CN116611850A - System for detecting and tracing engine assembly quality curve - Google Patents
System for detecting and tracing engine assembly quality curve Download PDFInfo
- Publication number
- CN116611850A CN116611850A CN202310873212.XA CN202310873212A CN116611850A CN 116611850 A CN116611850 A CN 116611850A CN 202310873212 A CN202310873212 A CN 202310873212A CN 116611850 A CN116611850 A CN 116611850A
- Authority
- CN
- China
- Prior art keywords
- assembly
- attribute information
- engine
- parts
- assembly quality
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000001514 detection method Methods 0.000 claims abstract description 16
- 238000012937 correction Methods 0.000 claims abstract description 6
- 238000000034 method Methods 0.000 claims description 139
- 238000012549 training Methods 0.000 claims description 22
- 238000013145 classification model Methods 0.000 claims description 19
- 230000007547 defect Effects 0.000 claims description 16
- 238000004364 calculation method Methods 0.000 claims description 15
- 238000000605 extraction Methods 0.000 claims description 7
- 230000015654 memory Effects 0.000 claims description 4
- 238000012216 screening Methods 0.000 claims description 4
- 238000002360 preparation method Methods 0.000 claims description 3
- 238000004519 manufacturing process Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 6
- 238000003860 storage Methods 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 4
- 238000003754 machining Methods 0.000 description 4
- 238000003062 neural network model Methods 0.000 description 4
- 210000002569 neuron Anatomy 0.000 description 3
- 230000003044 adaptive effect Effects 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000007792 addition Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000007787 long-term memory Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000012797 qualification Methods 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/018—Certifying business or products
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06395—Quality analysis or management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/04—Manufacturing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/22—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
- G06V10/765—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects using rules for classification or partitioning the feature space
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Multimedia (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Software Systems (AREA)
- Tourism & Hospitality (AREA)
- Medical Informatics (AREA)
- Evolutionary Computation (AREA)
- Educational Administration (AREA)
- Databases & Information Systems (AREA)
- Computing Systems (AREA)
- Game Theory and Decision Science (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Primary Health Care (AREA)
- Operations Research (AREA)
- Manufacturing & Machinery (AREA)
- Quality & Reliability (AREA)
- General Factory Administration (AREA)
Abstract
The application discloses a detection and traceability system of an engine assembly quality curve, which comprises the following components: the identification module is used for collecting assembly quality curves of a plurality of parts of the engine and identifying first attribute information and second attribute information of the corresponding parts according to the assembly quality curves, wherein the first attribute information at least comprises quality attributes of the parts, and the second attribute information at least comprises attributes of the parts and other parts matched with each other; the control module is capable of communicating with the identification module and is used for acquiring the first attribute information and the second attribute information, generating correction information of the part according to the first attribute information and generating coordination adjustment information of the part and other parts according to the second attribute information. The application can ensure the consistency and traceability of the assembly of key parts of the engine, ensure the consistency of products such as the engine and the like and facilitate the subsequent assembly quality improvement.
Description
Technical Field
The application relates to the technical field of engine assembly, in particular to a detection and traceability system of an engine assembly quality curve.
Background
The engine product is generally provided with a plurality of assembling stations, and has higher requirements and effects on assembling data, and particularly, the quality of key parts of the engine and the assembling process have more obvious influence on the consistency of the engine. In the assembly process of the engine, as a large number of parts are involved in the machining process, a part of the quality of most parts is controlled by the common negotiation of the quality level of suppliers and companies, and the other part is controlled by the machining process inside the companies, although all parts and the inspection standards of the machining process are qualified, the final consistency of the product is difficult to ensure, and the existing production system and equipment cannot effectively trace the assembly process, so that the continuous improvement of the subsequent assembly quality is not facilitated.
Disclosure of Invention
The application aims to provide a detection and traceability system for an engine assembly quality curve, which can ensure the consistency and traceability of the assembly of key parts of an engine, ensure the consistency of products and facilitate the improvement of the subsequent assembly quality.
Based on the above purpose, the application provides a detection and tracing system of an engine assembly quality curve, which comprises an identification module and a control module; the identification module is used for collecting assembly quality curves of a plurality of parts of the engine and identifying first attribute information and second attribute information of the corresponding parts according to the assembly quality curves, wherein the first attribute information at least comprises quality attributes of the parts, and the second attribute information at least comprises attributes of the parts and other parts matched with each other; the control module is capable of communicating with the identification module and is used for acquiring the first attribute information and the second attribute information, generating correction information of the part according to the first attribute information and generating coordination adjustment information of the part and other parts according to the second attribute information.
Further, the identification module is further configured to identify third attribute information of the corresponding part according to the assembly quality curve, where the third attribute information at least includes a process assembly attribute of the part, and the control module obtains the third attribute information and generates process adjustment information of the part according to the third attribute information.
Further, the system also comprises a detection module, the detection module performs detection screening on the assembly quality curve to obtain a qualified assembly quality curve, and the identification module identifies the first attribute information, the second attribute information and the third attribute information of the part according to the qualified assembly quality curve.
Further, the identification module includes:
the image recognition unit is used for carrying out image recognition on the assembly quality curve and determining an effective area of the assembly quality curve;
the key point identification unit is used for identifying the effective area and acquiring key points of the assembly quality curve, wherein the key points are at least one of the following: an assembly preparation starting point, an assembly ending point, an assembly starting point, a disassembly point, a secondary assembly starting point, a yield strength limit, a plastic region, an elastic region, a torque and a slope;
the feature extraction unit is used for extracting features of the key points to obtain feature values of the parts, wherein the feature values are at least one of the following: upper force angle, upper force torque, maximum angle, maximum torque, torque area integral, assembly slope, peak;
the model prediction unit is used for inputting the characteristic values into a preset classification model and outputting first attribute information, second attribute information and third attribute information of the part.
Further, the system further comprises a model training unit, wherein the model training unit constructs a training set by using a plurality of sample characteristic values, and trains the training set to obtain a classification model, and the classification model comprises an input layer, two hidden layers and an output layer.
Further, the process assembly attribute includes a plurality of process assembly types, the model prediction unit is capable of outputting the plurality of process assembly types of the part and an accuracy value corresponding to each process assembly type, determining the process assembly type corresponding to the maximum accuracy value as a final process assembly type of the part, determining the maximum accuracy value as an accuracy value of the final process assembly type, and obtaining a loss value of the final process assembly type, wherein the process assembly type is at least one of: overshoot, unhooking, misplacement, normal assembly.
Further, the identification module further comprises a defect unit, which is used for determining a corresponding defect type according to the final process assembly type, and determining a weight corresponding to the final process assembly type according to the defect type.
Further, the identification module further comprises an output unit, wherein the output unit is used for carrying out weighted average calculation according to the precision value and the weight of the final process assembly type of each part, determining the consistency parameter of the engine in the dimension direction of the precision value of the process assembly type, carrying out weighted average calculation according to the loss value and the weight of the final process assembly type of each part, and determining the consistency parameter of the engine in the dimension direction of the loss value of the process assembly type.
Further, the output unit performs weighted average calculation according to the positive deviation of each part and the weight of the final process assembly type of the corresponding part, determines the consistency parameter of the engine in the positive deviation dimension direction of the part, performs weighted average calculation according to the negative deviation of each part and the weight of the final process assembly type of the corresponding part, determines the consistency parameter of the engine in the negative deviation dimension direction of the part, and performs weighted average calculation according to the actual deviation of each part and the weight of the final process assembly type of the corresponding part, and determines the consistency parameter of the engine in the actual deviation dimension direction of the part.
Further, the system also comprises a storage module, the identification module acquires the engine model of the engine, and the storage module establishes and stores the corresponding relation between the engine model and the assembly quality curve of the plurality of parts.
The application can trace back the assembly quality curve data of the engine and other products in the assembly process, can continuously adjust and control each part in the assembly process of the engine from a plurality of dimensions such as the matching of the parts, and the like, can ensure the consistency and traceability of the assembly of key parts of the engine, can ensure the consistency of the engine and other products, and is convenient for the subsequent assembly quality improvement.
Drawings
FIG. 1 is a first system block diagram of an engine assembly quality curve detection and tracing system provided in accordance with the present application;
FIG. 2 is a second system block diagram of a system for detecting and tracing an engine assembly quality curve provided in accordance with the present application;
FIG. 3 is a schematic illustration of a standard assembly quality curve provided in accordance with the present application;
FIG. 4 is a third system block diagram of an engine assembly quality curve detection and tracing system provided in accordance with the present application;
FIG. 5 is a fourth system block diagram of a system for detecting and tracing an engine assembly quality curve provided in accordance with the present application;
FIG. 6 is a schematic diagram of a neural network classification model provided in accordance with the present application;
fig. 7 is a flowchart of a method for detecting and tracing an engine assembly quality curve according to the present application.
Detailed Description
The present application will be described in detail below with reference to the specific embodiments shown in the drawings, but these embodiments are not limited to the present application, and structural, method, or functional modifications made by those skilled in the art based on these embodiments are included in the scope of the present application.
Referring to fig. 1, an embodiment of the present application provides a system 100 for detecting and tracing an engine assembly quality curve, where the system 100 provided in the embodiment of the present application includes an identification module 11 and a control module 12. The identification module 11 collects assembly quality curves of a plurality of parts of the engine, and can identify first attribute information and second attribute information of corresponding parts according to the assembly quality curves, wherein the first attribute information at least comprises quality attributes of the parts, and the second attribute information at least comprises attributes of the parts and other parts matched with each other. The control module 12 is capable of communicating with the identification module 11 for acquiring the first attribute information and the second attribute information. The control module 12 generates correction information of the part according to the first attribute information, and generates coordination adjustment information of the part and other parts according to the second attribute information. The assembly quality curves of the various critical parts of the engine production line are collected by connection to assembly equipment or by integration of the manufacturing execution system. The first attribute information may be understood as quality problems caused by the part itself, such as the hardness or strength of the part not reaching a value. The correction information of the part is understood to mean that the part is adjusted or replaced, etc., based on quality problems caused by the part itself. The second attribute information may be understood as quality problems caused by the mutual matching between the two parts, for example, the matching degree of the part a and the part B does not reach the qualification index. Matching adjustment information can be understood as an adaptive adjustment of the degree of matching between two parts. For the whole car manufacturing industry, the assembly quality of the assembly equipment production line plays a very important role in the whole car assembly quality, especially for key procedures of high-end finish machining assembly, such as: the requirements and the roles of the assembly data of the engine and the gearbox of the whole vehicle are more obvious. On an engine assembly production line, two attributes of the parts, namely the quality attributes of the parts and the matching attributes among the parts, are identified through an artificial intelligence platform TensorFlow, the parts are correspondingly adjusted according to the two attributes, namely the parts are corrected if the quality attributes of the parts are the quality attributes of the parts, for example, if the hardness of the parts is unqualified, the parts on the assembly production line can be replaced; if the matching attribute of the parts and other parts is that the matching between the parts is adjusted, for example, the matching degree between the parts does not reach the qualified index, the matching degree between the parts or the assembling station between the parts can be adjusted, so that the matching degree between the parts reaches the index requirement, and the assembling of a plurality of parts of the engine is consistent. According to the method and the device, the assembly quality curve data in the engine assembling process can be traced, and each part in the part assembling process is continuously adjusted and controlled from the two dimensions of the part and the part, so that the assembly defects caused by the fact that the upper and lower limit floating value difference exists in the inspection standard of each part or part and the certain difference exists in the material of the part are reduced, the assembly of each part of the engine can be consistent, the consistency of the engine assembling process is ensured, and finally the consistency of products is achieved.
As an alternative implementation manner, the identification module 11 also identifies third attribute information of the corresponding part according to the assembly quality curve, where the third attribute information includes at least a process assembly attribute of the part. The control module 12 obtains the third attribute information and generates process adjustment information for the part based on the third attribute information to adjust the process set-up of the part. The third attribute information may be understood as quality problems caused by the process assembly process, such as overshooting, slipping, etc. The process adjustment information may be understood as an adaptive adjustment of the process assembly process of the part to meet the process assembly requirements of the part. Taking the hand-off as an example, the hand-off generally refers to that the actual torque of the part is low, so that the torque of the part needs to be adjusted, and the process assembly information of the part is adjusted to meet the torque assembly requirement of the part. By combining the above embodiments, three properties of the parts, namely, the quality properties of the parts themselves, the matching properties among the parts and the process assembly properties, are identified by the assembly quality curves of the parts, and the parts are correspondingly adjusted according to the properties, so that the parts in the assembly process of the engine can be continuously adjusted and controlled from the dimensions of the parts themselves, the matching of the parts, the process assembly and the like, the assembly of key parts of the engine can be consistent and traceable, the defect in the assembly process of the engine is further reduced, the assembly of the parts of the engine can be consistent, the consistency of the assembly process of the engine is further ensured, the consistency of products such as the engine is further satisfied, and the subsequent assembly quality improvement is facilitated.
As an alternative implementation, as shown in fig. 2, the system 100 further includes a detection module 13. The detection module 13 performs a preliminary screening of the assembly quality curve to obtain a qualified assembly quality curve. The identification module 11 identifies the first, second, and third attribute information of the part based on the qualified assembly quality curve. According to the method, the collected assembly quality curves are rapidly screened and detected, and the pictures and the assembly data are subjected to preliminary screening and judgment, so that the curves and the data with interference can be eliminated, the acquisition of the effective assembly quality curves is ensured, and the correctness of the follow-up picture identification results can be ensured. As shown in the standard assembly quality curve diagram of fig. 3, the assembly quality curve in the two-line range is a qualified quality assembly curve.
As an alternative implementation, as shown in fig. 4, the recognition module 11 includes an image recognition unit 111, a keypoint recognition unit 112, a feature extraction unit 113, and a model prediction unit 114. The image recognition unit 111 performs image recognition on the qualified assembly quality curve, and determines an effective area of the assembly quality curve. By identifying the effective area of the assembly quality curve and determining the positioning interval, the key point positions of the curve can be quickly identified. The key point identification unit 112 identifies the effective area of the assembly quality curve, and obtains key points of the assembly quality curve. Wherein the key point is at least one of the following: assembly preparation start point, assembly end point, assembly start point, disassembly point, secondary assembly start point, yield strength limit, plastic region, elastic region, torque, slope. The key point identification unit 112 performs key point detection on the positioning section of the curve, identifies key points of the curve, and provides data analysis for subsequent feature extraction. The key point identifying unit 112 also cuts the assembly quality curve with the key points identified, eliminates the abnormal curve with the key points not aligned, is convenient for subsequent curve feature extraction, and can further ensure the data correctness of the assembly quality curve. The feature extraction unit 113 performs feature extraction based on the key points of the assembly quality curve, and obtains a plurality of feature values corresponding to the obtained parts. The characteristic value is at least one of the following: upper force angle, upper force torque, maximum angle, maximum torque, torque area integral, assembly slope, peak. By processing the image of the assembly quality curve, positioning and extracting a plurality of characteristic values of the part, compared with the simple method of judging whether the assembly is in place by whether the torsion/angle is qualified or not, the method and the device analyze and judge whether the quality curve in the assembly process of the engine accords with the process installation requirement or not through the plurality of characteristic values, the analysis result is more accurate, powerful model input data is provided for the subsequent classification model, and the output result of the model is more accurate. The model prediction unit 114 inputs the feature values of the parts into a preset classification model, and outputs first, second, and third attribute information of the parts. Through artificial intelligence platform TensorFlow, utilize neural network model to train the eigenvalue of part, output the attribute category of part, according to the attribute category of part, go adjustment each part or part's station on the production line to can guarantee the uniformity in the product assembly process.
As an alternative implementation, as shown in fig. 5, the recognition module 11 further comprises a model training unit 115. The model training unit 115 constructs a training set with a plurality of sample feature values and trains the training set to obtain a classification model. Sample characteristic values include upper force angle, upper force torque, maximum angle, maximum torque, torque area integral, assembly slope, and peak value. And constructing a neural network according to the data of the assembly quality curve and the characteristic value, and finding the numerical value of the relation between the characteristic and the attribute by using the neural network. The neural network model includes an input layer, a hidden layer, and an output layer, each hidden layer containing a plurality of neurons. The fully connected neural network model employed by the present application, the neurons in each layer will take input connections from each neuron in the previous layer. A classification model is schematically shown in fig. 6, and includes an input layer, two hidden layers, and an output layer. And traversing each sample in the training set for each training set, acquiring characteristics of the samples, predicting according to the characteristics of the samples, comparing a prediction result with an attribute, measuring inaccuracy of the prediction result, and calculating loss and gradient of the model by using the obtained values. In the training and evaluation stage, a model optimization strategy and a loss function are set for the deviation of a prediction result and expected attributes of the model, and the variables of the model are updated until the model training is completed, so that the loss of the model is reduced, and the accuracy of the model is improved. The intensive neural network model is built through a platform TensorFlow, the training set is trained by deep machine learning, model variables are optimized and adjusted, and the accuracy of the model is improved, so that the accuracy of model output is guaranteed.
As an alternative implementation, different deep learning algorithms, such as convolutional neural networks, long and short term memory networks, recurrent neural networks, etc., may be implemented by the platform TensorFlow. The platform can support a plurality of machine learning algorithms, and the classification model is constructed through different algorithms.
As an alternative implementation manner, the process assembly attribute includes a plurality of process assembly types, the model prediction unit 114 inputs a plurality of feature values of the part into the classification model, outputs a plurality of process assembly types of the part and an accuracy value corresponding to each process assembly type, determines a process assembly type corresponding to a maximum accuracy value as a final process assembly type of the part, determines a maximum accuracy value as an accuracy value of the final process assembly type, and obtains a loss value of the final process assembly type. The process assembly type is at least one of the following: overshoot, out of hand, out of position, normal assembly, it should be noted that the process assembly types include, but are not limited to, the listed types. Based on the technical scheme, a plurality of characteristic values of each part are respectively input into a classification model, and the final process assembly type of each part, the precision value of the final process assembly type and the loss value of the final process assembly type are obtained through training and prediction of the model. In the classification model illustrated in fig. 4, characteristic values such as an upper force angle, an upper force torque, a maximum angle, a maximum torque, a torque area integral, an assembly slope, a peak value and the like of a part are taken as inputs of the model, and a plurality of process assembly types of the part and an accuracy value corresponding to each process assembly type are output at an output layer of the model through training and prediction of the model. In this embodiment, the classification model outputs 4 prediction results of the part, the process assembly type of the prediction result 1 is overshoot, the corresponding precision value is 0.02, the process assembly type of the prediction result 2 is out of hand, the corresponding precision value is 0.96, the process assembly type of the prediction result 3 is other conditions, the corresponding precision value is 0.01, the process assembly type of the prediction result 4 is normal, the corresponding precision value is 0.01, and the final process assembly type of the part is out of hand, and the corresponding precision value is 0.96 through judgment of the precision value. Meanwhile, the loss value of the final process assembly type which is the hands-off can be obtained through the classification model. Through continuous training of training set data for a long time, a classification model of a mature and stable assembly quality curve can be obtained, so that quality in an engine assembly process can be accurately analyzed, part states in the assembly process are controlled, data analysis of the assembly quality curve is simplified, analysis and judgment of the engine assembly curve are accelerated, and powerful evidence is provided for consistency of the product assembly process.
As an alternative implementation, as shown in fig. 5, the identification module 11 further includes a defect unit 116. The defect unit 116 determines a corresponding defect type according to a final process assembly type of the part, and determines a weight corresponding to the final process assembly type according to the defect type. Taking the above embodiment as an example for illustration, the final process assembly type of the part is a slipping, the defect type corresponding to the slipping is serious, and the weight corresponding to the slipping is set to be a% according to the serious level. The defect type is determined by the process assembly type, and the defect of the assembly quality can be analyzed later.
As an alternative implementation, as shown in fig. 5, the identification module 11 further comprises an output unit 117. The output unit 117 performs weighted average calculation according to the precision value of the final process assembly type of each part and the corresponding weight, and determines the consistency parameter of the engine in the dimension direction of the precision value of the process assembly type according to the calculated average. The output unit 117 also performs weighted average calculation according to the loss value of the final process assembly type of each part and the corresponding weight, and determines a consistency parameter of the engine in the dimension direction of the loss value of the process assembly type according to the calculated average. According to the technical scheme, in the aspect of process assembly, the consistency of process assembly of each part in the process of engine assembly is measured from the two dimension directions of accuracy and loss degree by analyzing the consistency parameter of the engine in the dimension direction of the accuracy value of the process assembly type and the consistency parameter of the engine in the dimension direction of the loss value of the process assembly type, so that powerful evidence is provided for subsequent analysis of product consistency, and process assembly information in the process of engine assembly can be adjusted according to the consistency parameter, so that the product consistency is further ensured.
As an alternative implementation manner, the output unit 117 performs weighted average calculation according to the positive deviation of each part and the weight of the final process assembly type of the corresponding part, determines the consistency parameter of the engine in the positive deviation dimension direction of the part, performs weighted average calculation according to the negative deviation of each part and the weight of the final process assembly type of the corresponding part, determines the consistency parameter of the engine in the negative deviation dimension direction of the part, and performs weighted average calculation according to the actual deviation of each part and the weight of the final process assembly type of the corresponding part, and determines the consistency parameter of the engine in the actual deviation dimension direction of the part. The positive deviation, the negative deviation and the actual deviation of the part can be obtained from the process assembly information of the part. The consistency of the process assembly of each part in the process assembly process of the engine is measured based on the dimension direction of the process assembly of the part per se by analyzing the consistency parameter of the engine in the positive deviation dimension direction of the part, the consistency parameter of the negative deviation dimension direction of the part and the consistency parameter of the actual deviation dimension direction of the part, so that powerful evidence is provided for the subsequent analysis of the product consistency, and the process assembly information in the process of the engine assembly process can be adjusted according to the consistency parameter so as to further ensure the product consistency.
As an alternative implementation, as shown in fig. 2, the system 100 further includes a storage module 14. The identification module 11 also obtains an engine model of the engine. The memory module 14 establishes and stores correspondence of engine model numbers to assembly quality curves of a plurality of parts. In the related art, the field device has limited amount of stored assembly data, and generally can only be stored for a limited time, and when the time is too long, the assembly data of the engine is easy to lose, so that the engine cannot be traced. Therefore, the assembly quality curve is stored through the artificial intelligent platform TensorFlow, so that the method is used for data and processing of the subsequent assembly quality curve, on the one hand, long-term effective and reliable data related to engine assembly can be guaranteed, traceability of the assembly quality curve data is guaranteed, and when quality problems occur after future products are put into the market, the method can be used for checking and analyzing based on the engine assembly quality data in the assembly process.
As an alternative implementation, the storage module 14 stores the final process assembly type of each part, and the plurality of characteristic values corresponding to the part. For example, the storage module 14 stores the final process assembly type of the part serial number 1 as an overshoot, the corresponding upper force angle characteristic value is a001, the corresponding upper force margin characteristic value is B001, the corresponding maximum angle characteristic value is C001, the corresponding maximum torque characteristic value is D001, and the like, so that the final process assembly type of each part and the corresponding multiple characteristic values can be stored, traceability of assembly quality curve data is ensured, and engine assembly quality data in the assembly process can be directly checked.
As an alternative implementation manner, the storage module 14 stores the defect type, the weight, the loss value and the precision value corresponding to the final process assembly type of each part, stores the positive deviation, the negative deviation and the actual deviation value of each part, stores the consistency parameter of the engine in the precision value dimension direction of the process assembly type and the consistency parameter of the engine in the loss value dimension direction of the process assembly type, stores the consistency parameter of the engine in the positive deviation dimension direction of the part, the consistency parameter of the part in the negative deviation dimension direction of the part and the consistency parameter of the part in the actual deviation dimension direction, and can directly check the running result of the classification model, thereby further ensuring the traceability of the assembly quality curve data and directly checking the assembly quality data in the assembly process of the engine.
In order to further describe the system 100 for detecting and tracing the engine assembly quality curve provided by the embodiment of the present application, the method for detecting and tracing the engine assembly quality curve provided by the embodiment of the present application will be further described below.
As shown in fig. 7, as an alternative implementation manner, the method provided by the embodiment of the present application includes:
s701, collecting assembly quality curves of a plurality of parts of an engine, and identifying first attribute information and second attribute information of corresponding parts according to the assembly quality curves, wherein the first attribute information at least comprises quality attributes of the parts, and the second attribute information at least comprises attributes of the parts and other parts matched with each other;
s702, acquiring the first attribute information and the second attribute information, generating correction information of the part according to the first attribute information, and generating coordination adjustment information of the part and other parts according to the second attribute information.
As an optional implementation manner, the control method provided by the embodiment of the present application further includes: and identifying third attribute information of the corresponding part according to the assembly quality curve, wherein the third attribute information at least comprises the process assembly attribute of the part, acquiring the third attribute information, and generating process adjustment information of the part according to the third attribute information.
Although the preferred embodiments of the present application have been disclosed for illustrative purposes, those skilled in the art will appreciate that various modifications, additions and substitutions are possible, without departing from the scope and spirit of the application as disclosed in the accompanying claims.
Claims (10)
1. A system for detecting and tracing an engine assembly quality curve, the system comprising:
the identification module is used for collecting assembly quality curves of a plurality of parts of the engine and identifying first attribute information and second attribute information of corresponding parts according to the assembly quality curves, wherein the first attribute information at least comprises quality attributes of the parts, and the second attribute information at least comprises attributes matched with other parts;
the control module is capable of communicating with the identification module and is used for acquiring the first attribute information and the second attribute information, generating correction information of the part according to the first attribute information and generating coordination adjustment information of the part and other parts according to the second attribute information.
2. The system of claim 1, wherein the identification module is further configured to identify third attribute information of a corresponding part according to the assembly quality curve, the third attribute information including at least a process assembly attribute of the part, and the control module obtains the third attribute information and generates process adjustment information of the part according to the third attribute information.
3. The system for detecting and tracing an engine assembly quality curve according to claim 2, further comprising a detection module, wherein said detection module performs a detection screening of said assembly quality curve to obtain a qualified assembly quality curve, and wherein said identification module identifies said first attribute information, said second attribute information, and said third attribute information of said part based on said qualified assembly quality curve.
4. The engine assembly quality curve detection and tracing system of claim 2, wherein said identification module comprises:
the image recognition unit is used for carrying out image recognition on the assembly quality curve and determining an effective area of the assembly quality curve;
the key point identification unit is used for identifying the effective area and acquiring key points of the assembly quality curve, wherein the key points are at least one of the following: an assembly preparation starting point, an assembly ending point, an assembly starting point, a disassembly point, a secondary assembly starting point, a yield strength limit, a plastic region, an elastic region, a torque and a slope;
the feature extraction unit is used for extracting features of the key points to obtain feature values of the parts, wherein the feature values are at least one of the following: upper force angle, upper force torque, maximum angle, maximum torque, torque area integral, assembly slope, peak;
and the model prediction unit is used for inputting the characteristic value into a preset classification model and outputting the first attribute information, the second attribute information and the third attribute information of the part.
5. The system for detecting and tracing an engine assembly quality curve according to claim 4, wherein said identification module further comprises a model training unit, said model training unit constructing a training set from a plurality of sample feature values and training said training set to obtain said classification model, wherein said classification model comprises an input layer, two hidden layers, and an output layer.
6. The system for detecting and tracing an engine assembly quality curve according to claim 4, wherein said process assembly attributes comprise a plurality of process assembly types, said model prediction unit is capable of outputting a plurality of process assembly types of said part and an accuracy value corresponding to each process assembly type, determining a process assembly type corresponding to a maximum accuracy value as a final process assembly type of said part, determining said maximum accuracy value as an accuracy value of said final process assembly type, and obtaining a loss value of said final process assembly type, wherein said process assembly type is at least one of: overshoot, unhooking, misplacement, normal assembly.
7. The system for detecting and tracing an engine assembly quality curve according to claim 6, wherein said identification module further comprises a defect unit for determining a corresponding defect type based on said final process assembly type and determining a weight corresponding to said final process assembly type based on said defect type.
8. The system of claim 7, wherein the identification module further comprises an output unit for performing a weighted average calculation based on the precision value and weight of the final process assembly type for each part, determining a consistency parameter of the engine in a process assembly type precision dimension, and performing a weighted average calculation based on the loss value and weight of the final process assembly type for each part, determining a consistency parameter of the engine in a process assembly type loss dimension.
9. The system of claim 8, wherein the output unit performs a weighted average calculation based on the positive bias of each part and the weight of the final process assembly type for the corresponding part, determines a consistency parameter of the engine in the positive bias dimension of the part, performs a weighted average calculation based on the negative bias of each part and the weight of the final process assembly type for the corresponding part, determines a consistency parameter of the engine in the negative bias dimension of the part, and performs a weighted average calculation based on the actual bias of each part and the weight of the final process assembly type for the corresponding part, and determines a consistency parameter of the engine in the actual bias dimension of the part.
10. The system for detecting and tracing an engine assembly quality curve according to claim 1, further comprising a memory module, wherein said memory module obtains an engine model of said engine, and wherein said memory module establishes and stores a correspondence of said engine model to an assembly quality curve of said plurality of parts.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310873212.XA CN116611850B (en) | 2023-07-14 | 2023-07-14 | System for detecting and tracing engine assembly quality curve |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310873212.XA CN116611850B (en) | 2023-07-14 | 2023-07-14 | System for detecting and tracing engine assembly quality curve |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116611850A true CN116611850A (en) | 2023-08-18 |
| CN116611850B CN116611850B (en) | 2023-10-24 |
Family
ID=87675005
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310873212.XA Active CN116611850B (en) | 2023-07-14 | 2023-07-14 | System for detecting and tracing engine assembly quality curve |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116611850B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20170079953A (en) * | 2015-12-31 | 2017-07-10 | 대우조선해양 주식회사 | The dimensional expert control system |
| CN107451666A (en) * | 2017-07-15 | 2017-12-08 | 西安电子科技大学 | Breaker based on big data analysis assembles Tracing back of quality questions system and method |
| AU2020103881A4 (en) * | 2020-12-03 | 2021-02-11 | Zhimin Li | A Diagnosis Method of Train Body Assembly Deviation in the Manufacturing Process of High-Speed Train |
| CN112613118A (en) * | 2020-12-16 | 2021-04-06 | 中国科学院沈阳自动化研究所 | Digital twin modeling and tracing method for undetectable assembly quality inside rocket engine |
| CN113487533A (en) * | 2021-05-19 | 2021-10-08 | 东北大学 | Part assembly quality digital detection system and method based on machine learning |
| CN114581632A (en) * | 2021-12-07 | 2022-06-03 | 北京航天测控技术有限公司 | Method, equipment and device for detecting assembly error of part based on augmented reality technology |
| CN115293629A (en) * | 2022-08-19 | 2022-11-04 | 福赛轴承（嘉兴）有限公司 | Production and processing method and system for rolling bearing part |
| CN115311618A (en) * | 2022-07-29 | 2022-11-08 | 华中科技大学 | Assembly quality inspection method based on deep learning and object matching |
| CN116401785A (en) * | 2023-03-28 | 2023-07-07 | 广州中望龙腾软件股份有限公司 | MLP-Mixer-based assembly data management method and system |
-
2023
- 2023-07-14 CN CN202310873212.XA patent/CN116611850B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20170079953A (en) * | 2015-12-31 | 2017-07-10 | 대우조선해양 주식회사 | The dimensional expert control system |
| CN107451666A (en) * | 2017-07-15 | 2017-12-08 | 西安电子科技大学 | Breaker based on big data analysis assembles Tracing back of quality questions system and method |
| AU2020103881A4 (en) * | 2020-12-03 | 2021-02-11 | Zhimin Li | A Diagnosis Method of Train Body Assembly Deviation in the Manufacturing Process of High-Speed Train |
| CN112613118A (en) * | 2020-12-16 | 2021-04-06 | 中国科学院沈阳自动化研究所 | Digital twin modeling and tracing method for undetectable assembly quality inside rocket engine |
| CN113487533A (en) * | 2021-05-19 | 2021-10-08 | 东北大学 | Part assembly quality digital detection system and method based on machine learning |
| CN114581632A (en) * | 2021-12-07 | 2022-06-03 | 北京航天测控技术有限公司 | Method, equipment and device for detecting assembly error of part based on augmented reality technology |
| CN115311618A (en) * | 2022-07-29 | 2022-11-08 | 华中科技大学 | Assembly quality inspection method based on deep learning and object matching |
| CN115293629A (en) * | 2022-08-19 | 2022-11-04 | 福赛轴承（嘉兴）有限公司 | Production and processing method and system for rolling bearing part |
| CN116401785A (en) * | 2023-03-28 | 2023-07-07 | 广州中望龙腾软件股份有限公司 | MLP-Mixer-based assembly data management method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116611850B (en) | 2023-10-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111815572B (en) | Method for detecting welding quality of lithium battery based on convolutional neural network | |
| CN105335692A (en) | Tire X-ray image detection and identification method and system | |
| CN108711148B (en) | Tire defect intelligent detection method based on deep learning | |
| CN110288032B (en) | Vehicle driving track type detection method and device | |
| CN103324937A (en) | Method and device for labeling targets | |
| CN112487708A (en) | Resistance spot welding quality prediction method based on ensemble learning | |
| CN111815555A (en) | Metal additive manufacturing image detection method and device combining anti-neural network with local binary | |
| CN115294033A (en) | Tire belt layer difference level and misalignment defect detection method based on semantic segmentation network | |
| CN116224930A (en) | Processing control method and system for numerically controlled grinder product | |
| CN113902710A (en) | Method and system for detecting surface defects of industrial parts based on anomaly detection algorithm | |
| CN111753877A (en) | Product quality detection method based on deep neural network transfer learning | |
| CN115035328A (en) | Converter image increment automatic machine learning system and establishment training method thereof | |
| CN116611850B (en) | System for detecting and tracing engine assembly quality curve | |
| CN116881530B (en) | Device surface defect detection system based on deep learning | |
| CN114782362B (en) | Welding intelligent detection system based on deep migration learning | |
| CN116223529A (en) | Intelligent detection method and system for production of film-pressed gloves | |
| CN114912637B (en) | Human-computer object knowledge graph manufacturing production line operation and maintenance decision method and system and storage medium | |
| CN115106615A (en) | Welding deviation real-time detection method and system based on intelligent working condition identification | |
| CN115984215A (en) | Fiber bundle defect detection method based on twin network | |
| CN117911412B (en) | Dimension detection method and system for caterpillar track section for engineering machinery | |
| CN117399435B (en) | Self-adaptive compensation adjustment method and system for copper pipe rolling process | |
| CN117787480B (en) | Res-LSTM-based weld joint forming quality real-time prediction method | |
| CN117911412A (en) | Dimension detection method and system for caterpillar track section for engineering machinery | |
| CN117910118B (en) | Building structure safety evaluation method and system based on BIM model | |
| CN114813838A (en) | Method and system for detecting welding spot quality of vehicle body based on dynamic resistance signal |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |