For more information about the threading process, see Figure 1 and related descriptions.

[0091] The finished product refers to a processed product obtained by processing a semi-finished product through a threading process.

[0092] In some embodiments, the processing module 240 can perform finished product inspection based on the finished product. For more information about finished product inspection, see Figure 8 and the related description.

[0093] In some embodiments of the present specification, by predicting the estimated quality of the to-be-processed blank after being processed into a pin shaft based on the feature information of the to-be-processed blank and the pin shaft processing process parameters, and only transporting the to-be-processed blank to the target station for subsequent processing if the estimated quality is qualified, a batch of blanks that are not suitable for processing can be effectively eliminated, resource waste can be avoided, and the quality and processing efficiency of the finished product can be improved. By performing semi-finished product inspection, problems in the pin shaft processing process can be found in time so that adjustments can be made in time to ensure the orderly progress of the pin shaft automated processing. The entire pin shaft processing process is controlled and executed by the management platform, realizing the automation and intelligentization of pin shaft processing, reducing labor costs, and improving production efficiency.

[0094] Figure 4 is an exemplary flowchart for determining three-dimensional feature information according to some embodiments of the present specification. In some embodiments, the flow 400 can be executed by the management platform 110 or the acquisition module 210. As Figure 4 shown, the flow 400 includes the following steps:

[0095] Step 410, acquiring sampling data of the to-be-processed blank.

[0096] The sampling data refers to data related to the actual situation of the to-be-processed blank obtained by sampling the outer surface of the to-be-processed blank. For example, the coordinates of a point on the outer surface of the to-be-processed blank, the length of a side on the outer surface of the to-be-processed blank, etc.

[0097] In some embodiments, the acquisition module 210 can collect the sampling data of the to-be-processed blank through various data sampling methods and store it in the storage device 113.

[0098] In some embodiments, the acquisition module 210 can collect point cloud data of the outer surface of the to-be-processed blank through devices such as depth cameras and laser scanners, thereby acquiring the sampling data of the to-be-processed blank. The point cloud data refers to a set of vectors representing the points on the outer surface of the to-be-processed blank in a three-dimensional coordinate system. Each point in the point cloud data of the to-be-processed blank can include the three-dimensional coordinates, color information, and reflection intensity information of the point.

[0099] Step 420, generating a three-dimensional model of the to-be-processed blank based on the sampling data.

[0100] The three-dimensional model refers to a model obtained by three-dimensional modeling of the blank to be processed, and reflecting the three-dimensional reality of the blank to be processed.

[0101] In some embodiments, the acquisition module 210 can generate the three-dimensional model of the blank to be processed based on the point cloud data of the blank to be processed through a point cloud modeling software (such as Context Capture, etc.).

[0102] In some embodiments, the sampling data includes real images of the blank to be processed taken from a plurality of preset angles. In some embodiments, the acquisition module 210 can determine generated images of the plurality of preset angles of the blank to be processed based on the real images of the blank to be processed taken from the plurality of preset angles; and generate the three-dimensional model of the blank to be processed based on the real images and the generated images.

[0103] The preset angle refers to a preset shooting angle, which includes a shooting height, a shooting direction, and a shooting distance. For example, the preset angle 1 can be that the shooting height is 10 cm, the shooting direction is 30° westward, and the shooting distance is 30 cm, with the bottom center of the blank to be processed as the reference; for another example, the preset angle 2 can be that the shooting height is 12 cm, the shooting direction is 25° westward, and the shooting distance is 33 cm, with the geometric center of the blank to be processed as the reference.

[0104] The real image refers to an image obtained by directly shooting the blank to be processed through an image acquisition device. For example, the image obtained by directly shooting from the preset angle 1 through the image acquisition device.

[0105] The generated image refers to an image obtained by processing a real image through an image generation technique. For example, the image obtained by processing the real image of the preset angle 1.

[0106] In some embodiments, the acquisition module 210 can process the real image of the blank to be processed through various feasible image generation techniques (such as an autoregressive model (Autoregressive model), a variational autoencoder (VAE), etc.) to obtain the generated image.

[0107] In some embodiments, the acquisition module 210 can process the real image of the blank to be processed through the generation layer of the three-dimensional model generation model to obtain the generated image. For more information about the generation layer of the three-dimensional model generation model, see Figure 5 and the related description.

[0108] In some embodiments, the acquisition module 210 can process the real image and the generated image using various 3D model generation methods (such as PixelNeRF, Pix2NeRF, GRF, etc.) to generate a 3D model of the blank to be processed.

[0109] In some embodiments, the acquisition module 210 can process the real image and the generated image using a 3D model generation method to generate a 3D model of the workpiece to be processed. For more information on 3D model generation, see [link to relevant documentation]. Figure 5 And its related descriptions.

[0110] In some embodiments of this specification, a generated image is obtained from a real image of the workpiece to be processed, and then a three-dimensional model is generated based on the real image and the generated image. This can improve the efficiency of generating the three-dimensional model and make the generated three-dimensional model more consistent with the real situation of the workpiece to be processed.

[0111] Step 430: Based on the 3D model, determine the 3D feature information of the blank to be processed.

[0112] Three-dimensional feature information refers to the feature information contained in the three-dimensional model of the workpiece to be processed. For example, the volume and surface area of ​​the three-dimensional model of the workpiece to be processed.

[0113] In some embodiments, the three-dimensional feature information includes morphological sampling features.

[0114] Morphological sampling features refer to the features contained in the data information obtained by sampling a 3D model. For example, the features contained in any point in the 3D model (such as the curvature of that point), or the features contained in any three points in the 3D model (such as the area of ​​the triangle formed by the three points).

[0115] In some embodiments, the morphological sampling features include one or more feature vectors. In some embodiments, the morphological sampling features include at least a distance feature vector.

[0116] A feature vector is a vector used to reflect the morphological sampling characteristics. Examples include single-distance feature vectors, area feature vectors, volume feature vectors, curvature feature vectors, and distance feature vectors.

[0117] Distance feature vectors are vectors used to reflect distance-related features in morphological sampling features.

[0118] In some embodiments, the acquisition module 210 can sample the 3D model using different sampling methods to obtain one or more feature vectors.

[0119] In some embodiments, the obtaining module 210 can determine the distance feature vector based on the multiple samplings. The way of determining the distance feature vector can be achieved by the following sub-step S1-sub-step S3:

[0120] Sub-step S1, for each sampling, obtain any two sampling points on the surface of the three-dimensional model.

[0121] A sampling point refers to any point on the three-dimensional model of the blank to be machined, and the sampling point can be represented by three-dimensional coordinates. For example, sampling point 1 can be (0, 1, 0), and sampling point 2 can be (0, 0, 1).

[0122] Sub-step S2, calculate the distance between the two sampling points, and take the distance as a feature value.

[0123] A feature value refers to the value of a specific factor between different sampling points when sampling in different sampling ways. For example, for a single-distance feature vector, the sampling way is to collect any one sampling point on the surface of the three-dimensional model for each sampling, and the single-distance feature value can be the distance from the sampling point to the center of the three-dimensional model. For example, for a distance feature vector, the sampling way is to collect any two sampling points on the surface of the three-dimensional model for each sampling, and the distance feature value can be the distance between the two sampling points. For example, for an area feature vector, the sampling way is to collect any three points on the surface of the three-dimensional model for each sampling, and the area feature value can be the area of the triangle formed by the three points.

[0124] In some embodiments, the obtaining module 210 can calculate the distance between the two sampling points by a coordinate calculation method. For example, the distance between sampling point 1 (0, 1, 0) and sampling point 2 (0, 0, 1) is √2.

[0125] In some embodiments, the obtaining module 210 can obtain other types of feature values by a coordinate calculation method and a geometric analysis method. For example, the obtaining module 210 can calculate the area of the triangle formed by any three sampling points by a coordinate calculation method and a geometric analysis method, and take it as an area feature value. For example, the obtaining module 210 can calculate the volume of the tetrahedron formed by any four sampling points by a coordinate calculation method and a geometric analysis method, and take it as a volume feature value.

[0126] Sub-step S3, the sub-step S1-sub-step S2 is executed enough times (such as, million, million times), a plurality of characteristic values of a plurality of categories are obtained, and the number of occurrences of each characteristic value under the corresponding category is counted, that is, the characteristic vector corresponding to the characteristic value of the category is obtained, such as counting the number of occurrences of each distance characteristic value in the plurality of distance characteristic values, the distance characteristic vector can be obtained. Each element in the distance characteristic vector corresponds to a distance characteristic value, for example, the distance characteristic value corresponding to the first element is 1, the distance characteristic value corresponding to the second element is 2, and so on. Then the distance characteristic vector can be (20, 30, 40, …), indicating that the distance characteristic value 1 occurs 200,000 times, the distance characteristic value 2 occurs 300,000 times, the distance characteristic value 3 occurs 400,000 times, and so on.

[0127] In some embodiments, the acquisition module 210 can determine other characteristic vectors (such as single-distance characteristic vectors, area characteristic vectors, volume characteristic vectors, etc.) based on multiple samplings. The method of determining other characteristic vectors is similar to the method of determining distance characteristic vectors, which will not be described here.

[0128] In some embodiments, the acquisition module 210 can obtain the distance between any two sampling points and the feature difference between the two sampling points for each sampling. The number of occurrences of each distance characteristic value under different feature differences is counted to obtain the distance characteristic vector.

[0129] The feature difference refers to the difference in other features between any two sampling points in addition to the distance. For example, the curvature difference.

[0130] The curvature difference refers to the difference in curvature between any two sampling points. For example, the curvature of sampling point 1 is 0.2, and the curvature of sampling point 2 is 0.5, so the curvature difference between the two sampling points is 0.3.

[0131] Curvature can be used to describe the curvature of a curve or surface. The curvature of a point on a curve can be the rate of rotation of the tangent direction angle of the point with respect to the arc length. The curvature of a point on a three-dimensional model can be the maximum, minimum, average or any other feasible representation of the curvature of the point on the curve passing through the point.

[0132] In some embodiments, the acquisition module 210 can directly obtain the curvature of the sampling point based on the three-dimensional model. In some embodiments, the acquisition module 210 can calculate the curvature of the sampling point based on the three-dimensional coordinates of the sampling point through a curvature calculation method.

[0133] In some embodiments, the acquisition module 210 can update the distance feature vector in combination with the curvature difference, and count the number of occurrences of each distance feature value under different curvature differences to obtain a two-dimensional distance feature vector, i.e., a distance matrix. Each row of the distance matrix corresponds to a curvature difference, and each column corresponds to the number of occurrences of the same distance feature value under different curvature differences. For example, the distance matrix can be The curvature difference corresponding to the first row is 0, the curvature difference corresponding to the second row is 0.1, the curvature difference corresponding to the third row is 0.2, and so on; the distance feature value corresponding to the first column is 1, the distance feature value corresponding to the second column is 2, the distance feature value corresponding to the third column is 3, and so on; and the representation of the distance matrix is: when the curvature difference is 0, the distance feature value 1 occurs 230,000 times, the distance feature value 2 occurs 220,000 times, and the distance feature value 3 occurs 380,000 times; when the curvature difference is 0.1, the distance feature value 1 occurs 330,000 times, the distance feature value 2 occurs 450,000 times, and the distance feature value 3 occurs 470,000 times; and when the curvature difference is 0.2, the distance feature value 1 occurs 320,000 times, the distance feature value 2 occurs 550,000 times, and the distance feature value 3 occurs 600,000 times.

[0134] In some embodiments of the present specification, when determining the distance feature vector, the feature difference between the sampling points, such as the curvature difference, is introduced, so that the two-dimensional vector is optimized to a three-dimensional or multi-dimensional vector, the dimension of the distance feature vector is enriched, and the richness of the three-dimensional feature information is increased, which helps to improve the accuracy of the prediction result when predicting the estimated quality of the processed blank processed into a pin shaft based on the feature information and the pin shaft processing process parameters.

[0135] In some embodiments of the present specification, one or more feature vectors can be obtained by different sampling methods, which can make the collected three-dimensional feature information more rich and comprehensive, and help to improve the accuracy of the prediction result when predicting the estimated quality of the processed blank processed into a pin shaft based on the feature information and the pin shaft processing process parameters.

[0136] In some embodiments, the acquisition module 210 can determine the three-dimensional feature information of the processed blank by various information acquisition methods. For example, based on the three-dimensional model, the related information (such as the length of any edge of the three-dimensional model) is derived.

[0137] In some embodiments, the acquisition module 210 can obtain one or more feature vectors by the above-mentioned different sampling methods, determine the morphological sampling feature based on the one or more feature vectors, and then obtain the three-dimensional feature information based on the morphological sampling feature.

[0138] In some embodiments of the present specification, by acquiring the sampling data of the blank to be processed, generating a three-dimensional model based on the sampling data, and then acquiring three-dimensional feature information based on the three-dimensional model, a large amount of three-dimensional feature information can be acquired in a short time, making it more convenient to acquire three-dimensional feature information, and ensuring that the acquired three-dimensional feature information conforms to the actual situation of the blank to be processed.

[0139] Figure 5 is an exemplary schematic diagram of a three-dimensional model generation model according to some embodiments of the present specification.

[0140] In some embodiments, the acquisition module 210 can process the real images of each preset angle of the blank to be processed through the three-dimensional model generation model to generate a three-dimensional model of the blank to be processed.

[0141] The three-dimensional model generation model refers to a model for generating a three-dimensional model of the blank to be processed. As shown in Figure 5 , the three-dimensional model generation model includes a generation layer 520 and a reconstruction layer 540. The three-dimensional model generation model can process the real images 510 of each preset angle of the blank to be processed to generate a three-dimensional model 550 of the blank to be processed.

[0142] In some embodiments, the generation layer 520 can process the real images 510 of each preset angle of the blank to be processed to acquire generated images 530 of each preset angle of the blank to be processed. In some embodiments, the generation layer 520 can be various feasible neural network models. For example, Convolutional Neural Networks (CNN), Deep Neural Networks (DNN), Generative Adversarial Networks (GAN), Neural Radiance Fields (NeRF), PixelNeRF, etc. or combinations thereof.

[0143] As shown in Figure 5 , the input of the generation layer 520 can include the real images 510 of each preset angle of the blank to be processed. For example, the real image 511 of the preset angle 1, the real image 512 of the preset angle 2, etc.

[0144] The output of the generation layer 520 can include the generated images 530 of each preset angle of the blank to be processed. For example, the generated image 531 of the preset angle 1, the generated image 532 of the preset angle 2, etc.

[0145] In some embodiments, the generation layer 520 can be obtained through adversarial training. Adversarial training refers to adding generated adversarial samples to the training set to achieve data augmentation, so that the model learns the adversarial samples in advance during training, thereby improving the processing capability of the model.

[0146] In some embodiments, the generation layer 520 is composed of a generator and a discriminator. The generator can be used to generate fake images, and the fake images generated by the generator are input into the discriminator together with real images. The discriminator can be used to determine whether the images input into the discriminator are real images.

[0147] In some embodiments, the generation layer can be trained based on the second and third training samples to obtain the trained generation layer 520. The training of the generation layer 520 can include multiple stages:

[0148] In the first stage, the parameters of the generator are fixed, and the discriminator is trained. Based on the historical data, sample real images are obtained, and the sample real images are input into the generator to generate fake images. The fake images and the sample real images are used as the second training samples in the first stage. The second label corresponding to the second training sample is whether the image input into the discriminator is a real image. If it is a real image, the second label is 1, otherwise, the second label is 0. The discriminator is trained based on the second training sample and the second label, so that the discriminator can distinguish whether the input image is a real image as much as possible.

[0149] In the second stage, the parameters of the discriminator are fixed, and the generator is trained. The generator and the discriminator obtained in the first stage form a composite model. In the composite model, the fake images output by the generator can be input into the discriminator for judgment. The sample real images are input into the generator to generate fake images, and the fake images are used as the third training samples in the second stage and input into the composite model for training. The third label corresponding to the third training sample is that the image input into the composite model is a real image, i.e., the third label of the composite model is 1. In the second stage training, the parameters of the discriminator in the composite model are fixed, and the parameters of the generator in the composite model are updated. With the continuous training in the second stage, the data processing capability of the generator becomes stronger and stronger, and the similarity between the fake images output by the generator and the sample real images is continuously improved, until the discriminator judges the fake images as real images.

[0150] The first stage and the second stage are cycled, and finally through continuous cycling, the capabilities of the generator and the discriminator become stronger and stronger, and finally the model converges to obtain the trained generation layer 520.

[0151] In some embodiments, the reconstruction layer 540 can process the generated images 530 of each preset angle of the workpiece to be processed to obtain a three-dimensional model 550 of the workpiece to be processed. In some embodiments, the reconstruction layer 540 can be various feasible neural network models. For example, a convolutional neural network (CNN), a deep neural network (DNN), a generative adversarial network (GAN), a neural radiance field (NeRF), a PixelNeRF, or a combination thereof.

[0152] As shown in FIG. 5, the input of the reconstruction layer 540 can include the generated images 530 of each preset angle of the workpiece to be processed. For example, the generated image 531 of the preset angle 1, the generated image 532 of the preset angle 2, and the like. Figure 5

[0153] The output of the reconstruction layer 540 can include the three-dimensional model 550 of the workpiece to be processed.

[0154] In some embodiments, the reconstruction layer 540 can be obtained through adversarial training. For details of the specific process of obtaining the reconstruction layer through adversarial training, please refer to the above description of obtaining the generation layer 520 through adversarial training. The difference between the two is that the input of the generator of the generation layer 520 is a real image, and the output is a fake image. The fake image generated by the generator is input into the discriminator of the generation layer 520 together with the real image. The discriminator can be used to determine whether the input image is a real image. The input of the generator of the reconstruction layer 540 is the image output by the generation layer 520, and the output of the generator is a three-dimensional model. The three-dimensional model generated by the generator is input into the discriminator of the reconstruction layer 540 together with a historical three-dimensional model similar to the workpiece to be processed by more than a threshold (for example, 99%). The discriminator can be used to determine whether the input model is a historical three-dimensional model. For other details of adversarial training, please refer to the above description of obtaining the generation layer 520 through adversarial training. Here, it will not be repeated.

[0155] ​In some embodiments of the present specification, the three-dimensional model of the blank to be processed is generated by a three-dimensional model generation model, which can improve the efficiency of generating the three-dimensional model; and the generation layer and reconstruction of the three-dimensional model generation model are trained in the manner of a generative adversarial network, a high-difficulty sample is generated to train the discriminator, and through the self-game of the generator of the generation layer and the discriminator in the training process, the accuracy of the discriminator in identifying confusing samples can be greatly improved, and the ability of the generator and the discriminator to process data can be enhanced, thereby improving the quality of the false image or false model output by the generator, and finally making the three-dimensional model generated by the three-dimensional model generation model more similar to the actual blank to be processed.

[0156] Figure 6 is an example flowchart of performing semi-product inspection according to some embodiments of the present specification. In some embodiments, the flow 600 can be performed by the management platform 110 or the processing module 240. As shown in Figure 6 , the flow 600 includes the following steps:

[0157] Step 610, based on the semi-product obtained after each blank to be processed is processed by each cold heading sub-process, performing semi-product inspection to obtain a semi-product inspection result.

[0158] In some embodiments, the cold heading process can include at least one cold heading sub-process. As shown in Figure 7 , the pin shaft processing process can include a cold heading process and a thread rolling process, wherein the cold heading process includes three cold heading sub-processes. In the automatic processing process of the pin shaft, the blank to be processed is processed by the cold heading sub-process 1 to obtain the semi-product 1; the semi-product 1 is processed by the cold heading sub-process 2 to obtain the semi-product 2; the semi-product 2 is processed by the cold heading sub-process 3 to obtain the semi-product 3; and the semi-product 3 is processed by the thread rolling process to obtain the finished product. Among them, the semi-product 1 and the semi-product 2 are process semi-products, and the semi-product 3 is a final semi-product. For more information about the cold heading process, see Figure 1 and related descriptions.

[0159] For more information about the semi-product, the process semi-product, and the final semi-product, see Figure 3 and related descriptions.

[0160] The semi-product inspection result refers to the result reflecting the processing of the semi-product after the semi-product inspection is performed on the semi-product. For more information about the semi-product inspection, see Figure 3 and related descriptions. The semi-product inspection result can be represented by a specific percentage, for example, the semi-product inspection result can be 80%.

[0161] During the processing of the cold heading sub-process, each semi-finished product obtained after processing of a cold heading sub-process can correspond to a semi-finished product inspection result. As shown in FIG. 1, the semi-finished product 1 can correspond to a semi-finished product inspection result, and the semi-finished product 2 can also correspond to a semi-finished product inspection result. Figure 7

[0162] In some embodiments, the semi-finished product inspection result can be determined based on matching the inspection vector of the semi-finished product to be inspected with the historical vectors in the vector database, obtaining similar vectors, and determining the semi-finished product inspection result based on the situation of the similar vectors.

[0163] The inspection vector refers to a vector constructed based on the sensing information obtained during and after the processing of the semi-finished product in the cold heading sub-process. As shown in FIG. 2, for the process of obtaining the semi-finished product 1 by processing the blank 1 in the cold heading sub-process 1, the semi-finished product 1 corresponds to an inspection vector, which is constructed based on the sensing information obtained during the processing of the cold heading sub-process 1 and the sensing information obtained after the processing of the cold heading sub-process 1. In some embodiments, the inspection vector can be obtained by directly encoding the sensing information, or by embedding the sensing information. Figure 7

[0164] The sensing information obtained during the processing of the cold heading sub-process 1 can include pressure sensing information, which refers to the pressure generated during the processing of the cold heading sub-process. In some embodiments, the pressure sensing information can be collected by a pressure sensor. For example, the pressure sensor can be deployed at a certain position of the cold heading machine mold, so that the sensor is in good contact with the force receiving member, thereby allowing real-time collection of the pressure sequence during the processing of the cold heading sub-process.

[0165] The sensing information obtained after the processing of the cold heading sub-process 1 can include image sensing information, which refers to the information contained in the image of the outer surface of the semi-finished product. In some embodiments, the image sensing information can be collected by an image acquisition device. For example, a camera can be deployed at a position where the semi-finished product is output from the cold heading machine, to acquire images of the semi-finished product in real time.

[0166] In some embodiments, the vector database of the management platform 110 stores a large number of historical vectors. The historical vectors in the vector database include standard vectors corresponding to standard semi-finished products and defect vectors corresponding to defective semi-finished products. The standard vector refers to a vector constructed based on the sensing information collected under a standard and normal cold heading sub-process. The defect vector refers to a vector constructed based on the sensing information collected under a condition where there is a fault or problem in the processing of the cold heading sub-process (e.g., the blank to be processed has a defect, a certain component of the cold heading machine has a fault, etc.). The historical vectors in the vector database and their construction methods are the same as the inspection vector, and will not be described here.

[0167] ​​The similar vector refers to a vector in the vector database that meets a preset condition. The preset condition can be a condition set in advance by a person or a system. For example, the preset condition can be that the distance (e.g., Euclidean distance, etc.) between the test vector and the similar vector meets a certain threshold; for another example, the preset condition can be that the distances between the test vector and the similar vectors are arranged in ascending order, and the top 50 vectors are selected.

[0168] In some embodiments, the processing module 240 can determine the semi-finished product test result based on the case of the similar vector. For example, based on the obtained similar vector, the proportion of the standard vector in the similar vector is calculated, and the proportion is taken as the semi-finished product test result. For example, a total of 1000 similar vectors are obtained, of which 800 are standard vectors, and the semi-finished product test result is 80%.

[0169] In step 620, in response to the semi-finished product test result not meeting the test value preset condition and the continuity preset condition, the pin shaft automatic processing is stopped.

[0170] The test value preset condition refers to a condition set in advance for judging whether the semi-finished product test result meets the requirement in terms of specific numerical value. For example, the test value preset condition can be that the semi-finished product test result is greater than a test threshold (e.g., 80%). When performing semi-finished product testing on the semi-finished product obtained by each cold heading sub-process, a test value preset condition corresponding to the semi-finished product test result is determined.

[0171] In some embodiments, each cold heading sub-process corresponds to a deformation degree, and the test value preset condition is related to the deformation degree. The greater the deformation degree, the greater the random factors that can occur, and the greater the tolerance of the semi-finished product test result to the test value preset condition. For example, the greater the deformation degree, the smaller the test threshold in the test value preset condition that the semi-finished product test result is greater than the test threshold.

[0172] The deformation degree refers to a degree of deformation of the blank to be processed after the blank is processed by the cold heading sub-process. For example, the deformation degree can be the ratio of the compression amount of the length of the cold heading part of the blank to be processed or the process semi-finished product to the original height after the cold heading sub-process; for another example, the deformation degree can be the ratio of the increase amount of the cross-sectional area of the cold heading part of the blank to be processed or the process semi-finished product to the original cross-sectional area. Each cold heading sub-process corresponds to a deformation degree, such as deformation degree 1, deformation degree 2, etc. Figure 7

[0173] In some embodiments, the processing module 240 can analyze the workpieces before and after the cold heading sub-process by image analysis technology to determine the deformation degree. For example, the processing module 240 can obtain the length of the cold heading part of the workpiece before and after the cold heading sub-process by image recognition technology, calculate the ratio of the lengths before and after, and determine the deformation degree. ​

[0174] In some embodiments of the present disclosure, the preset condition of the inspection value is related to the deformation degree. The preset condition of the inspection value can be adjusted according to the deformation degree to make the setting of the preset condition of the inspection value more reasonable, considering the random factors in the cold heading process.

[0175] The preset condition of the continuity is a condition set in advance for judging whether the inspection results of a plurality of continuous semi-finished products meet the requirements. For example, the preset condition of the continuity can be that the number of semi-finished products whose inspection results meet the preset condition of the inspection value is less than a continuity threshold (e.g., 3). When performing the semi-finished product inspection on the semi-finished products obtained by each cold heading sub-process, a corresponding preset condition of the continuity is used.

[0176] In some embodiments, the preset condition of the continuity is related to the stability of the semi-finished product inspection results. The higher the stability of the semi-finished product inspection results, the smaller the tolerance of the preset condition of the continuity to the semi-finished product inspection results. For example, the higher the stability, the smaller the continuity threshold in the preset condition of the continuity that “the number of semi-finished products whose inspection results meet the preset condition of the inspection value is less than the continuity threshold”.

[0177] In some embodiments, the preset condition of the continuity corresponding to a subsequent cold heading sub-process is related to the stability of the semi-finished product inspection results of a previous cold heading sub-process. For example, as shown in FIG. 2, each cold heading sub-process has a preset condition of the continuity. The preset condition of the continuity of the cold heading sub-process 2 is related to the stability of the semi-finished product inspection results of the semi-finished products produced by the cold heading sub-process 1 in the last 10 minutes. Figure 7

[0178] The stability of the semi-finished product inspection results refers to the case that the semi-finished product inspection results of a plurality of semi-finished products processed by the same cold heading sub-process in a certain time period (e.g., 10 minutes) meet the preset condition of the inspection value. The stability of the semi-finished product inspection results can be expressed as a percentage, for example, 70%.

[0179] In some embodiments, a plurality of semi-finished products processed by the same cold heading sub-process in a certain time period can be obtained, and the proportion of the semi-finished products whose semi-finished product inspection results meet the preset condition of the inspection value can be taken as the stability of the semi-finished product inspection results. For example, in a certain time period (e.g., 10 minutes), 10 semi-finished products processed by the same cold heading sub-process are obtained, among which 7 semi-finished products have semi-finished product inspection results meeting the preset condition of the inspection value, and the stability of the semi-finished product inspection results is 70%.

[0180] ​In some embodiments of the present specification, the continuity preset condition corresponding to the next cold heading sub-process is determined based on the stability of the semi-finished product inspection result of the previous cold heading sub-process. The judgment standard of the next cold heading sub-process can be adaptively adjusted according to the actual situation of the previous cold heading sub-process, so that the setting of the continuity preset condition is more scientific and reasonable, and the problems in the cold heading sub-process processing process can be found in time, and the processing is stopped in time to reduce the loss.

[0181] In some embodiments, the continuity preset condition is related to the abnormal type distribution and the estimated quality of the plurality of continuous to-be-processed blank parts. The more concentrated the abnormal type distribution is, the higher the estimated quality of the plurality of continuous to-be-processed blank parts is. If the semi-finished product inspection result of the plurality of semi-finished parts is low, the possibility of problems existing in the pin shaft machining process, the mold, etc. is greater, and a relatively uniform abnormality is more likely to occur. Therefore, the greater the continuity preset condition is to the tolerance of the semi-finished product inspection result. For example, the more concentrated the abnormal type distribution is, the higher the estimated quality of the plurality of continuous to-be-processed blank parts is, and the greater the continuity threshold in the continuity preset condition that the "continuous number of semi-finished product inspection results meeting the inspection value preset condition is less than the continuity threshold".

[0182] For more information about the abnormal type distribution, see Figure 8 and the related description. For more information about the estimated quality, see Figure 3 and the related description.

[0183] In some embodiments of the present specification, the continuity preset condition is determined in combination with the abnormal type distribution and the estimated quality of the plurality of continuous to-be-processed blank parts. Whether the cause of the machining abnormality is a problem existing in the pin shaft machining process, the mold, etc. can be further considered, so that the continuity preset condition is reasonably adjusted according to the actual situation to ensure the stable progress of the pin shaft machining process.

[0184] In some embodiments of the present specification, by performing semi-finished product inspection on the semi-finished parts and judging whether to stop machining based on the semi-finished product inspection result, it can be determined in time whether a problem exists in a certain process in the pin shaft machining process, and the machining is stopped in time after the problem is found to reduce the waste of resources. In addition, the inspection value preset condition and the continuity preset condition are determined in combination with factors such as the deformation degree, the stability, the abnormal type distribution, and the estimated quality, so that the set inspection value preset condition and the continuity preset condition are more reasonable, and the accuracy of the judgment result of judging whether to stop machining is ensured.

[0185] Figure 8 is an exemplary flowchart of determining the pin shaft machining process optimization parameter according to some embodiments of the present specification. In some embodiments, the flow 800 can be executed by the management platform 110. As shown in Figure 8 , the flow 800 includes the following steps:

[0186] At step 810, based on the finished product, a finished product inspection is performed to obtain a finished product inspection result. The finished product inspection result includes an abnormality rate of the finished product and an abnormality type distribution of the finished product. In some embodiments, step 810 can be performed by the inspection module 250.

[0187] For more information about the finished product, see Figure 3 and related descriptions.

[0188] In some embodiments, the inspection module 250 can perform inspection on the finished product based on various processing piece inspection methods (e.g., sampling inspection, etc.).

[0189] In some embodiments, the inspection module 250 can perform sampling on the finished product, compare the sample finished product with historical qualified finished products in terms of various elements, and thus achieve finished product inspection. For example, the inspection module 250 can obtain image information of the sample finished product through an image acquisition device, obtain historical image information of historical qualified finished products from the storage device 113, and process the image information of the sample finished product and the historical image information of the historical qualified finished products through an image analysis method, thus achieving finished product inspection. For another example, the inspection module 250 can obtain the weight of the sample finished product through a weight sensor, obtain the historical weight of the historical qualified finished products from the storage device 113, and compare the weight of the sample finished product with the historical weight of the historical qualified finished products directly, thus achieving finished product inspection.

[0190] The finished product inspection result refers to a result reflecting the processing of the finished product after the finished product inspection is performed on the finished product. The finished product inspection result includes the abnormality rate of the finished product and the abnormality type distribution of the finished product.

[0191] The abnormality rate refers to the proportion of abnormal products in the total products subjected to inspection. For example, the abnormality rate can be 80%.

[0192] The abnormality type distribution refers to the proportion distribution of different abnormality types in the abnormal products subjected to inspection. The abnormality type distribution can be represented by a vector. For example, the abnormality type distribution can be (30, 20, 30, 20), indicating that the abnormal products of the abnormality type of “surface crack” account for 30% of all abnormal products, the abnormal products of the abnormality type of “internal crack” account for 20% of all abnormal products, the abnormal products of the abnormality type of “surface folding” account for 30% of all abnormal products, and the abnormal products of the abnormality type of “axial bending” account for 20% of all abnormal products.

[0193] In some embodiments, the inspection module 250 can determine whether to perform a stoppage inspection on the pin shaft machining device based on the finished product inspection result. For example, in response to the abnormality rate being greater than a threshold (e.g., 80%) and the abnormality type distribution being concentrated on a certain abnormality (e.g., the abnormality type distribution is (10, 70, 10, 10)), a stoppage inspection is performed on the pin shaft machining device. The abnormality type distribution being concentrated on a certain abnormality means that a certain value A in the abnormality type distribution is significantly greater than other values, and thus the abnormality type distribution is concentrated on the abnormality type corresponding to the value A. For example, the abnormality type distribution is (10, 70, 10, 10), and thus the abnormality type distribution is concentrated on the abnormality type corresponding to the value 70; for another example, the abnormality type distribution is (5, 10, 5, 10, 60, 10), and thus the abnormality type distribution is concentrated on the abnormality type corresponding to the value 60.

[0194] At step 820, a pin shaft machining process optimization parameter is determined based on the finished product inspection result. In some embodiments, step 820 can be performed by the optimization module 260.

[0195] The pin shaft machining process optimization parameter refers to a parameter obtained after the pin shaft machining process parameter is optimized. For more information about the pin shaft machining process parameter, see Figure 3 and the related description.

[0196] In some embodiments, the optimization module 260 can determine the pin shaft machining process optimization parameter based on historical data. For example, the optimization module 260 can obtain the best finished product inspection result and the pin shaft machining process parameter corresponding to the best finished product inspection result from the historical finished product inspection results stored in the storage device 113, and take the pin shaft machining process parameter corresponding to the best finished product inspection result as the pin shaft machining process optimization parameter.

[0197] In some embodiments, the optimization module 260 can determine the pin shaft machining process optimization parameter through sub-steps 821-824.

[0198] At sub-step 821, a constraint condition is determined based on the shape difference degree between the to-be-machined blank and the finished product.

[0199] The shape difference degree refers to the degree of shape difference between the to-be-machined blank and the finished product obtained after the blank is machined. The smaller the shape difference degree, the smaller the degree of shape difference between the to-be-machined blank and the finished product obtained after the blank is machined.

[0200] In some embodiments, the shape difference degree can be determined based on the morphological sampling features of the blank and the morphological sampling features of the finished product. For example, a certain type of feature vector in the morphological sampling features of the blank is obtained, a feature vector of the same type in the morphological sampling features of the finished product is obtained, the distance (e.g., Euclidean distance, etc.) between the two vectors is calculated, and the calculation result is taken as the shape difference degree. The smaller the calculation result, the smaller the shape difference degree.

[0201] For example, for different types of feature vectors, based on one type, a feature vector of the same type in the morphological sampling features of the blank is obtained, a feature vector of the same type in the morphological sampling features of the finished product is obtained, the distance between the two vectors is calculated to obtain the vector distance corresponding to the feature vector of the same type, and the vector distances corresponding to the feature vectors of other types are calculated in the same way. Finally, the average value of all vector distances is calculated, and the calculation result is taken as the shape difference degree. The smaller the calculation result, the smaller the shape difference degree, and the smaller the shape difference between the blank and the finished product obtained after processing the blank. For more information about morphological sampling features and feature vectors, see Figure 4 and related descriptions.

[0202] In some embodiments of the present specification, determining the shape difference degree based on the morphological sampling features can make the way of obtaining the shape difference degree more convenient, and can also ensure the accuracy of the obtained shape difference degree.

[0203] The constraint condition refers to the adjustment range of the pin shaft processing parameters. The constraint condition can include various types, such as the constraint condition of the number of cold heading sub-processes, the constraint condition of the tonnage of the selected extrusion equipment, the constraint condition of the extrusion rate, the constraint condition of the lifting rate, the constraint condition of the pipeline moving rate, etc.

[0204] In some embodiments, the constraint condition can be obtained based on historical data. For example, the pin shaft processing parameters corresponding to the finished products whose inspection results are greater than a threshold value (e.g., 80%) are obtained, and are taken as the constraint condition.

[0205] In some embodiments, the constraint range of the number of cold heading sub-processes can be determined based on the shape difference degree. For example, the greater the shape difference degree, the more the number of cold heading sub-processes in the pin shaft processing parameters, so a larger parameter adjustment range (e.g., the number of cold heading sub-processes is 5-20) can be preset. For example, the smaller the shape difference degree, the fewer the number of cold heading sub-processes in the pin shaft processing parameters, so a smaller parameter adjustment range (e.g., the number of cold heading sub-processes is 2-5) can be preset.

[0206] Sub-step 822, based on the constraint condition, generate multiple groups of candidate pin shaft processing parameters.

[0207] The candidate pin shaft machining process parameters refer to the pin shaft machining process parameters generated based on the constraints. The candidate pin shaft machining process parameters contain different types of candidate parameters, such as the candidate number of cold heading sub-processes, the candidate tonnage of selected extrusion equipment, the candidate speed of line movement, and the like.

[0208] Based on different types of constraints, different types of candidate parameters in the candidate pin shaft machining process parameters can be generated. For example, the number of cold heading sub-processes in the constraints is 2-5, and the candidate number of cold heading sub-processes in the candidate pin shaft machining process parameters has 4 groups, which are: the number of cold heading sub-processes is 2, the number of cold heading sub-processes is 3, the number of cold heading sub-processes is 4, and the number of cold heading sub-processes is 5.

[0209] For each type of candidate parameter, one parameter in each type of candidate parameter is randomly selected to form a group of candidate pin shaft machining process parameters, and then a plurality of groups of candidate pin shaft machining process parameters are determined by the same method.

[0210] Sub-step 823, determine the evaluation value of each group of candidate pin shaft machining process parameters.

[0211] The evaluation value refers to a numerical value used to represent the effect achieved by processing the to-be-processed blank based on the group of candidate pin shaft machining process parameters. The evaluation value can be represented by a score, for example, the evaluation value can be represented by 1-10 points, and the higher the score, the better the processing effect of the candidate pin shaft machining process parameter corresponding to the evaluation value.

[0212] In some embodiments, the optimization module 260 can determine the evaluation value based on the estimated cost, the estimated efficiency, and the estimated abnormal rate. The lower the estimated cost, the higher the estimated efficiency, and the lower the estimated abnormal rate, the higher the evaluation value.

[0213] The estimated cost refers to the cost consumed by the pin shaft machining based on the group of candidate pin shaft machining process parameters. In some embodiments, the estimated cost can be determined based on the number of cold heading sub-processes in the candidate pin shaft machining process parameters, and the more the number of cold heading sub-processes, the higher the estimated cost.

[0214] The estimated efficiency refers to the generation efficiency of the pin shaft machining based on the group of candidate pin shaft machining process parameters. In some embodiments, the estimated efficiency can be determined based on the extrusion speed, the lifting speed, and the line movement speed in the candidate pin shaft machining process parameters, and the greater the extrusion speed, the greater the lifting speed, and the greater the line movement speed, the faster the estimated efficiency.

[0215] The estimated abnormality rate refers to the proportion of abnormal products in the finished products obtained by processing the pin shaft based on the set of candidate pin shaft processing parameters. In some embodiments, the estimated abnormality rate can be determined based on all parameters in the set of candidate pin shaft processing parameters. For example, based on all parameters in the set of candidate pin shaft processing parameters, 10,000 pin shafts are simulated to be manufactured by the simulation system, and the proportion of abnormal products in the obtained finished products is determined. Random noise can be added in some steps of the simulation to simulate the actual pin shaft processing environment and improve the accuracy of the simulation results.

[0216] In sub-step 824, the target pin shaft processing parameter is determined based on the evaluation value, and the target pin shaft processing parameter is used as the pin shaft processing optimization parameter.

[0217] The target pin shaft processing parameter refers to the parameter in the set of candidate pin shaft processing parameters that meets the selection condition. The selection condition refers to a condition for determining whether the candidate pin shaft processing parameter can be used as the target pin shaft processing parameter. For example, the selection condition can be that the evaluation value is the largest, and the candidate pin shaft processing parameter with the largest evaluation value can be directly used as the target pin shaft processing parameter. For another example, the selection condition can be that the evaluation value is greater than a threshold value (e.g., 8 points), and the average value of the multiple sets of candidate pin shaft processing parameters with evaluation values greater than the threshold value can be used as the target pin shaft processing parameter.

[0218] In some embodiments of the present specification, by obtaining multiple sets of candidate pin shaft processing parameters and their evaluation values, and then determining the final pin shaft processing optimization parameter, the final determination result can be more accurate. Moreover, by comprehensively considering the estimated cost, the estimated efficiency, and the estimated abnormality rate when determining the evaluation value, the final determined pin shaft processing optimization parameter can be more practical.

[0219] In some embodiments of the present specification, by performing the finished product inspection, the quality of the finished product can be further checked based on the semi-finished product inspection, and abnormalities in the pin shaft processing process can be found in time and the processing is stopped, thereby reducing the loss. Moreover, by optimizing the pin shaft processing parameter based on the finished product inspection result, the parameter can be automatically adjusted in combination with the actual processing situation, so that the pin shaft processing equipment is in the most efficient and optimal processing state, and the processing efficiency and quality of the pin shaft are ensured.

[0220] It should be noted that the above description of the processes 300, 400, 600, and 800 is only for example and illustration, and does not limit the scope of the present specification. Those skilled in the art can make various modifications and changes to the processes 300, 400, 600, and 800 under the guidance of the present specification. However, these modifications and changes are still within the scope of the present specification.

[0221] Having described the basic concepts, it is obvious that the above detailed disclosure is merely intended for purposes of illustration and is not intended to limit the present specification. Although the present specification has not explicitly described, those skilled in the art can make various modifications, improvements and corrections to the present specification. Such modifications, improvements and corrections are suggested in the present specification, and thus still belong to the spirit and scope of the exemplary embodiments of the present specification.

[0222] Meanwhile, the present specification uses specific terms to describe the embodiments of the present specification. As "one embodiment", "an embodiment", and / or "some embodiments" mean that a certain feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the present specification. Therefore, it is emphasized and noted that "a" or "one" embodiment or "an" or "one" alternative embodiment appearing in various positions of the present specification does not necessarily refer to the same embodiment. In addition, certain features, structures, or characteristics of one or more embodiments of the present specification can be properly combined.

[0223] Further, unless the claim explicitly states otherwise, the order of the processing elements and sequences, the use of the lettering, or the use of other designations in the present specification are not intended to limit the order of the processes and methods of the present specification. Although some presently preferred inventive embodiments are discussed in the above disclosure by various examples, it is to be understood that such details are merely for the purpose of illustration and additional claims are not limited to the disclosed embodiments, but rather intended to cover all modifications and equivalent arrangements included within the spirit and scope of the inventive embodiments. For example, although the system components described above can be implemented by hardware devices, they can also be implemented by only software solutions, such as installing the described system on an existing server or mobile device.

[0224] Similarly, it is to be noted that, in order to simplify the description of the present specification and to help the understanding of one or more inventive embodiments, the above description of the embodiments of the present specification sometimes combines various features into one embodiment, drawing, or description thereof. However, this disclosure method does not mean that the features required by the present specification are more than those mentioned in the claims. In fact, the features of the embodiments are less than all the features of the above-disclosed single embodiment.

[0225] In some embodiments, numbers describing the quantity of components and attributes are used. It should be understood that such numbers used in the description of embodiments are modified in some examples with the terms "approximately," "approximately," or "generally." Unless otherwise stated, "approximately," "approximately," or "generally" indicates that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may be changed depending on the characteristics required by individual embodiments. In some embodiments, numerical parameters should take into account specified significant digits and employ a general method of digit reservation. Although the numerical ranges and parameters used to confirm their breadth of range in some embodiments of this specification are approximate values, in specific embodiments, such values ​​are set as precisely as feasible.

[0226] For each patent, patent application, patent application publication, and other material, such as articles, books, specifications, publications, and documents, referenced in this specification, the entire contents of which are incorporated herein by reference. This excludes historical application documents that are inconsistent with or conflict with the content of this specification, as well as documents that limit the broadest scope of the claims in this specification (currently or subsequently appended to this specification). It should be noted that in the event of any inconsistency or conflict between the descriptions, definitions, and / or terminology used in the supplementary materials to this specification and the content of this specification, the descriptions, definitions, and / or terminology used in this specification shall prevail.

[0227] Finally, it should be understood that the embodiments described in this specification are merely illustrative of the principles of the embodiments described herein. Other variations may also fall within the scope of this specification. Therefore, alternative configurations of the embodiments described herein are intended to be consistent with the teachings of this specification, rather than as examples or limitations. Accordingly, the embodiments described herein are not limited to the implementations explicitly presented and described herein.

Claims

1. An automated machining method for pins, characterized in that, The method is executed by the management platform, and the method includes: Obtain feature information of the blank to be processed, the feature information including two-dimensional feature information and three-dimensional feature information of the blank to be processed, the two-dimensional feature information being related to image information of the blank to be processed and / or image features of the image information; the three-dimensional feature information includes morphological sampling features, the morphological sampling features including one or more feature vectors, the feature vectors including single-distance feature vectors, area feature vectors, volume feature vectors, curvature feature vectors or distance feature vectors, the morphological sampling features including at least the distance feature vectors, the distance feature vectors used to determine the feature vectors include: S1. For each sampling, obtain any two sampling points on the surface of the three-dimensional model of the blank to be processed; S2. Use the distance between two sampling points as the feature value; S3. Repeat S1-S2 to obtain multiple feature values, count the number of times different feature values ​​appear, and use them as the distance feature vector; Based on the aforementioned feature information, cold heading process parameters, and thread rolling process parameters, the estimated quality of the blank to be processed into a pin is predicted. In response to the estimated quality meeting the preset quality conditions, the blank to be processed is transported to the target workstation by the gripping device; Based on the cold heading process parameters, the blank to be processed is subjected to a cold heading process to obtain a semi-finished part. The cold heading process includes multiple cold heading sub-processes, which are predetermined based on the pin machining requirements. During the cold heading process, a semi-finished product inspection is performed, including: Based on the semi-finished parts obtained after each cold heading process of each blank part to be processed, a semi-finished product inspection is performed to obtain the semi-finished product inspection results. In response to the semi-finished product inspection results not meeting the preset conditions for inspection values ​​and continuity, the automated processing of the pin shaft is stopped; the preset conditions for inspection values ​​are related to the degree of deformation, and the preset conditions for continuity are related to the stability of the semi-finished product inspection results, the distribution of abnormality types, and the estimated quality of multiple consecutive blanks to be processed; Based on the aforementioned thread rolling process parameters, the semi-finished part is subjected to a thread rolling process to obtain the finished part.

2. The method according to claim 1, characterized in that, The method for obtaining the three-dimensional feature information includes: Obtain the sampling data of the blank to be processed; Based on the sampled data, a three-dimensional model of the blank to be processed is generated; Based on the three-dimensional model, the three-dimensional feature information of the blank to be processed is determined.

3. The method according to claim 1, characterized in that, The method further includes: Based on the finished parts, perform finished product inspection and obtain finished product inspection results, including the abnormality rate and abnormality type distribution of the finished parts; Based on the finished product inspection results, the optimized parameters for the pin machining process were determined.

4. An automated pin machining system, characterized in that, The system includes a management platform, and the system also includes: The acquisition module is used to acquire feature information of the blank to be processed. The feature information includes two-dimensional and three-dimensional feature information of the blank. The two-dimensional feature information is related to image information of the blank and / or image features of the image information. The three-dimensional feature information includes morphological sampling features, which include one or more feature vectors. These feature vectors include single-distance feature vectors, area feature vectors, volume feature vectors, curvature feature vectors, or distance feature vectors. The morphological sampling features at least include the distance feature vector. The acquisition module is further used to determine the distance feature vector of the feature vector. S1. For each sampling, obtain any two sampling points on the surface of the three-dimensional model of the blank to be processed; S2. Use the distance between two sampling points as the feature value; S3. Repeat S1-S2 to obtain multiple feature values, and count the number of times different feature values ​​appear, which is used as the distance feature vector; The prediction module is used to predict the estimated quality of the blank to be processed into a pin based on the feature information, cold heading process parameters and thread rolling process parameters. The gripping module, in response to the estimated quality meeting the preset quality conditions, transports the blank to be processed to the target station through the gripping device; Processing module, used for Based on the cold heading process parameters, the blank to be processed is subjected to a cold heading process to obtain a semi-finished part. The cold heading process includes multiple cold heading sub-processes, which are predetermined based on the pin machining requirements. During the cold heading process, a semi-finished product inspection is performed, including: Based on the semi-finished parts obtained after each cold heading process of each blank part to be processed, a semi-finished product inspection is performed to obtain the semi-finished product inspection results. In response to the semi-finished product inspection results not meeting the preset conditions for inspection values ​​and continuity, the automated processing of the pin shaft is stopped; the preset conditions for inspection values ​​are related to the degree of deformation, and the preset conditions for continuity are related to the stability of the semi-finished product inspection results, the distribution of abnormality types, and the estimated quality of multiple consecutive blanks to be processed; Based on the aforementioned thread rolling process parameters, the semi-finished part is subjected to a thread rolling process to obtain the finished part.

5. The system according to claim 4, characterized in that, The acquisition module is also used for: Obtain the sampling data of the blank to be processed; Based on the sampled data, a three-dimensional model of the blank to be processed is generated; Based on the three-dimensional model, the three-dimensional feature information of the blank to be processed is determined.

6. The system according to claim 4, characterized in that, The system also includes: The inspection module is used to perform finished product inspection based on the finished parts and obtain the finished product inspection results, which include the abnormality rate and abnormality type distribution of the finished parts. The optimization module is used to determine the optimization parameters of the pin machining process based on the finished product inspection results.

7. An automated pin machining device, the device comprising at least one processor and at least one memory; The at least one memory is used to store computer instructions; The at least one processor is used to execute at least some of the instructions in the computer instructions to implement the pin automatic machining method as described in any one of claims 1 to 3.

8. A computer-readable storage medium storing computer instructions that, when executed by a processor, implement the pin-axis automated machining method as described in any one of claims 1 to 3.

Images