CN116910619A - Ship welding defect recognition system based on artificial intelligence - Google Patents
Ship welding defect recognition system based on artificial intelligence Download PDFInfo
- Publication number
- CN116910619A CN116910619A CN202310874183.9A CN202310874183A CN116910619A CN 116910619 A CN116910619 A CN 116910619A CN 202310874183 A CN202310874183 A CN 202310874183A CN 116910619 A CN116910619 A CN 116910619A
- Authority
- CN
- China
- Prior art keywords
- welding
- defect
- data
- module
- welding defects
- 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.)
- Pending
Links
- 238000003466 welding Methods 0.000 title claims abstract description 258
- 230000007547 defect Effects 0.000 title claims abstract description 248
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 12
- 238000000605 extraction Methods 0.000 claims abstract description 27
- 238000000034 method Methods 0.000 claims description 36
- 230000008569 process Effects 0.000 claims description 28
- 238000003062 neural network model Methods 0.000 claims description 23
- 230000010354 integration Effects 0.000 claims description 8
- 239000002893 slag Substances 0.000 claims description 8
- 230000007704 transition Effects 0.000 claims description 8
- 238000012549 training Methods 0.000 claims description 7
- 238000012545 processing Methods 0.000 claims description 5
- 238000005336 cracking Methods 0.000 claims description 4
- 239000000284 extract Substances 0.000 claims description 3
- 230000004044 response Effects 0.000 abstract description 4
- 238000005457 optimization Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 230000004913 activation Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 239000002245 particle Substances 0.000 description 2
- 238000003908 quality control method Methods 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- HSFWRNGVRCDJHI-UHFFFAOYSA-N Acetylene Chemical compound C#C HSFWRNGVRCDJHI-UHFFFAOYSA-N 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 238000002922 simulated annealing Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- 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)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- General Factory Administration (AREA)
Abstract
The invention provides a ship welding defect identification system based on artificial intelligence, which is characterized by comprising a data acquisition module, a feature extraction module, a defect identification module and a defect prediction module; the device comprises a data acquisition module, a characteristic extraction module, a defect identification module and a defect prediction module, wherein the data acquisition module is used for acquiring multidimensional data related to ship welding defects, the characteristic extraction module is used for extracting dimension data with larger correlation with the welding defects in the multidimensional data, the defect identification module is used for identifying the ship welding defects, and the defect prediction module predicts the possible ship welding defects based on the identification result; the system not only can identify the real-time ship welding defects, but also can predict the ship welding defects possibly occurring in the future, thereby ensuring that a user can better evaluate the ship welding defects and make corresponding response measures according to the real-time identification result and the prediction result.
Description
Technical Field
The invention relates to the technical field of ship welding, in particular to a ship welding defect identification system based on artificial intelligence.
Background
Ship welding is a common connection technology in ship manufacturing, but various defects such as cracks, air holes, unfused and the like can be generated in the welding process; these welding defects have serious influence on the structural strength and safety of the ship, so that it is important to discover and accurately identify the welding defects of the ship as soon as possible; the development of artificial intelligence technology provides a new solution for identifying welding defects of ships; machine learning algorithms are capable of learning patterns and features from a large amount of data and intelligently identifying and classifying.
Referring to the related disclosed technical scheme, for example, CN115564249a discloses a ship welding quality control system and method, including a basic database, a monitoring module, a functional module and a functional processing module; according to the scheme, the state of a welding gun before welding, welding parameters and environment parameters in welding and flaw detection results after welding are subjected to full-flow quality control, the problems that in the prior art, element monitoring of welding quality is incomplete, the welding control system is not complete enough, a control action instruction is lack, a target is lack in welding quality improvement and the like are overcome, the controllability of welding parameters is comprehensively ensured, the welding construction quality is comprehensively improved, invalid welding task amount is timely restrained, and welding auxiliary lifting suggestions are accurately pushed; another exemplary prior art publication No. CN113537621a discloses a method for predicting welding quality of marine thin plates driven by big data, comprising the steps of: utilizing the established Internet of things to establish a frame, and collecting and uplink transmitting real-time welding data; performing key data feature optimization on the acquired welding quality influence factor data to finish data dimension reduction of an initial feature set in the welding big data; the BP neural network prediction model optimized by the self-adaptive simulated annealing particle swarm algorithm is established, and a ship sheet welding quality prediction result is output according to the data feature set after the data dimension reduction; the scheme establishes an SAPSO_BP prediction model, and overcomes the defects of easy sinking into local minimum points, low convergence speed, poor robustness and the like of the BP neural network prediction model, thereby realizing the on-line monitoring of the ship sheet welding process and the accurate prediction of the welding quality, improving the precision of a prediction result and providing high-reliability reference value for the optimization decision of the welding process; according to the scheme, the ship welding defects can be predicted only by collecting the data in real time, the user can make relevant response measures after receiving the defect identification result, the possible ship welding defects cannot be predicted in advance, if the user can obtain the prediction information related to the possible ship welding defects in advance, the ship welding defects can be better evaluated and corresponding response measures can be made, and the time for remedying the welding defects by the user can be increased.
Disclosure of Invention
The invention aims to provide a ship welding defect identification system based on artificial intelligence aiming at the defects.
The invention adopts the following technical scheme:
the ship welding defect identification system based on artificial intelligence is characterized by comprising a data acquisition module, a feature extraction module, a defect identification module and a defect prediction module;
the device comprises a data acquisition module, a characteristic extraction module, a defect identification module and a defect prediction module, wherein the data acquisition module is used for acquiring multidimensional data related to ship welding defects, the characteristic extraction module is used for extracting dimension data with larger correlation with the welding defects in the multidimensional data, the defect identification module is used for identifying the ship welding defects, and the defect prediction module predicts the possible ship welding defects based on the identification result;
the data acquisition module comprises a real-time data acquisition module and a historical data acquisition module, wherein the real-time data acquisition module is used for acquiring multidimensional data in a real-time welding process, and the historical data acquisition module is used for acquiring the multidimensional data and corresponding time point defect information in the historical welding process; the real-time data acquisition module comprises a data sensing unit and a data integration unit, wherein the data sensing unit is used for acquiring multidimensional data in a real-time welding process, and the data integration unit is used for performing calibration, formatting and time synchronization operation processing on the multidimensional data;
the feature extraction module extracts dimension data with larger correlation with welding defects from the multi-dimension data through the following steps:
s101: preparing data: acquiring multidimensional data and corresponding time point defect information in the history welding process;
s102: constructing a rough set: based on the data obtained in the previous step, a rough set model is established, wherein the rough set model can be expressed as (U, C U D), and U is a domain, namely a data set containing multidimensional data in the history welding process; c is a conditional feature set comprising all dimensions in the domain U, D is a decision feature set comprising various defect categories and defect category values; the defect type value is the dimension corresponding to the defect type;
s103: extracting important dimensions: calculating the approximate classification quality of each dimension, wherein the approximate classification quality is calculated by the ratio of the lower approximation of each dimension to the domain; comparing the approximate classification quality with a set extraction threshold value, and taking the dimension corresponding to the approximate classification quality larger than the threshold value as the dimension with larger correlation with the welding defect;
further, the defect recognition module establishes a BP neural network model, uses multidimensional data with larger correlation with welding defects and corresponding time point defect information in the history welding process as a training set to train the neural network model, and uses the trained neural network model for identifying the welding defects of the ship; the neural network model is input into multidimensional data with larger correlation with welding defects, and output into occurrence probabilities of various welding defects, wherein the various welding defects are respectively welding cracks, welding holes, welding slag inclusion, welding misalignment, insufficient welding transition areas and welding deformation;
further, the specific mode of predicting the possible ship welding defect by the defect prediction module comprises the following steps:
s301: intercepting 10 groups of neural network model recognition results containing time sequence forwards by taking the current time as the cut-off time; each group of identification results comprise welding cracking, welding holes, welding slag inclusion, welding misalignment, insufficient welding transition area and welding deformation probability during ship welding;
s302: calculating the average probability value of each welding defect in 10 groups of identification results;
s303: extracting the identification results corresponding to the two welding defects with the largest average probability values in the step S302;
s304: calculating an early warning value warning according to the data extracted in the step S303:
wherein ,a is the average probability value corresponding to the welding defect with the largest average probability value in the welding defects max A is the maximum value of the 10 groups of identification results of the welding defects with the maximum average probability value among the welding defects min The method comprises the steps that the minimum value of the welding defects with the largest average probability value in 10 groups of identification results is used as the welding defect; delta a is the difference value between the recognition result closest to the current time and the recognition result second closest to the current time among 10 groups of recognition results of the welding defects with the largest average probability value among the welding defects; b max B, maximum value of 10 groups of identification results of the welding defects with the average probability value of the welding defects being the second largest min For the minimum value of the second largest welding defect in the 10 groups of recognition results of the welding defects, delta b is the difference value between the recognition result closest to the current time and the recognition result closest to the current time of the 10 groups of recognition results of the welding defects with the second largest average probability value;
s305: and comparing the early warning value with a defect threshold, and outputting the welding defect with the largest average probability value among the welding defects in the step S304 as the predicted possible ship welding defect when the early warning value is larger than the defect threshold.
The beneficial effects obtained by the invention are as follows:
the feature extraction module of the system is based on the rough set theory, and the importance of each dimension is measured by calculating the approximate classification quality of each dimension, so that the dimension which has no obvious influence on defect judgment can be filtered out, and the dimension with more distinguishing and predicting capabilities is extracted; the cost of calculation and storage is reduced, the efficiency of feature selection and data analysis is improved, and the task of identifying and predicting the welding defects of the subsequent ships is better supported; the real-time ship welding defects are identified through the defect identification module, and the ship welding defects which possibly occur in the future are predicted through the defect prediction module, so that a user can be guaranteed to evaluate the ship welding defects and make corresponding response measures according to the real-time identification result and the prediction result, the time for remedying the defects for the user is increased, and the structural strength and the safety of ship welding are improved.
For a further understanding of the nature and the technical aspects of the present invention, reference should be made to the following detailed description of the invention and the accompanying drawings, which are provided for purposes of reference only and are not intended to limit the invention.
Drawings
FIG. 1 is a schematic diagram of the overall structural framework of the present invention;
FIG. 2 is a flow chart of the feature extraction module of the present invention for extracting dimension data having a greater correlation with welding defects;
FIG. 3 is a schematic flow chart of the defect recognition module for recognizing welding defects of a ship according to the present invention;
fig. 4 is a schematic diagram of a process of predicting a welding defect of a ship by using the defect prediction module of the present invention.
Detailed Description
The following embodiments of the present invention are described in terms of specific examples, and those skilled in the art will appreciate the advantages and effects of the present invention from the disclosure herein. The invention is capable of other and different embodiments and its several details are capable of modification and variation in various respects, all without departing from the spirit of the present invention. The drawings of the present invention are merely schematic illustrations, and are not intended to be drawn to actual dimensions. The following embodiments will further illustrate the related art content of the present invention in detail, but the disclosure is not intended to limit the scope of the present invention.
Embodiment one.
The embodiment provides a ship welding defect identification system based on artificial intelligence, which is characterized by comprising a data acquisition module, a feature extraction module, a defect identification module and a defect prediction module;
the device comprises a data acquisition module, a characteristic extraction module, a defect identification module and a defect prediction module, wherein the data acquisition module is used for acquiring multidimensional data related to ship welding defects, the characteristic extraction module is used for extracting dimension data with larger correlation with the welding defects in the multidimensional data, the defect identification module is used for identifying the ship welding defects, and the defect prediction module predicts the possible ship welding defects based on the identification result;
the data acquisition module comprises a real-time data acquisition module and a historical data acquisition module, wherein the real-time data acquisition module is used for acquiring multidimensional data in a real-time welding process, and the historical data acquisition module is used for acquiring the multidimensional data and corresponding time point defect information in the historical welding process; the real-time data acquisition module comprises a data sensing unit and a data integration unit, wherein the data sensing unit is used for acquiring multidimensional data in a real-time welding process, and the data integration unit is used for performing calibration, formatting and time synchronization operation processing on the multidimensional data;
the feature extraction module extracts dimension data with larger correlation with welding defects from the multi-dimension data through the following steps:
s101: preparing data: acquiring multidimensional data and corresponding time point defect information in the history welding process;
s102: constructing a rough set: based on the data obtained in the previous step, a rough set model is established, wherein the rough set model can be expressed as (U, C U D), and U is a domain, namely a data set containing multidimensional data in the history welding process; c is a conditional feature set comprising all dimensions in the domain U, D is a decision feature set comprising various defect categories and defect category values; the defect type value is the dimension corresponding to the defect type;
s103: extracting important dimensions: calculating the approximate classification quality of each dimension, wherein the approximate classification quality is calculated by the ratio of the lower approximation of each dimension to the domain; comparing the approximate classification quality with a set extraction threshold value, and taking the dimension corresponding to the approximate classification quality larger than the threshold value as the dimension with larger correlation with the welding defect;
further, the defect recognition module establishes a BP neural network model, uses multidimensional data with larger correlation with welding defects and corresponding time point defect information in the history welding process as a training set to train the neural network model, and uses the trained neural network model for identifying the welding defects of the ship; the neural network model is input into multidimensional data with larger correlation with welding defects, and output into occurrence probabilities of various welding defects, wherein the various welding defects are respectively welding cracks, welding holes, welding slag inclusion, welding misalignment, insufficient welding transition areas and welding deformation;
further, the specific mode of predicting the possible ship welding defect by the defect prediction module comprises the following steps:
s301: intercepting 10 groups of neural network model recognition results containing time sequence forwards by taking the current time as the cut-off time; each group of identification results comprise welding cracking, welding holes, welding slag inclusion, welding misalignment, insufficient welding transition area and welding deformation probability during ship welding;
s302: calculating the average probability value of each welding defect in 10 groups of identification results;
s303: extracting the identification results corresponding to the two welding defects with the largest average probability values in the step S302;
s304: calculating an early warning value warning according to the data extracted in the step S303:
wherein ,a is the average probability value corresponding to the welding defect with the largest average probability value in the welding defects max A is the maximum value of the 10 groups of identification results of the welding defects with the maximum average probability value among the welding defects min The method comprises the steps that the minimum value of the welding defects with the largest average probability value in 10 groups of identification results is used as the welding defect; delta a is the difference value between the recognition result closest to the current time and the recognition result second closest to the current time among 10 groups of recognition results of the welding defects with the largest average probability value among the welding defects; b max B, maximum value of 10 groups of identification results of the welding defects with the average probability value of the welding defects being the second largest min For the minimum value of the second largest welding defect in the 10 groups of recognition results of the welding defects, delta b is the difference value between the recognition result closest to the current time and the recognition result closest to the current time of the 10 groups of recognition results of the welding defects with the second largest average probability value;
s305: and comparing the early warning value with a defect threshold, and outputting the welding defect with the largest average probability value among the welding defects in the step S304 as the predicted possible ship welding defect when the early warning value is larger than the defect threshold.
Embodiment two.
This embodiment should be understood to include at least all of the features of any one of the foregoing embodiments, and be further modified based thereon;
the embodiment provides a ship welding defect identification system based on artificial intelligence, which is characterized by comprising a data acquisition module, a feature extraction module, a defect identification module and a defect prediction module;
the device comprises a data acquisition module, a characteristic extraction module, a defect identification module and a defect prediction module, wherein the data acquisition module is used for acquiring multidimensional data related to ship welding defects, the characteristic extraction module is used for extracting dimension data with larger correlation with the welding defects in the multidimensional data, the defect identification module is used for identifying the ship welding defects, and the defect prediction module predicts the possible ship welding defects based on the identification result;
the data acquisition module comprises a real-time data acquisition module and a historical data acquisition module, wherein the real-time data acquisition module is used for acquiring multidimensional data in a real-time welding process, and the historical data acquisition module is used for acquiring the multidimensional data and corresponding time point defect information in the historical welding process; the real-time data acquisition module comprises a data sensing unit and a data integration unit, wherein the data sensing unit is used for acquiring multidimensional data in a real-time welding process, and the data integration unit is used for performing calibration, formatting and time synchronization operation processing on the multidimensional data;
the data sensing unit includes, but is not limited to, the following sensing devices or information acquisition devices: an imaging device, a current sensor, a voltage sensor, an inductance sensor, a welding gas flow sensor, and the like; these sensing devices are merely examples and other types of sensing devices may be used to obtain multidimensional data during welding of a vessel;
the system also comprises a communication module, a feature extraction module and a data acquisition module, wherein the communication module is used for completing data transmission between the data acquisition module and the feature extraction module;
the feature extraction module is used for extracting dimension data with larger correlation with welding defects from the dimension data by the following steps:
s101: preparing data: acquiring multidimensional data and corresponding time point defect information in the history welding process;
s102: constructing a rough set: based on the data obtained in the previous step, a rough set model is established, wherein the rough set model can be expressed as (U, C U D), and U is a domain, namely a data set containing multidimensional data in the history welding process; c is a conditional feature set comprising all dimensions in the domain U, D is a decision feature set comprising various defect categories and defect category values; the defect type value is the dimension corresponding to the defect type;
s103: extracting important dimensions: calculating the approximate classification quality of each dimension, wherein the approximate classification quality is calculated by the ratio of the lower approximation of each dimension to the domain; comparing the approximate classification quality with a set extraction threshold value, and taking the dimension corresponding to the approximate classification quality larger than the threshold value as the dimension with larger correlation with the welding defect;
the defect identification module is used for identifying the welding defects of the ship by establishing a BP neural network model; the method specifically comprises the following steps:
s201: establishing a BP neural network model, and determining the node numbers of an input layer, a hidden layer and an output layer in the neural network model; the number of nodes of the input layer is the number of dimensions in multidimensional data with larger correlation with welding defects, the number of nodes of the output layer is 7, and the types of the welding defects are welding cracks, welding holes, welding slag inclusion, welding misalignment, insufficient welding transition areas and welding deformation; hidden layer node number n 1 The method comprises the following steps:
wherein n is the number of nodes of the feature layer, m is the number of nodes of the output layer, k is a constant with a value range of [1,10], and the value is obtained through experimental verification;
s202: determining an activation function: setting activation functions of a hidden layer and an output layer in the neural network model as a sigm oid function;
s203: model training optimization: training the weight and the threshold value in the optimization model of the neural network model by using the data acquired by the historical data acquisition module as a training set; the model training optimization can be realized through algorithms such as a gradient descent method, a genetic algorithm or a particle swarm algorithm in the prior art, and the description is omitted here;
s204: the trained neural network model is used for real-time prediction, and the input of the neural network model is multidimensional data with larger correlation with welding defects in the real-time welding process; the output identification result is the occurrence probability of the defect type corresponding to the output layer node;
the specific mode of predicting the welding defects of the ship which possibly occur by the defect prediction module is as follows:
s301: intercepting 10 groups of neural network model recognition results containing time sequence forwards by taking the current time as the cut-off time; each group of identification results comprise welding cracking, welding holes, welding slag inclusion, welding misalignment, insufficient welding transition area and welding deformation probability during ship welding;
s302: calculating the average probability value of each welding defect in 10 groups of identification results;
s303: extracting the identification results corresponding to the two welding defects with the largest average probability values in the step S302;
s304: calculating an early warning value warning according to the data extracted in the step S303:
wherein ,a is the average probability value corresponding to the welding defect with the largest average probability value in the welding defects max A is the maximum value of the 10 groups of identification results of the welding defects with the maximum average probability value among the welding defects min The method comprises the steps that the minimum value of the welding defects with the largest average probability value in 10 groups of identification results is used as the welding defect; delta a is the difference value between the recognition result closest to the current time and the recognition result second closest to the current time among 10 groups of recognition results of the welding defects with the largest average probability value among the welding defects; b max B, maximum value of 10 groups of identification results of the welding defects with the average probability value of the welding defects being the second largest min For the minimum value of the second largest welding defect in the 10 groups of recognition results of the welding defects, delta b is the difference value between the recognition result closest to the current time and the recognition result closest to the current time of the 10 groups of recognition results of the welding defects with the second largest average probability value;
s305: comparing the early warning value with a defect threshold, and outputting the welding defect with the largest average probability value in the welding defects in the step S304 as the predicted possible ship welding defect when the early warning value is larger than the defect threshold;
in the welding of ships, there is often an interconnection between various welding defects, and different defect types may occur for the same or similar reasons; in the step S304, the related factors of two welding defects are considered, so that the calculation of the early warning value is more close to reality;
the system also comprises an early warning module which is respectively connected with the defect recognition module and the defect prediction module, and sends out real-time defect early warning information to a user side when judging that the situation that the recognition result of the defect recognition module exceeds a preset warning threshold exists; when the defect prediction module outputs the predicted ship welding defect, the predicted defect early warning information is sent to the user side; the specific forms of the real-time defect early warning information and the predicted defect early warning information can be alarm information, short messages, mails, push notifications and the like, so that the user side can be ensured to know in time and take corresponding measures.
The foregoing disclosure is only a preferred embodiment of the present invention and is not intended to limit the scope of the invention, so that all equivalent technical changes made by applying the description of the present invention and the accompanying drawings are included in the scope of the present invention, and in addition, elements in the present invention can be updated as the technology develops.
Claims (3)
1. The ship welding defect identification system based on artificial intelligence is characterized by comprising a data acquisition module, a feature extraction module, a defect identification module and a defect prediction module;
the device comprises a data acquisition module, a characteristic extraction module, a defect identification module and a defect prediction module, wherein the data acquisition module is used for acquiring multidimensional data related to ship welding defects, the characteristic extraction module is used for extracting dimension data with larger correlation with the welding defects in the multidimensional data, the defect identification module is used for identifying the ship welding defects, and the defect prediction module predicts the possible ship welding defects based on the identification result;
the data acquisition module comprises a real-time data acquisition module and a historical data acquisition module, wherein the real-time data acquisition module is used for acquiring multidimensional data in a real-time welding process, and the historical data acquisition module is used for acquiring the multidimensional data and corresponding time point defect information in the historical welding process; the real-time data acquisition module comprises a data sensing unit and a data integration unit, wherein the data sensing unit is used for acquiring multidimensional data in a real-time welding process, and the data integration unit is used for performing calibration, formatting and time synchronization operation processing on the multidimensional data;
the feature extraction module extracts dimension data with larger correlation with welding defects from the multi-dimension data through the following steps:
s101: preparing data: acquiring multidimensional data and corresponding time point defect information in the history welding process;
s102: constructing a rough set: based on the data obtained in the previous step, a rough set model is established, wherein the rough set model can be expressed as (U, C U D), and U is a domain, namely a data set containing multidimensional data in the history welding process; c is a conditional feature set comprising all dimensions in the domain U, D is a decision feature set comprising various defect categories and defect category values; the defect type value is the dimension corresponding to the defect type;
s103: extracting important dimensions: calculating the approximate classification quality of each dimension, wherein the approximate classification quality is calculated by the ratio of the lower approximation of each dimension to the domain; and comparing the approximate classification quality with a set extraction threshold value, and taking the dimension corresponding to the approximate classification quality larger than the threshold value as the dimension with larger correlation with the welding defect.
2. The ship welding defect recognition system based on artificial intelligence according to claim 1, wherein the defect recognition module establishes a BP neural network model, trains the neural network model by using multidimensional data with larger correlation with welding defects and corresponding time point defect information in a history welding process as a training set, and uses the trained neural network model for ship welding defect recognition; the neural network model is input into multidimensional data with larger correlation with welding defects, and output into occurrence probabilities of various welding defects, wherein the various welding defects are respectively welding cracks, welding holes, welding slag inclusion, welding misalignment, insufficient welding transition areas and welding deformation.
3. The artificial intelligence based ship welding defect recognition system of claim 2, wherein the defect prediction module predicts the possible occurrence of the ship welding defect in a specific manner comprising the steps of:
s301: intercepting 10 groups of neural network model recognition results containing time sequence forwards by taking the current time as the cut-off time; each group of identification results comprise welding cracking, welding holes, welding slag inclusion, welding misalignment, insufficient welding transition area and welding deformation probability during ship welding;
s302: calculating the average probability value of each welding defect in 10 groups of identification results;
s303: extracting the identification results corresponding to the two welding defects with the largest average probability values in the step S302;
s304: calculating an early warning value warning according to the data extracted in the step S303:
wherein ,a is the average probability value corresponding to the welding defect with the largest average probability value in the welding defects max A is the maximum value of the 10 groups of identification results of the welding defects with the maximum average probability value among the welding defects min The method comprises the steps that the minimum value of the welding defects with the largest average probability value in 10 groups of identification results is used as the welding defect; delta a is the difference value between the recognition result closest to the current time and the recognition result second closest to the current time among 10 groups of recognition results of the welding defects with the largest average probability value among the welding defects; b max B, maximum value of 10 groups of identification results of the welding defects with the average probability value of the welding defects being the second largest min For the minimum value of the second largest welding defect in the 10 groups of recognition results of the welding defects, delta b is the difference value between the recognition result closest to the current time and the recognition result closest to the current time of the 10 groups of recognition results of the welding defects with the second largest average probability value;
s305: and comparing the early warning value with a defect threshold, and outputting the welding defect with the largest average probability value among the welding defects in the step S304 as the predicted possible ship welding defect when the early warning value is larger than the defect threshold.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310874183.9A CN116910619A (en) | 2023-07-17 | 2023-07-17 | Ship welding defect recognition system based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310874183.9A CN116910619A (en) | 2023-07-17 | 2023-07-17 | Ship welding defect recognition system based on artificial intelligence |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116910619A true CN116910619A (en) | 2023-10-20 |
Family
ID=88362429
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310874183.9A Pending CN116910619A (en) | 2023-07-17 | 2023-07-17 | Ship welding defect recognition system based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116910619A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117984024A (en) * | 2024-04-03 | 2024-05-07 | 中国水利水电第十工程局有限公司 | Welding data management method and system based on automatic production of ship lock lambdoidal doors |
-
2023
- 2023-07-17 CN CN202310874183.9A patent/CN116910619A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117984024A (en) * | 2024-04-03 | 2024-05-07 | 中国水利水电第十工程局有限公司 | Welding data management method and system based on automatic production of ship lock lambdoidal doors |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111353413B (en) | Low-missing-report-rate defect identification method for power transmission equipment | |
| WO2021042935A1 (en) | Bearing service life prediction method based on hidden markov model and transfer learning | |
| CN116625438B (en) | Gas pipe network safety on-line monitoring system and method thereof | |
| JP2020123330A (en) | Method for acquiring sample image for label acceptance inspection from among auto-labeled images utilized for neural network learning, and sample image acquisition device utilizing the same | |
| CN116910619A (en) | Ship welding defect recognition system based on artificial intelligence | |
| CN106198749B (en) | A kind of data fusion method of the multiple sensor based on Metal Crack monitoring | |
| CN112631240A (en) | Spacecraft fault active detection method and device | |
| CN111862065B (en) | Power transmission line diagnosis method and system based on multitask deep convolutional neural network | |
| CN111461784B (en) | Multi-model fusion-based fraud detection method | |
| CN110851422A (en) | Data anomaly monitoring model construction method based on machine learning | |
| CN111717753A (en) | Self-adaptive elevator fault early warning system and method based on multi-dimensional fault characteristics | |
| CN116205265A (en) | Power grid fault diagnosis method and device based on deep neural network | |
| CN115507309A (en) | Gas leakage alarm method and system | |
| CN115035328A (en) | Converter image increment automatic machine learning system and establishment training method thereof | |
| CN114398992A (en) | Intelligent fault diagnosis method based on unsupervised domain adaptation | |
| Zhang et al. | Fault diagnosis of natural gas pipeline leakage based on 1D-CNN and self-attention mechanism | |
| CN112232235B (en) | Intelligent factory remote monitoring method and system based on 5G sum algorithm | |
| CN114283367B (en) | Artificial intelligent open fire detection method and system for garden fire early warning | |
| CN115393714A (en) | Power transmission line bolt missing pin detection method based on fusion graph theory reasoning | |
| CN113328986A (en) | Network flow abnormity detection method based on combination of convolutional neural network and LSTM | |
| CN113008998A (en) | Hidden engineering internal defect judgment method based on PCNN | |
| CN112802011A (en) | Fan blade defect detection method based on VGG-BLS | |
| Pan | Corrosion detection method of substation Aboveground Steel Structure Based on deep learning | |
| CN116787017B (en) | Control method and system for hydraulic turbine seat ring welding robot | |
| CN117854215A (en) | Fire alarm method and device based on time sequence image |
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 |