Product quality online monitoring method and system applied to intelligent manufacturing

Abstract

The embodiment of the invention provides a product quality online monitoring method and system applied to intelligent manufacturing, and belongs to the field of control systems.The method comprises the steps that welding related data in the process that a welding device welds a circuit board to be subjected to quality inspection is obtained, and whether the welding process is abnormal or not is judged; if it is judged that the welding process is abnormal, welding parameters are adjusted, and the welding equipment welds the circuit board with the same model as the circuit board to be subjected to quality inspection based on the adjusted welding parameters; if it is judged that the welding process is normal, point cloud data of welding spots of the circuit to be subjected to quality inspection are obtained, and whether the forms of the welding spots are abnormal or not is judged; if it is judged that the form of the welding spot is abnormal, the state of the welding equipment is determined based on historical welding data of the welding equipment; if it is judged that the form of the welding spot is normal, circuit testing is conducted on the welding spot of the circuit to be subjected to quality inspection through electrical testing equipment, and testing data are obtained; based on the test data, the state of the welding spot of the circuit board to be subjected to quality inspection is judged, and the method has the advantage of improving the welding spot detection efficiency and accuracy.

Description

TECHNICAL FIELD

[0001] The present specification relates to the field of control systems, and in particular to a product quality online monitoring method and system applied to intelligent manufacturing. BACKGROUND

[0002] After the third scientific and technological revolution, microelectronic technology has developed extremely rapidly. In order to meet the needs of the people for a better life, various powerful digital products and household appliances are also becoming more and more common in people's lives. As the functions of electronic products become more and more powerful, the integration of circuits is also becoming higher and higher, and the components that need to be assembled are also becoming more and more numerous and precise, so the solder joints on the circuit board have become more complex and delicate. The traditional circuit board solder joint detection is completed manually, but manual detection is low in efficiency and high in cost, which cannot meet the production needs today. The Chinese patent with the application number CN202110336125.1 discloses a soldering quality processing method, which comprises: obtaining warping data of each layer of a multi-layer circuit board under a preset soldering temperature change curve; generating warping levels of each region of the multi-layer circuit board in the stacking state according to the stacking state and the warping data simulation of the multi-layer circuit board; processing the multi-layer circuit board according to the warping levels; and processing the multi-layer circuit board according to the warping levels, which includes adjusting the design parameters of the multi-layer circuit board, including one or more of the following: residual copper rate of the multi-layer circuit board; relative position of the multi-layer circuit board; material of the multi-layer circuit board; and soldering position of the components of the multi-layer circuit board. The above-mentioned soldering quality processing method only judges the soldering quality according to the warping data of the circuit board, which has low accuracy.

[0003] Therefore, it is necessary to provide a product quality online monitoring method and system applied to intelligent manufacturing, which is used to improve the solder joint detection efficiency and accuracy. SUMMARY

[0004] One of the embodiments of the present specification provides a product quality online monitoring method applied to intelligent manufacturing, which comprises: obtaining soldering related data in the process of soldering a circuit board to be inspected by a soldering device; judging whether an abnormality occurs in the soldering process based on the soldering related data; if it is judged that an abnormality occurs in the soldering process, adjusting the soldering parameters, and the soldering device soldering a circuit board of the same model as the circuit board to be inspected based on the adjusted soldering parameters; if it is judged that the soldering process is normal, obtaining point cloud data of solder joints of the circuit to be inspected; judging whether the shape of the solder joints is abnormal based on the point cloud data of the solder joints of the circuit to be inspected; if it is judged that the shape of the solder joints is abnormal, determining the state of the soldering device based on the historical soldering data of the soldering device; if it is judged that the shape of the solder joints is normal, performing a circuit test on the solder joints of the circuit to be inspected by an electrical detection device to obtain test data; and judging the state of the solder joints of the circuit board to be inspected based on the test data.

[0005] In some embodiments, the welding-related data is obtained during a welding process of the welding device on the circuit board to be inspected, including: obtaining, during the welding process, first vibration data of the welding device, second vibration data of the circuit board to be inspected, temperature information of at least one position of the circuit board to be inspected, and environmental smoke concentration information at multiple time points; generating a welding-related sequence, wherein an element of the welding-related sequence corresponds to the first vibration data of the welding device, the second vibration data of the circuit board to be inspected, the temperature information of at least one position of the circuit board to be inspected, and the environmental smoke concentration information at a time point; and determining whether an abnormality occurs in the welding process based on the welding-related sequence by an abnormality determination model.

[0006] In some embodiments, the temperature information of at least one position of the circuit board to be inspected is obtained during the welding process, including: obtaining the temperature information of at least one position of the circuit board to be inspected by an infrared thermal imager.

[0007] In some embodiments, the welding parameters are adjusted, including: adjusting the welding parameters based on the welding-related data by a parameter adjustment model, wherein the welding parameters include at least one of a welding temperature, a thickness of solder paste, and a welding sequence of welding points.

[0008] In some embodiments, the point cloud data of the welding points of the circuit to be inspected is obtained, including: obtaining point cloud information of the circuit to be inspected by the point cloud acquisition device; and performing a separation operation on the point cloud information of the circuit to be inspected to obtain the point cloud data of the welding points of the circuit to be inspected.

[0009] In some embodiments, the shape of the welding points is determined based on the point cloud data of the welding points of the circuit to be inspected, including: obtaining a contour shape of the welding points based on the point cloud data of the welding points of the circuit to be inspected; and determining whether the shape of the welding points is abnormal based on the contour shape of the welding points.

[0010] In some embodiments, the point cloud acquisition device comprises a shooting pose adjustment apparatus, a structured light binocular vision apparatus and a data processor. The structured light binocular vision apparatus is arranged on the shooting pose adjustment apparatus. The shooting pose adjustment apparatus is used to adjust the pose of the structured light binocular vision apparatus. The structured light binocular vision apparatus is used to shoot the circuit board to be inspected to obtain left-eye and right-eye images of the circuit board to be inspected. The data processor is used to generate a target pose of a circuit diagram of the circuit board to be inspected, which is marked with solder joints, and control the shooting pose adjustment apparatus to adjust the pose of the shooting pose adjustment apparatus to the target pose. The data processor is also used to control the structured light binocular vision apparatus to obtain the left-eye and right-eye images of the circuit board to be inspected at the target pose. The data processor is further used to generate point cloud information of the circuit to be inspected based on the left-eye and right-eye images of the circuit board to be inspected obtained by the structured light binocular vision apparatus.

[0011] In some embodiments, the circuit test of the solder joints of the circuit to be inspected by the electrical detection device to obtain test data comprises: for each solder joint, determining at least one pair of detection points of the solder joint on the circuit board to be inspected based on the design information of the circuit board to be inspected by a solder joint determination model; and obtaining the test data based on the at least one pair of detection points by the electrical detection device.

[0012] In some embodiments, the circuit test of the solder joints of the circuit to be inspected by the electrical detection device to obtain test data comprises: determining the state of each solder joint of the circuit board to be inspected based on the design information of the circuit board to be inspected and the test data by a state determination model, and outputting a test report, wherein the test report comprises a circuit diagram of the circuit board to be inspected, which is marked with faulty solder joints and the state of each faulty solder joint.

[0013] One of the embodiments of the present specification provides a product quality online monitoring system applied to intelligent manufacturing, the system comprises: a data acquisition module, configured to acquire welding related data in a process of welding a to-be-inspected circuit board by a welding device; an abnormality judgment module, configured to judge whether an abnormality occurs in the welding process based on the welding related data; a parameter adjustment module, configured to adjust welding parameters if the abnormality judgment module judges that an abnormality occurs in the welding process, and the welding device welds a circuit board of the same model as the to-be-inspected circuit board based on the adjusted welding parameters; a point cloud acquisition module, configured to acquire point cloud data of a welding point of the to-be-inspected circuit if the abnormality judgment module judges that the welding process is normal; a welding point judgment module, configured to judge whether the form of the welding point is abnormal based on the point cloud data of the welding point of the to-be-inspected circuit; a state determination module, configured to determine the state of the welding device based on historical welding data of the welding device if the welding point judgment module judges that the form of the welding point is abnormal; an electrical test module, configured to perform circuit test on the welding point of the to-be-inspected circuit by an electrical test device to acquire test data if the welding point judgment module judges that the form of the welding point is normal; and a welding point detection module, configured to judge the state of the welding point of the to-be-inspected circuit board based on the test data.

[0014] Compared with the prior art, the product quality online monitoring method and system applied to intelligent manufacturing at least have the following beneficial effects:

[0015] 1. First, whether an abnormality occurs in the welding process is judged according to the welding related data in the process of welding the to-be-inspected circuit board by the welding device, and only when the welding process is judged to be normal, the point cloud acquisition and the judgment of whether the form of the welding point is abnormal are performed, and only when the form of the welding point is judged to be normal, the final circuit test is performed, and the state of the welding point of the to-be-inspected circuit board is judged according to the test data of the circuit test, thereby avoiding the circuit board with welding abnormality from entering the subsequent welding point form judgment and circuit test, and also avoiding the circuit board with welding point form abnormality from entering the subsequent circuit test, effectively improving the welding point detection efficiency and accuracy;

[0016] 2. When the welding process is judged to be abnormal, the welding parameters are adjusted in time to avoid the subsequent circuit board from continuing to have welding abnormality, thereby effectively avoiding loss;

[0017] 3. When the form of the welding point is judged to be abnormal, the state of the welding device is determined based on the historical welding data of the welding device, so that the welding device with fault can be found in time for repair or replacement, thereby avoiding the subsequent circuit board from continuing to have welding abnormality, and effectively avoiding loss;

[0018] 4, the circuit diagram corresponding to the to-be-inspected circuit and having the welding point is generated with a target pose, and the shooting pose adjusting device is controlled to adjust the pose of the shooting pose adjusting device to the target pose, and the structured light binocular vision device is controlled to acquire the left-eye image and the right-eye image of the to-be-inspected circuit board at the target pose, so that the image of the welding point is clearer, and the point cloud data generated subsequently is more accurate. BRIEF DESCRIPTION OF DRAWINGS

[0019] The present specification will be further illustrated in the manner of exemplary embodiments, which will be described in detail through the accompanying drawings. These embodiments are not restrictive, and in these embodiments, the same numbers represent the same structures, in which:

[0020] Figure 1 is a module diagram of a product quality online monitoring system applied to intelligent manufacturing according to some embodiments of the present specification;

[0021] Figure 2 is a flowchart of a product quality online monitoring method applied to intelligent manufacturing according to some embodiments of the present specification. DETAILED DESCRIPTION

[0022] In order to more clearly illustrate the technical solutions of the embodiments of the present specification, the drawings needed to be used in the embodiment description will be briefly introduced below. 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 be applied to other similar scenarios without creative labor on the basis of these drawings. Unless it is clear from the language environment or otherwise stated, the same reference numbers in the drawings represent the same structures or operations.

[0023] 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.

[0024] As shown in the specification and claims, unless the context clearly indicates otherwise, "one", "a", "an", and / or "the" do not refer to the singular, but can also include the plural. Generally speaking, the terms "comprise" and "include" 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.

[0025] Flowcharts are used in the specification to illustrate the operation of systems in accordance with embodiments of the specification. It should be understood that the operations in the figures do not necessarily have to be performed in the precise order shown. Rather, various steps can be handled in an inverse order or simultaneously with each other. Also, other operations can be added or removed from the processes.

[0026] Figure 1 is a module diagram of an exemplary product quality online monitoring system for intelligent manufacturing according to some embodiments of the specification. As shown in Figure 1 the product quality online monitoring system for intelligent manufacturing can include a data acquisition module, an abnormality judgment module, a parameter adjustment module, a point cloud acquisition module, a solder joint judgment module, a state determination module, an electrical test module, and a solder joint detection module.

[0027] The data acquisition module can be configured to acquire welding-related data during the welding of the circuit board to be inspected by the welding equipment.

[0028] The abnormality judgment module can be configured to determine whether an abnormality occurs in the welding process based on the welding-related data.

[0029] In some embodiments, the abnormality judgment module can be further configured to: acquire, at multiple time points during the welding process, first vibration data of the welding equipment, second vibration data of the circuit board to be inspected, temperature information of at least one position of the circuit board to be inspected, and environmental smoke concentration information; generate a welding-related sequence, wherein one element of the welding-related sequence corresponds to the first vibration data of the welding equipment, the second vibration data of the circuit board to be inspected, the temperature information of at least one position of the circuit board to be inspected, and the environmental smoke concentration information at a time point; and determine whether an abnormality occurs in the welding process based on the welding-related sequence by using an abnormality judgment model.

[0030] In some embodiments, the abnormality judgment module can be further configured to: acquire the temperature information of at least one position of the circuit board to be inspected by using an infrared thermal imager.

[0031] The parameter adjustment module can be configured to adjust the welding parameters if the abnormality judgment module determines that an abnormality occurs in the welding process, and the welding equipment welds a circuit board of the same model as the circuit board to be inspected based on the adjusted welding parameters.

[0032] In some embodiments, the parameter adjustment module can be further configured to: adjust the welding parameters based on the welding-related data by using a parameter adjustment model, wherein the welding parameters include at least one of a welding temperature, a solder paste thickness, and a welding sequence of a solder joint.

[0033] The point cloud acquisition module can be configured to acquire point cloud data of a solder joint of the circuit board to be inspected if the abnormality judgment module determines that the welding process is normal.

[0034] In some embodiments, the point cloud obtaining module can also be configured to: obtain, by the point cloud obtaining device, the point cloud information of the circuit to be inspected; and perform a separation operation on the point cloud information of the circuit to be inspected to obtain the point cloud data of the solder joint of the circuit to be inspected.

[0035] The solder joint determining module can be configured to determine, based on the point cloud data of the solder joint of the circuit to be inspected, whether the morphology of the solder joint is abnormal.

[0036] In some embodiments, the solder joint determining module can also be configured to: obtain, based on the point cloud data of the solder joint of the circuit to be inspected, a solder joint contour shape; and determine, based on the solder joint contour shape, whether the morphology of the solder joint is abnormal.

[0037] The state determining module can be configured to, if the solder joint determining module determines that the morphology of the solder joint is abnormal, determine the state of the welding device based on historical welding data of the welding device.

[0038] The electrical test module can be configured to, if the solder joint determining module determines that the morphology of the solder joint is normal, perform an electrical test on the solder joint of the circuit to be inspected by the electrical detection device to obtain test data.

[0039] In some embodiments, the electrical test module can also be configured to: for each solder joint, determine, by the solder joint determining model, at least one pair of detection points of the solder joint on the circuit board to be inspected based on design information of the circuit board to be inspected; and obtain, by the electrical detection device, the test data based on the at least one pair of detection points.

[0040] The solder joint detection module can be configured to determine, based on the test data, the state of the solder joint of the circuit board to be inspected.

[0041] In some embodiments, the solder joint detection module can also be configured to: determine, by the state determining model, the state of each solder joint of the circuit board to be inspected based on the design information of the circuit board to be inspected and the test data, and output a test report, wherein the test report includes a circuit diagram of the circuit board to be inspected with faulty solder joints and the state of each faulty solder joint marked.

[0042] For more descriptions of the data obtaining module, the abnormality determining module, the parameter adjusting module, the point cloud obtaining module, the solder joint determining module, the state determining module, the electrical test module, and the solder joint detection module, please refer to Figure 2 and related descriptions thereof, which will not be repeated here.

[0043] Figure 2 is a flowchart of an exemplary product quality online monitoring method applied to intelligent manufacturing according to some embodiments of the present specification. In some embodiments, the product quality online monitoring method applied to intelligent manufacturing can be executed by a product quality online monitoring system applied to intelligent manufacturing.

[0044] At step 210, welding-related data is acquired during the welding process of the welding equipment on the circuit board to be inspected. In some embodiments, step 210 can be performed by a data acquisition module.

[0045] In some embodiments, the data acquisition module can acquire the first vibration data of the welding equipment, the second vibration data of the circuit board to be inspected, the temperature information of at least one position of the circuit board to be inspected, and the environmental smoke concentration information at multiple time points during the welding process. For example, the data acquisition module can acquire the first vibration data of the welding equipment through the first vibration sensor installed on the welding equipment, acquire the second vibration data of the circuit board to be inspected through the second vibration sensor arranged on the circuit board to be inspected, acquire the temperature information of at least one position of the circuit board to be inspected through the infrared thermal imager, and acquire the environmental smoke concentration information through the smoke sensor arranged on the welding equipment. During the welding process, the smoke sensor can be arranged above the circuit board to be inspected.

[0046] In some embodiments, the data acquisition module can generate a welding-related sequence, wherein one element of the welding-related sequence corresponds to the first vibration data of the welding equipment, the second vibration data of the circuit board to be inspected, the temperature information of at least one position of the circuit board to be inspected, and the environmental smoke concentration information at a time point. The elements in the welding-related sequence can be arranged in chronological order.

[0047] At step 220, whether an abnormality occurs in the welding process is determined based on the welding-related data. In some embodiments, step 220 can be performed by an abnormality determination module.

[0048] In some embodiments, the abnormality determination module can determine whether an abnormality occurs in the welding process based on the welding-related sequence through an abnormality determination model. The input of the abnormality determination module can include the welding-related sequence, and the output of the abnormality determination module can include a determination result indicating whether an abnormality occurs in the welding process.

[0049] In some embodiments, the abnormality determination model can include, but is not limited to, a neural network (NN), a convolutional neural network (CNN), a deep neural network (DNN), a recurrent neural network (RNN), or any combination thereof, for example, the abnormality determination model can be a model formed by combining a convolutional neural network and a deep neural network.

[0050] In some embodiments, if it is determined that the welding process is normal, step 240 is performed, and if it is determined that an abnormality occurs in the welding process, step 230 is performed.

[0051] At step 230, if it is determined that an abnormality occurs in the welding process, the welding parameters are adjusted, and the welding equipment welds a circuit board of the same model as the circuit board to be inspected based on the adjusted welding parameters. In some embodiments, step 230 can be performed by a parameter adjustment module.

[0052] In some embodiments, the parameter adjustment module can adjust the welding parameters based on the welding-related data by a parameter adjustment model, wherein the welding parameters include at least one of a soldering temperature, a solder paste thickness, and a soldering sequence of the solder joints. The input of the parameter adjustment model can include the welding-related sequence and the current welding parameters (e.g., at least one of the soldering temperature, the solder paste thickness, and the soldering sequence of the solder joints), and the output of the parameter adjustment model can include at least one of the adjusted soldering temperature, the solder paste thickness, and the soldering sequence of the solder joints.

[0053] In some embodiments, the parameter adjustment model can include, but is not limited to, a neural network (NN), a convolutional neural network (CNN), a deep neural network (DNN), a recurrent neural network (RNN), or the like, or any combination thereof, for example, the parameter adjustment model can be a model formed by a combination of a convolutional neural network and a deep neural network.

[0054] At step 240, if it is determined that the welding process is normal, the point cloud data of the solder joints of the circuit to be inspected is acquired. In some embodiments, step 240 can be performed by a point cloud acquisition module.

[0055] In some embodiments, the point cloud acquisition module can acquire the point cloud information of the circuit to be inspected by a point cloud acquisition device.

[0056] In some embodiments, the point cloud acquisition device includes a shooting pose adjustment apparatus, a structured light binocular vision apparatus, and a data processor, the structured light binocular vision apparatus is arranged on the shooting pose adjustment apparatus, the shooting pose adjustment apparatus is used to adjust the pose of the structured light binocular vision apparatus, the structured light binocular vision apparatus is used to shoot the circuit board to be inspected to acquire the left-eye image and the right-eye image of the circuit board to be inspected.

[0057] In some embodiments, the data processor is used to generate a target pose for the circuit diagram of the circuit to be inspected, which is marked with the solder joints, and control the shooting pose adjustment apparatus to adjust the pose of the shooting pose adjustment apparatus to the target pose, and control the structured light binocular vision apparatus to acquire the left-eye image and the right-eye image of the circuit board to be inspected at the target pose.

[0058] In some embodiments, the data processor is further used to generate the point cloud information of the circuit to be inspected based on the left-eye image and the right-eye image of the circuit board to be inspected acquired by the structured light binocular vision apparatus.

[0059] For example, the data processor can generate the point cloud information of the circuit board to be inspected by reconstructing the left-eye image and the right-eye image of the circuit board to be inspected through a reconstruction algorithm. In some embodiments, the reconstruction algorithm includes a PMVS (the patch-based MVSalgorithm) algorithm, a MC (Marching Cube) algorithm, a DC (Dual Contouring) algorithm, etc.

[0060] In some embodiments, the point cloud acquisition module can perform a separation operation on the point cloud information of the circuit to be inspected to obtain the point cloud data of the solder joint of the circuit to be inspected.

[0061] For example, the point cloud acquisition module can first determine a separation plane based on the height information of the point cloud, i.e., the plane on which the upper surface of the circuit to be inspected is located, and take the point cloud on the separation plane as the point cloud of the solder joint and take the point cloud under the separation plane as the point cloud of the circuit to be inspected. When there are multiple solder joints, the point cloud acquisition module can perform clustering on the separated point cloud based on the distance between two point clouds to generate multiple clustering clusters, each of which corresponds to the point cloud of a solder joint.

[0062] Step 250: judging whether the morphology of the solder joint is abnormal based on the point cloud data of the solder joint of the circuit to be inspected. In some embodiments, step 250 can be performed by a solder joint judgment module.

[0063] In some embodiments, the solder joint judgment module can obtain the contour shape of the solder joint based on the point cloud data of the solder joint of the circuit to be inspected, and judge whether the morphology of the solder joint is abnormal based on the contour shape of the solder joint.

[0064] In some embodiments, if it is judged that the morphology of the solder joint is abnormal, step 260 is performed, and if it is judged that the morphology of the solder joint is normal, step 270 is performed.

[0065] Step 260: if it is judged that the morphology of the solder joint is abnormal, determining the state of the welding equipment based on the historical welding data of the welding equipment. In some embodiments, step 260 can be performed by a state determination module.

[0066] The historical welding data can include the welding-related data and the point cloud data of multiple circuit boards welded by the welding equipment.

[0067] In some embodiments, the state determination module can determine the state of the welding device based on historical welding data of the welding device by a state judgment model. The input of the state judgment model can include the historical welding data of the welding device, and the output of the state judgment model can include the state of the welding device. For example, the input of the state judgment model can include welding-related data of a plurality of welding tasks completed by the welding device (first vibration data of the welding device, second vibration data of the circuit board to be inspected, temperature information and environmental smoke concentration information of at least one position of the circuit board to be inspected) and point cloud data of the corresponding circuit board, and the output of the state judgment model can include the state of the welding device.

[0068] In some embodiments, the state judgment model can include, but is not limited to, a neural network (NN), a convolutional neural network (CNN), a deep neural network (DNN), a recurrent neural network (RNN), or any combination thereof, for example, the state judgment model can be a model formed by combining a convolutional neural network and a deep neural network.

[0069] If it is determined that the state of the welding device is a fault state, the welding device can be repaired or replaced.

[0070] In step 270, if it is determined that the morphology of the welding spot is normal, the welding spot of the circuit to be inspected is subjected to a circuit test by an electrical detection device to obtain test data. In some embodiments, step 270 can be performed by an electrical test module.

[0071] In some embodiments, for each welding spot, the electrical test module can determine at least one pair of detection points of the welding spot on the circuit board to be inspected based on design information of the circuit board to be inspected by a welding spot determination model; and obtain test data based on the at least one pair of detection points by an electrical detection device.

[0072] The design information of the circuit board to be inspected can include circuit schematic diagram, PCB diagram, and other information of the circuit board to be inspected.

[0073] A pair of detection points refers to the positive and negative electrodes for electrical detection.

[0074] The test data can include current, voltage, and other information.

[0075] In some embodiments, the welding spot determination model can include, but is not limited to, a neural network (NN), a convolutional neural network (CNN), a deep neural network (DNN), a recurrent neural network (RNN), or any combination thereof, for example, the welding spot determination model can be a model formed by combining a convolutional neural network and a deep neural network.

[0076] In step 280, the state of the welding spot of the circuit board to be inspected is determined based on the test data. In some embodiments, step 280 can be performed by a welding spot detection module.

[0077] In some embodiments, the solder joint detection module can determine the state of each solder joint of the circuit board to be inspected based on the design information and the test data of the circuit board to be inspected by the state determination model, and output a test report, wherein the test report comprises a circuit diagram corresponding to the circuit board to be inspected, and the circuit diagram is marked with the faulty solder joints and the state of each faulty solder joint.

[0078] In some embodiments, the state determination model can include, but is not limited to, a neural network (NN), a convolutional neural network (CNN), a deep neural network (DNN), a recurrent neural network (RNN), etc., or any combination thereof, for example, the state determination model can be a model formed by combining a convolutional neural network and a deep neural network.

[0079] The above has described the basic concept, and it is obvious that the above detailed disclosure is only used as an example and does not limit the present specification for those skilled in the art. Although it is not explicitly stated here, 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, so such modifications, improvements and corrections still belong to the spirit and scope of the exemplary embodiments of the present specification.

[0080] At the same time, specific words are used in the present specification to describe the embodiments of the present specification. As "one embodiment", "an embodiment", and / or "some embodiments" means a certain feature, structure or characteristic related to at least one embodiment of the present specification. Therefore, it should be emphasized and noted that the "an embodiment" or "one embodiment" or "one alternative embodiment" mentioned in different places in the present specification does not necessarily refer to the same embodiment. In addition, some features, structures or characteristics in one or more embodiments of the present specification can be properly combined.

[0081] In addition, unless the claim explicitly states otherwise, the order of the processing elements and sequences described in the present specification, the use of numerals and letters, or the use of other names, is not intended to limit the order of the processes and methods of the present specification. Although some currently considered useful embodiments of the invention are discussed in the above disclosure through various examples, it should be understood that such details are only for the purpose of illustration, and the additional claims are not limited to the disclosed embodiments, on the contrary, the claims are intended to cover all modifications and equivalent combinations that meet the spirit and scope of the embodiments of the present specification. For example, although the system components described above can be implemented by hardware devices, they can also be implemented only by software solutions, such as installing the described system on existing servers or mobile devices.

[0082] It should be appreciated that, for simplicity of illustration, the foregoing methodology can be illustrated or described in terms such as the following: in one embodiment, another embodiment, an implementation, or variations thereof, etc. Although described in one or more implementations, these teachings are hereby applied to any and all such variations. Where the description defines implementations by using such terms as "in one embodiment," "in another embodiment," or "in an implementation", such terminology is used for descriptive purposes only, and does not delimit the subject matter described. It is recognized that "in one embodiment" or "in an implementation" can sometimes appear to be repetitive herein. However, such language is not intended to cause a limitation of the number of embodiments or implementations which can be claimed. A description or definition of an embodiment or implementation as that used herein should not be construed to mean that the described embodiment or implementation is the only embodiment or implementation. Other embodiments or implementations that are not explicitly described or defined can be considered to be within the scope of the subject matter described.

[0083] Finally, it should be appreciated that the foregoing description is by way of example only, and is not intended to limit the scope of the subject matter described herein. Other variations that are within the scope of the described subject matter can occur to those in the art upon reading or understanding the previous description. Accordingly, the embodiments described herein are intended to embrace all such variations.

Claims

1. A product quality online monitoring method applied to intelligent manufacturing, characterized in that, The method comprises the following steps: acquiring welding-related data in the process of welding a to-be-inspected circuit board by a welding device; judging whether an abnormality occurs in the welding process based on the welding-related data; if it is judged that an abnormality occurs in the welding process, adjusting welding parameters, and the welding device welding a circuit board of the same model as the to-be-inspected circuit board based on the adjusted welding parameters; if it is judged that the welding process is normal, acquiring point cloud data of a solder joint of the to-be-inspected circuit; judging whether the shape of the solder joint is abnormal based on the point cloud data of the solder joint of the to-be-inspected circuit; if it is judged that the shape of the solder joint is abnormal, determining the state of the welding device based on historical welding data of the welding device; if it is judged that the shape of the solder joint is normal, performing a circuit test on the solder joint of the to-be-inspected circuit by an electrical detection device to acquire test data; judging the state of the solder joint of the to-be-inspected circuit based on the test data. 2.The product quality online monitoring method applied to intelligent manufacturing according to claim 1, wherein, The acquiring of the welding-related data in the process of welding a to-be-inspected circuit board by a welding device comprises the following steps: acquiring first vibration data of the welding device, second vibration data of the to-be-inspected circuit board, temperature information of at least one position of the to-be-inspected circuit board, and environmental smoke concentration information at multiple time points in the welding process; generating a welding-related sequence, wherein one element of the welding-related sequence corresponds to the first vibration data of the welding device, the second vibration data of the to-be-inspected circuit board, the temperature information of at least one position of the to-be-inspected circuit board, and the environmental smoke concentration information at a time point. 3.The product quality online monitoring method applied to intelligent manufacturing according to claim 2, characterized in that, The acquiring of the temperature information of at least one position of the to-be-inspected circuit board in the welding process comprises the following steps: acquiring the temperature information of at least one position of the to-be-inspected circuit board by an infrared thermal imager.

4. The product quality online monitoring method applied to intelligent manufacturing according to claims 1-3, characterized in that, The adjusting of the welding parameters comprises the following steps: adjusting the welding parameters based on the welding-related data by a parameter adjustment model, wherein the welding parameters comprise at least one of a welding temperature, a solder paste thickness, and a welding sequence of a solder joint.

5. The product quality online monitoring method applied to intelligent manufacturing according to claims 1-3, characterized in that, The acquiring of the point cloud data of the solder joint of the to-be-inspected circuit comprises the following steps: acquiring point cloud information of the to-be-inspected circuit by the point cloud acquisition device; performing a separation operation on the point cloud information of the to-be-inspected circuit to acquire the point cloud data of the solder joint of the to-be-inspected circuit.

6. The product quality online monitoring method applied to intelligent manufacturing according to claim 5, characterized in that, The judging of whether the shape of the solder joint is abnormal based on the point cloud data of the solder joint of the to-be-inspected circuit comprises the following steps: acquiring a contour shape of the solder joint based on the point cloud data of the solder joint of the to-be-inspected circuit; judging whether the shape of the solder joint is abnormal based on the contour shape of the solder joint.

7. The product quality online monitoring method applied to intelligent manufacturing according to claim 5, characterized in that, The point cloud acquisition device comprises a shooting pose adjustment device, a structured light binocular vision device, and a data processor, the structured light binocular vision device is arranged on the shooting pose adjustment device, the shooting pose adjustment device is used for adjusting the pose of the structured light binocular vision device, and the structured light binocular vision device is used for shooting the to-be-inspected circuit board to acquire a left-eye image and a right-eye image of the to-be-inspected circuit board. The data processor is configured to generate a target pose of the circuit diagram with the solder joint marked corresponding to the circuit to be inspected, and control the shooting pose adjusting device to adjust the pose of the shooting pose adjusting device to the target pose; The data processor is further configured to control the structured light binocular vision device to acquire a left-eye image and a right-eye image of the circuit board to be inspected in the target pose; The data processor is further configured to generate point cloud information of the circuit to be inspected based on the left-eye image and the right-eye image of the circuit board to be inspected acquired by the structured light binocular vision device.

8. The product quality online monitoring method applied to intelligent manufacturing according to claims 1-3, characterized in that, The circuit test on the solder joint of the circuit to be inspected by the electrical detection equipment includes: For each solder joint, A solder joint determination model determines at least one pair of detection points of the solder joint on the circuit board to be inspected based on design information of the circuit board to be inspected; The electrical detection equipment acquires the test data based on the at least one pair of detection points.

9. The product quality online monitoring method applied to intelligent manufacturing according to claims 1-3, characterized in that, The state of the solder joint of the circuit board to be inspected is determined based on the test data, including: A state determination model determines the state of each solder joint of the circuit board to be inspected based on the design information of the circuit board to be inspected and the test data, and outputs a test report, wherein the test report includes a circuit diagram of the circuit board to be inspected corresponding to the solder joint marked with a faulty solder joint and the state of each faulty solder joint.

10. A product quality online monitoring system applied to intelligent manufacturing, characterized in that, It includes: A data acquisition module is configured to acquire welding-related data during welding of a welding device on a circuit board to be inspected; An abnormality judgment module is configured to determine whether an abnormality occurs in the welding process based on the welding-related data; A parameter adjustment module is configured to adjust the welding parameters if the abnormality judgment module determines that an abnormality occurs in the welding process, and the welding device welds circuit boards of the same model as the circuit board to be inspected based on the adjusted welding parameters; A point cloud acquisition module is configured to acquire point cloud data of the solder joint of the circuit to be inspected if the abnormality judgment module determines that the welding process is normal; A solder joint determination module is configured to determine whether the shape of the solder joint is abnormal based on the point cloud data of the solder joint of the circuit to be inspected; A state determination module is configured to determine the state of the welding device based on historical welding data of the welding device if the solder joint determination module determines that the shape of the solder joint is abnormal; An electrical test module is configured to acquire test data by performing circuit test on the solder joint of the circuit to be inspected by an electrical detection equipment if the solder joint determination module determines that the shape of the solder joint is normal; A solder joint detection module is configured to determine the state of the solder joint of the circuit board to be inspected based on the test data.

Images