CN115563549B - Welding defect cause diagnosis method and system and electronic equipment - Google Patents
Welding defect cause diagnosis method and system and electronic equipment Download PDFInfo
- Publication number
- CN115563549B CN115563549B CN202211323505.2A CN202211323505A CN115563549B CN 115563549 B CN115563549 B CN 115563549B CN 202211323505 A CN202211323505 A CN 202211323505A CN 115563549 B CN115563549 B CN 115563549B
- Authority
- CN
- China
- Prior art keywords
- diagnosis
- fuzzy
- rule
- neural network
- cause
- 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.)
- Active
Links
Classifications
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/04—Inference or reasoning models
- G06N5/048—Fuzzy inferencing
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Automation & Control Theory (AREA)
- Fuzzy Systems (AREA)
- Investigating And Analyzing Materials By Characteristic Methods (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a welding defect cause diagnosis method, a system and an electronic device, wherein the method comprises the steps of establishing a fuzzy diagnosis rule and a cause diagnosis neural network database, combining the fuzzy diagnosis rule and the cause diagnosis neural network database to perform defect diagnosis, performing front item expansion based on a traditional fuzzy rule, adding a threshold, performing fuzzy reasoning by using a sample rule selected from the fuzzy diagnosis rule, obtaining a target diagnosis result according to the sample rule if the actual credibility is greater than a preset threshold, and performing complementary processing by using the cause diagnosis neural network if the actual credibility is not present, so as to obtain an ideal target diagnosis result. Compared with the traditional fuzzy neural network, the fuzzy neural network and the fuzzy rule processing method based on the fuzzy neural network combine two theories, namely, the fuzzy neural network has the advantage that fuzzy reasoning can process uncertainty knowledge, also has the advantages of self-learning, high efficiency and high precision of the neural network, greatly improves the reliability of the reasoning process, and is more suitable for welding the scene of multiple influencing factors.
Description
Technical Field
The present invention relates to the field of welding defect diagnosis technologies, and in particular, to a method, a system, and an electronic device for diagnosing a cause of a welding defect.
Background
The analysis of the cause of the occurrence of the welding defect is a difficulty in the welding field, and the defect type and form are various, and the defect formation cause is complex, so that the diagnosis of the defect cause is too dependent on manual experience, and the efficiency and the reliability are extremely low. To solve this problem, welding expert systems having intelligent diagnosis function are increasingly being used.
Aiming at the research of welding defect cause detection, china patent (CN 109447403A) provides a welding defect analysis system based on big data, but the quality of a selected data set directly influences the diagnosis of the last welding defect cause, and a knowledge base does not have corresponding self-learning capability and lacks versatility. The Chinese patent (CN 105223275A) proposes a sparse matrix-based welding defect diagnosis method, but the database sources obtained by the method are all homemade single prefabricated defect standard samples, and are not applicable to detection of multi-defect nonstandard weldments.
According to the research, the prior art can be known to perform deduction based on the generation rule, namely, the prior art can be deduced based on the existing sample and the existing data set, and cannot well reflect the complex nonlinear relation among the defect characteristics, the process parameters and the welding defect causes, so that the diagnosis accuracy is low, the method is not applicable to the diagnosis and identification of the welding defects, such as nonlinear and multi-classification problems, and uncertainty knowledge cannot be well processed.
Disclosure of Invention
In view of the foregoing, it is necessary to provide a method, a system and an electronic device for diagnosing a welding defect cause, which are used for solving the problems that the conventional welding defect cause diagnosis technology is too dependent on the existing sample, and can not well process the non-linearity, multi-classification and uncertainty knowledge of the welding defect diagnosis.
In order to achieve the technical purpose, the invention adopts the following technical scheme:
in a first aspect, the present invention provides a welding defect cause diagnosis method, including:
determining an evaluation factor and a defect cause, and establishing a plurality of fuzzy diagnosis rules for representing fuzzy relations between the evaluation factor and the defect cause based on fuzzy rules of an expansion front;
establishing a cause diagnosis neural network model, wherein input parameters of the cause diagnosis neural network model comprise welding defect types and evaluation factors, and output parameters of the cause diagnosis neural network model comprise diagnosis results;
Acquiring the actual welding defect type and evaluation factor data corresponding to the defect to be diagnosed;
based on the evaluation factor data, selecting a matched fuzzy diagnosis rule from a plurality of fuzzy diagnosis rules as a sample rule, and performing fuzzy reasoning on the sample rule to obtain the actual credibility of each sample rule;
judging whether the sample rule with the actual credibility larger than a preset threshold exists or not, if so, obtaining a target diagnosis result according to the defect cause of the sample rule with the actual credibility larger than the preset threshold, and if not, obtaining the target diagnosis result according to the cause diagnosis neural network model.
Further, the determining the evaluation factor and the defect cause establishes a plurality of fuzzy diagnosis rules for representing fuzzy relations between the evaluation factor and the defect cause based on the fuzzy rules of the expansion foreterm, and the method comprises the following steps:
establishing a welding characteristic information base, and obtaining the evaluation factors and the defect causes according to the welding characteristic information base;
establishing a plurality of fuzzy diagnosis rules based on fuzzy rules of the extension antecedents according to the evaluation factors;
each fuzzy diagnosis rule comprises a front term, a rear term, theoretical credibility and the preset threshold value, wherein the front term comprises a plurality of evaluation factors, and the rear term comprises the defect cause.
Further, the selecting, based on the evaluation factor data, a matched fuzzy diagnosis rule from a plurality of fuzzy diagnosis rules as a sample rule, and performing fuzzy reasoning on the sample rule to obtain an actual credibility of each sample rule includes:
selecting a matched fuzzy diagnosis rule from a plurality of diagnosis rules as the sample rule based on the evaluation factor data;
according to the actual welding defect type, obtaining a weight coefficient of each evaluation factor for the actual welding defect type based on an analytic hierarchy process;
obtaining the membership degree of each evaluation factor in the sample rule according to the evaluation factor data;
and calculating the actual credibility of each sample rule according to the weight coefficient and the membership degree.
Further, the obtaining, based on the analytic hierarchy process, a weight coefficient of each evaluation factor for the actual welding defect type according to the actual welding defect type includes:
establishing a pair comparison matrix of the evaluation factors according to the actual welding defect type;
consistency test is carried out on the paired comparison matrixes, and the paired comparison matrixes are adjusted according to test results;
Obtaining the feature vector of the pair of adjusted comparison matrixes according to the pair of adjusted comparison matrixes;
and normalizing the feature vector to obtain a weight coefficient of each evaluation factor for the actual welding defect type.
Further, the step of judging whether the sample rule with the actual reliability being greater than a preset threshold exists, if yes, obtaining a target diagnosis result according to a defect cause of the sample rule with the actual reliability being greater than the preset threshold, and if no, obtaining the target diagnosis result according to the cause diagnosis neural network model, including:
sequencing the sample rule according to the numerical value of the actual credibility to obtain a sample rule sequence;
judging whether the actual credibility of the sample rule is larger than the corresponding preset threshold value, if so, selecting the defect causes of the set number of the sample rules with the highest actual credibility according to the sample rule sequence to obtain the target diagnosis result, and if not, inputting the evaluation factor data and the actual welding defect type into the cause diagnosis neural network model to obtain the target diagnosis result.
Further, the establishing the causal diagnosis neural network model includes:
establishing an initial BP neural network model, wherein the input parameters of the initial BP neural network model comprise a plurality of evaluation factors, and the output parameters of the initial BP neural network model comprise the diagnosis results;
acquiring initial parameters of the initial BP neural network model, and optimizing through a PSO-BP algorithm to obtain optimized parameters;
and optimizing the initial BP neural network model according to the optimization parameters to obtain the causal diagnosis neural network model.
Further, the obtaining the initial parameters of the initial BP neural network model, and optimizing the initial parameters through a PSO-BP algorithm to obtain optimized parameters includes:
acquiring initial parameters of the initial BP neural network model, wherein the initial parameters comprise weights, thresholds and loss functions of neurons;
establishing a solution space and a particle population according to the weight and the threshold of the neuron, and initializing the position of the particle population in the solution space;
establishing an adaptability function according to the loss function;
optimizing the position of the particle population in the solution space according to the fitness function;
And acquiring the position of the optimized particle population in the solution space to obtain the optimization parameters.
In a second aspect, the present invention also provides a welding defect cause diagnosis system, comprising:
the rule establishing module is used for determining an evaluation factor and a defect cause, and establishing a plurality of fuzzy diagnosis rules for representing fuzzy relations between the evaluation factor and the defect cause based on fuzzy rules of an expansion front term;
the network establishment module is used for establishing a cause diagnosis neural network model, wherein the input parameters of the cause diagnosis neural network model comprise welding defect types and evaluation factors, and the output parameters of the cause diagnosis neural network model comprise diagnosis results;
the data acquisition module is used for acquiring the actual welding defect type and evaluation factor data corresponding to the defect to be diagnosed;
the fuzzy reasoning module is used for selecting a matched fuzzy diagnosis rule from a plurality of fuzzy diagnosis rules as a sample rule based on the evaluation factor data, and carrying out fuzzy reasoning on the sample rule to obtain the actual credibility of each sample rule;
and the result diagnosis module is used for judging whether the sample rule with the actual credibility being larger than a preset threshold exists or not, if so, obtaining a target diagnosis result according to the defect cause of the sample rule with the actual credibility being larger than the preset threshold, and if not, obtaining the target diagnosis result according to the neural network model.
In a third aspect, the invention also provides an electronic device comprising a memory and a processor, wherein,
a memory for storing a program;
and a processor coupled to the memory for executing the program stored in the memory to implement the steps in the welding defect cause diagnosis method in any one of the above implementations.
The invention provides a welding defect cause diagnosis method, a system and electronic equipment, wherein the method comprises the steps of establishing a fuzzy diagnosis rule and a cause diagnosis neural network database, combining the fuzzy diagnosis rule and the cause diagnosis neural network database to diagnose the welding defect, specifically, performing prior expansion based on the traditional fuzzy rule, adding a threshold value, performing fuzzy reasoning by using a sample rule selected from the fuzzy diagnosis rule, obtaining a target diagnosis result according to the sample rule if the sample rule with actual credibility being greater than a preset threshold value exists, and considering that the existing fuzzy diagnosis rule cannot process the existing fact if the sample rule does not exist, and performing complementary processing by utilizing the cause diagnosis neural network at the moment, thereby obtaining an ideal target diagnosis result. Compared with the prior art, the fuzzy rule and the neural network are combined, so that the diagnosis process has the advantages of fuzzy reasoning and uncertainty knowledge processing, the neural network self-learning and high efficiency and high precision, the reliability of the reasoning process is greatly improved, and the fuzzy neural network model is more suitable for welding the scene of multiple influencing factors compared with the traditional fuzzy neural network model.
Drawings
FIG. 1 is a flow chart of a method for diagnosing a cause of a weld defect according to an embodiment of the present application;
FIG. 2 is a schematic diagram of a welding feature information library established in an embodiment of a welding defect cause diagnosis method according to the present application;
FIG. 3 is a flowchart illustrating a method according to an embodiment of step S104 in FIG. 1;
FIG. 4 is a flowchart illustrating a method according to an embodiment of step S302 in FIG. 3;
FIG. 5 is a schematic diagram of a system for diagnosing a cause of a weld defect according to an embodiment of the present application;
fig. 6 is a schematic structural diagram of an electronic device provided by the present application.
Detailed Description
The following detailed description of preferred embodiments of the application is made in connection with the accompanying drawings, which form a part hereof, and together with the description of the embodiments of the application, are used to explain the principles of the application and are not intended to limit the scope of the application.
In the description of the present application, the meaning of "a plurality" is two or more, unless explicitly defined otherwise.
Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the application. The appearances of such phrases in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those of skill in the art will explicitly and implicitly appreciate that the embodiments described herein may be combined with other embodiments.
According to the welding defect cause diagnosis method, the system and the electronic equipment, on one hand, the fuzzy theory is utilized to improve the capability of processing uncertainty knowledge; on the other hand, the fact that fuzzy reasoning cannot be processed is supplemented by utilizing a neural network connection mechanism, so that the method has the advantages of two theories, is beneficial to diagnosing welding defect causes under the condition of lack of rules or low matching degree, and has important engineering practice significance for promoting automation of a welding process.
The invention provides a welding defect cause diagnosis method, a system and electronic equipment, and the method and the system are respectively described below.
Referring to fig. 1, in one embodiment of the present invention, a method for diagnosing a cause of a welding defect is disclosed, comprising:
s101, determining an evaluation factor and a defect cause, and establishing a plurality of fuzzy diagnosis rules for representing fuzzy relations between the evaluation factor and the defect cause based on fuzzy rules of an expansion front;
s102, establishing a cause diagnosis neural network model, wherein input parameters of the cause diagnosis neural network model comprise welding defect types and evaluation factors, and output parameters of the cause diagnosis neural network model comprise diagnosis results;
S103, acquiring actual welding defect type and evaluation factor data corresponding to the defect to be diagnosed;
s104, based on the evaluation factor data, selecting a matched fuzzy diagnosis rule from a plurality of fuzzy diagnosis rules as a sample rule, and performing fuzzy reasoning on the sample rule to obtain the actual credibility of each sample rule;
s105, judging whether the sample rule with the actual credibility larger than a preset threshold exists, if so, obtaining a target diagnosis result according to the defect cause of the sample rule with the actual credibility larger than the preset threshold, and if not, obtaining the target diagnosis result according to the cause diagnosis neural network model.
The invention provides a welding defect cause diagnosis method, a system and electronic equipment, wherein the method comprises the steps of establishing a fuzzy diagnosis rule and a cause diagnosis neural network database, combining the fuzzy diagnosis rule and the cause diagnosis neural network database to diagnose the welding defect, specifically, performing prior expansion based on the traditional fuzzy rule, adding a threshold value, performing fuzzy reasoning by using a sample rule selected from the fuzzy diagnosis rule, obtaining a target diagnosis result according to the sample rule if the sample rule with actual credibility being greater than a preset threshold value exists, and considering that the existing fuzzy diagnosis rule cannot process the existing fact if the sample rule does not exist, and performing complementary processing by utilizing the cause diagnosis neural network at the moment, thereby obtaining an ideal target diagnosis result. Compared with the prior art, the fuzzy rule and the neural network are combined, so that the diagnosis process has the advantages of fuzzy reasoning and uncertainty knowledge processing, the neural network self-learning and high efficiency and high precision, the reliability of the reasoning process is greatly improved, and the fuzzy neural network model is more suitable for welding the scene of multiple influencing factors compared with the traditional fuzzy neural network model.
It should be noted that, the evaluation factors in the present embodiment include any influencing factors that can cause a welding defect, and the defect causes refer to specific quantification situations of influencing factors that form a defect to be diagnosed, which indicate specific causes of forming the defect, and the diagnosis results include the defect causes, and diagnosis conclusions that can be derived according to the defect causes, such as the causes of possible occurrence of the defect, the possibility of each cause, the process optimization scheme, and the like.
Specifically, as a preferred embodiment, in step S101 in the present embodiment, an evaluation factor and a defect cause are determined, and a plurality of fuzzy diagnosis rules for characterizing a fuzzy relationship between the evaluation factor and the defect cause are established based on a fuzzy rule of an extension precursor, specifically including:
establishing a welding characteristic information base, and obtaining the evaluation factors and the defect causes according to the welding characteristic information base;
and establishing a plurality of fuzzy diagnosis rules based on fuzzy rules of an extended precursor according to the evaluation factors, wherein each fuzzy diagnosis rule comprises a precursor, a postion, theoretical credibility and the preset threshold, the precursor comprises a plurality of evaluation factors, and the postion comprises the defect cause.
Compared with the traditional fuzzy rule model, the welding source obtained by the method meets the diagnosis requirement of actual welding.
Fig. 2 shows a welding characteristic information base established in the present embodiment, which may be directly obtained from the prior art, and then a test or the like is selected according to actual needs, so as to determine required evaluation factors from the welding characteristic information base.
In this embodiment, the determined evaluation factors include: welding method C 0 Distance C between defect relative centers 1 Nearest distance C between two defects 2 Aspect ratio of defects C 3 Thickness C of plate 4 Welding current C 5 Welding voltage C 6 Welding speed C 7 Wire feed speed C 8 Flow rate C of gas 9 Ten factors.
After the evaluation factors are determined, a fuzzy diagnosis rule can be established, in the embodiment, the fuzzy diagnosis rule is designed by adopting a C-F model based on the fuzzy rule, and further expansion is carried out, particularly, the previous item is expanded, so that the membership degree of each condition is increased. The fuzzy diagnostic rules in this embodiment are designed as follows:
IF E 1 (α 1 ,η 1 )AND E 2 (α 2 ,η 2 )…AND E n (α n ,η n )THEN H(CF(H),λ)
wherein E is 1 、E 2 、…、E n Corresponds to various known facts, namely the actual situation of each evaluation factor; alpha 1 、α 2 、…、α n As the weight coefficient, i.e. the influence degree of each fact on the defect cause, is ensured in the embodiment η 1 ,η 2 ,…,η n Representing the membership of this fact, CF (H) represents the theoretical confidence level of the whole rule, λ being the preset threshold of this rule. Among the above parameters, the weight coefficient and the membership degree will be calculated correspondingly in the subsequent step according to the actual situation.
The following table is a part of the fuzzy diagnostic rules established in the present embodiment:
part rule for diagnosing defect cause
The fuzzy diagnosis rule is used for reasoning, is more suitable for welding scenes with multiple influencing factors compared with a traditional fuzzy rule model, improves judging accuracy, can provide a reasonable analysis scheme, effectively replaces manual specialists, saves a large amount of time and labor cost, helps operators to effectively reduce influences caused by welding defects in time, and improves automation degree of welding production. Meanwhile, as a preferred embodiment, in step S102 in the present embodiment, a cause diagnosis neural network model is established, where input parameters of the cause diagnosis neural network model include a welding defect type and the evaluation factor, and output parameters of the cause diagnosis neural network model include a diagnosis result, and specifically includes:
establishing an initial BP neural network model, wherein the input parameters of the initial BP neural network model comprise a plurality of evaluation factors, and the output parameters of the initial BP neural network model comprise diagnosis results;
Acquiring initial parameters of the initial BP neural network model, and optimizing through a PSO-BP algorithm to obtain optimized parameters;
and optimizing the initial BP neural network model according to the optimization parameters to obtain the causal diagnosis neural network model.
Specifically, in a preferred embodiment, the obtaining the initial parameters of the initial BP neural network model in the above process, and optimizing the initial parameters by using a PSO-BP algorithm, to obtain optimized parameters specifically includes:
acquiring initial parameters of the initial BP neural network model, wherein the initial parameters comprise weights, thresholds and loss functions of neurons;
establishing a solution space and a particle population according to the weight and the threshold of the neuron, and initializing the position of the particle population in the solution space;
establishing an adaptability function according to the loss function;
optimizing the position of the particle population in the solution space according to the fitness function;
and acquiring the position of the optimized particle population in the solution space to obtain the optimization parameters.
To facilitate an understanding of the above process, the present invention also provides a more specific embodiment in which:
(1) Collecting welding experimental data, and establishing a BP neural network welding defect analysis model of a single hidden layer as an initial BP neural network model according to the process experimental data, wherein input parameters of the initial BP neural network model comprise a plurality of evaluation factors, and output parameters of the initial BP neural network model comprise a diagnosis result;
(2) In order to further improve the convergence efficiency and accuracy of the model, the initial value and the parameter of the initial BP neural network model are updated and improved by utilizing the initial parameter of the initial BP neural network through an improved PSO-BP algorithm, and the specific flow comprises the following steps:
(1) setting specific parameters of a particle swarm algorithm by using initial parameters of an initial BP neural network, setting weights and thresholds of all neurons in the initial BP neural network structure as position variables of particles, and establishing a solution space and a particle population according to the weights and thresholds of the neurons; setting a loss function of an initial BP neural network as an fitness function of particles, initializing a population of particles, namely initializing the position of the population of particles in the solution space;
(2) for each example, the particle fitness f can be calculated i And minimum fitness f with the particle min Comparing if f i ＜f min The updated fitness is taken as the minimum fitness f of the particle min The corresponding position is the optimal position P i ；
(3) The particle swarm searches the whole state space by updating the speed and the position of the particle swarm, and updates the particle swarm by tracking an individual extremum (an optimal solution of the best fitness found by the particle swarm) and a global extremum (an optimal solution of the best fitness found by the whole population at present) until the optimal solution is found;
the evolution formula of the particle speed and the position in the process is as follows:
wherein ω is nonlinear dynamic change inertial weight; c 1 And c 2 Is a learning factor; r is (r) 1 And r 2 To take the value of 0,1]Random numbers in between; p is p in (k) Representing individual position extremum of the particles; p is p gn (k) Representing a global position extremum of the particle; x is x in (k) The position of particle i at time k; x is x in (k+1) is the position of particle i at time k+1; v in (k) The velocity of particle i at time k; v in (k+1) is the velocity of particle i at time k+1;
wherein, the nonlinear dynamic change inertia weight ω is expressed as:
wherein omega min Is the minimum inertial weight; omega max For maximum inertial weight, l 1 (t) is a nonlinear function, l 2 (t) is a linear function, f i For the fitness of the current particle, f min Is the minimum fitness of the particles. l (L) 1 (t)、l 2 (t) is determined by the following formula:
where t is the current iteration number of the model, t max The maximum iteration number of the model is set;
(4) calculating the minimum fitness f of the whole particle swarm g ＝min{f 1 ，f 2 ，…，f M Judging whether the current iteration number reaches the set maximum iteration number, stopping iteration if the current iteration number reaches the set maximum iteration number, otherwise, turning to (2) to continuously update;
(5) outputting the neural network parameters P determined by the whole particle swarm g ＝(p g1 ,p g2 ,...,p gD ) As an initial value for the neural network structure. And combining the particle swarm algorithm with the gradient descent algorithm, and further iteratively updating the neural network parameters by using the dynamic weight factors. And outputting the initial parameters of the neural network determined by the improved PSO-BP algorithm, and replacing the corresponding parameters of the initial BP neural network to optimize the initial BP neural network model, so as to obtain the causal diagnosis neural network model.
The following table shows specific values of some other parameters required in the above-described process of the present embodiment:
adaptive PSO-BP algorithm parameter setting based on dynamic weight
In the neural network structure of the embodiment, the original gradient descent algorithm is reserved in the updating mode of the network parameters, the dynamic weight factors are introduced, the PSO algorithm and the BP algorithm are combined, the advantages of different algorithms are exerted at different stages of iteration, and the training speed and the iteration efficiency are improved. The PSO-BP neural network model obtained through training is utilized to carry out supplementary diagnosis on the fact that fuzzy rule reasoning cannot be determined, the forming reason of the welding defect is determined, and the model is corrected according to the evaluation result, for example, the output result is used as a new fuzzy diagnosis rule, so that the whole method has the characteristic of self-learning, the problem that the prior art is too dependent on the existing sample is effectively solved, and the training efficiency and the diagnosis precision of the model are effectively improved; the rule reasoning and the model reasoning are combined to form a rule-model collaborative reasoning mechanism, compared with a traditional neural network or a fuzzy neural network, the self-learning capability of the rule-model collaborative reasoning mechanism is improved, the rule-model collaborative reasoning mechanism is more suitable for welding a scene with multiple influencing factors, and the reliability of a reasoning mode is improved.
After the fuzzy diagnosis rule and the causal diagnosis neural network model are established, the actual diagnosis can be started, and step S103 is firstly required to be carried out to obtain the actual welding defect type and the evaluation factor data corresponding to the defect to be diagnosed, wherein the evaluation factor data is an actual numerical value corresponding to the evaluation factor of the defect to be diagnosed.
Further, referring to fig. 3, as a preferred embodiment, step S104 in this embodiment selects, based on the evaluation factor data, a matched fuzzy diagnosis rule from a plurality of fuzzy diagnosis rules as a sample rule, and performs fuzzy reasoning on the sample rule to obtain an actual credibility of each sample rule, which specifically includes:
s301, selecting a matched fuzzy diagnosis rule from a plurality of diagnosis rules as the sample rule based on the evaluation factor data;
s302, obtaining a weight coefficient of each evaluation factor for the actual welding defect type based on a hierarchical analysis method according to the actual welding defect type;
s303, obtaining the membership degree of each evaluation factor in the sample rule according to the evaluation factor data;
s304, calculating the actual credibility of each sample rule according to the weight coefficient and the membership degree.
The above-mentioned process is the main step of fuzzy inference in this embodiment, and it can be understood that, because fuzzy inference is the prior art that can be understood by those skilled in the relevant arts, in practice, other fuzzy inference methods may be adopted to obtain the reliability.
As a preferred embodiment, the present embodiment firstly eliminates a rule that is obviously unreasonable from a plurality of fuzzy diagnosis rules, selects an appropriate rule as a sample rule, and then proceeds to step S302, specifically, please refer to fig. 4, and in this embodiment, step S302 is performed to obtain, according to the actual welding defect type, a weight coefficient of each evaluation factor for the actual welding defect type based on a hierarchical analysis method, which specifically includes:
s401, establishing a pair comparison matrix of the evaluation factors according to the actual welding defect type;
s402, carrying out consistency test on the pair of comparison matrixes, and adjusting the pair of comparison matrixes according to a test result;
s403, obtaining the feature vector of the adjusted paired comparison matrix according to the adjusted paired comparison matrix;
s404, carrying out normalization processing on the characteristic vector to obtain a weight coefficient of each evaluation factor for the actual welding defect type.
The present invention also provides a more specific embodiment for explaining the above processes S401 to S404:
according to the actual welding type, determining the relative importance of each evaluation factor by a field expert, and comparing every two, wherein the comparison adopts the relative comparison 1-5 scale shown in the following table:
relative comparison scale
The pair comparison matrix A is obtained by using the relative comparison scale as follows:
the calculation of each element in matrix a is calculated according to the following formula:
where i, j, k=0, 1,2,3,..n.
Obtaining the maximum characteristic root lambda according to the obtained paired comparison matrix A max The corresponding feature vector is taken as a weight vector ω, i.e. aω=λ max . And carrying out normalization processing on the weight vector to obtain a final weight coefficient.
Because different evaluation factors have different influences on different types of defect characteristics, the process needs to be flexibly set according to the actual welding defect types, in the embodiment, the welding defect causes are compared with the importance of the different evaluation factors by taking a 'welding through' defect as an example, and a pair comparison matrix A is obtained as follows:
after the comparison matrix is obtained, consistency test is needed to be carried out to verify rationality, and the comparison process can be carried out by calculating a consistency index CR:
The weight distribution of the established paired comparison matrix to the related parameters is reasonable, the reliability is high, and the weight coefficient can be calculated. In this embodiment, the final weight coefficient vector β of "penetration" obtained according to the above matrix is:
β＝(0.25,0.05,0.05,0.06,0.09,0.13,0.13,0.09,0.09,0.06)
above-mentionedThe process is only a weight coefficient of one actual welding defect type, and in practice, different welding defect types all need to be purposefully and independently established into a pair comparison matrix and calculated into the weight coefficient. In the present embodiment, there are ten influencing factors C i Corresponding to four different types of defects S i The final defect weight coefficient table is obtained as shown in the following table.
Defect weight coefficient table
After determining the weight coefficients, step S303 and step S304 may be performed, where membership degrees of each fact are determined by using membership functions according to the evaluation factor data, and specific processes thereof are in the prior art, which are not described herein. Then step S304 is performed, in this embodiment, the actual credibility lambda of each fuzzy diagnosis rule g Can be calculated by the following formula:
for the convenience of understanding, the present invention also provides a specific embodiment for explaining the above-mentioned actual credibility calculation process. The evaluation factor data acquired in this embodiment are: welding method M "pulse arc welding", defect relative center distance l 1 "3mm", distance between two defects/ 2 "1.2mm", defect aspect ratio δ "1.1", plate thickness b "6mm", welding current I "142A", welding voltage U "19.8V", welding speed V h "0.36m/min", wire feed speed V s "9.0m/min", gas flow V q "25L". The actual welding defect type is 'air hole', and the weight coefficient S of each evaluation factor for the defects such as 'air hole' can be known through the process 3 0.25,0.05,0.05,0.06,0.08,0.06,0.06,0.13,0.13,0.13, the actual credibility lambda of the three fuzzy diagnostic rules (namely rule 1, rule 2 and rule 3 in the previous table) can be obtained g1 、λ g2 、λ g3 ：
λ g1 ＝0.95×(0.25×1+0.05×1+0.05×1+0.06×1+0.08×1+0.06×0.9+0.06×0.9+0.13×1+0.13×1+0.13×1)＝0.9386
λ g2 ＝0.95×(0.25×1+0.05×1+0.05×1+0.06×1+0.08×1+0.06×0.9+0.06×0.9+0.13×1+0.13×1+0.13×0)＝0.8151
λ g3 ＝0.90×(0.25×1+0.05×1+0.05×1+0.06×1+0.08×1+0.06×0.9+0.06×0.9+0.13×0.5+0.13×1+0.13×0.5)＝0.7722
After calculating the actual reliability, step S105 may be performed to determine whether the sample rule with the actual reliability greater than the preset threshold exists, if yes, a target diagnosis result is obtained according to a defect cause of the sample rule with the actual reliability greater than the preset threshold, if no, a target diagnosis result is obtained according to the neural network model, and in this embodiment, the step specifically includes:
sequencing the sample rule according to the numerical value of the actual credibility to obtain a sample rule sequence;
Judging whether the actual credibility of the sample rule is larger than the corresponding preset threshold value, if so, selecting the defect causes of the set number of the sample rules with the highest actual credibility according to the sample rule sequence to obtain the target diagnosis result, and if not, inputting the evaluation factor data and the actual welding defect type into the cause diagnosis neural network model to obtain the target diagnosis result.
In this embodiment, the three actual credibilities are compared with the preset threshold λ of each rule to obtain the integrated credibility λ of rule 1 g1 0.9386 is greater than a threshold λ=0.88, λ for the remaining two rules g And is smaller than lambda, so that only rule 1 is activated and used, and finally, the cause of the defect with diagnosis can be obtained according to rule 1, namely, the gas flow is possibly overlarge and the reliability is 0.9386.
It will be appreciated that in this embodiment, each will beThe actual credibility is compared with the preset threshold value of the corresponding rule, and the same preset threshold value can be uniformly set for comparison in practice. When lambda is present g And when the actual credibility of a plurality of fuzzy diagnosis rules is larger than a preset threshold value after all the rules are searched, a set number (for example three) fuzzy diagnosis rules can be selected from the rules to serve as final diagnosis results.
And when the comprehensive credibility lambdag of all the rules is smaller than the threshold lambdag, the existing fuzzy diagnosis rules are not credible, and then the neural network model can be further selected for reasoning. The input characteristics are normalized and then are led into a model, the neural network model for cause diagnosis obtained through training is utilized for reasoning, the network model outputs a reasonable welding defect cause in a specific form (such as probability of each defect cause), and then a user evaluates the model output result to finally obtain a defect cause diagnosis scheme or a model correction report.
In order to verify the application effect of the method in actual production, 5 test samples are randomly selected in the production of the enterprise tank body, and the test samples are shown in the following table.
Test sample
In the test, the system gives the reasoning results according to the fuzzy rule for the samples 2 to 5, as shown in the following table, wherein the defect cause true value is the actually acquired evaluation factor data in the previous text, and the rule reasoning credibility is the actual credibility in the previous text:
defect cause true value and rule reasoning result
Whereas for sample 1, there is a rule integrated confidence lambda g Under the condition of being lower than a preset threshold lambda, the auxiliary reasoning is carried out by adopting a causal diagnosis neural network model, and the true value of the defect causal value and the network output result are shown in the following table:
Defect cause true value and model output result
Taking the diagnosis of aluminum alloy welding air hole defects (sample 1) in enterprises as an example, it can be seen that by the auxiliary diagnosis of the self-adaptive PSO-BP neural network model based on dynamic weight, it is inferred that 99.51% of input air hole defects are caused by overlarge welding speed, 0.38% of input air hole defects are caused by overlarge gas flow, and 0.11% of input air hole defects are caused by overlarge gas flow.
In order to better implement the welding defect cause diagnosis method according to the embodiment of the present invention, referring to fig. 5 correspondingly, fig. 5 is a schematic structural diagram of an embodiment of a welding defect cause diagnosis system according to the present invention, where the welding defect cause diagnosis system 500 includes:
a rule establishing module 510, configured to determine an evaluation factor and a defect cause, and establish a plurality of fuzzy diagnostic rules for characterizing a fuzzy relationship between the evaluation factor and the defect cause based on a fuzzy rule of an extension precursor;
a network establishment module 520 configured to establish a cause diagnosis neural network model, wherein input parameters of the cause diagnosis neural network model include a welding defect type and the evaluation factor, and output parameters of the cause diagnosis neural network model include a diagnosis result;
The data obtaining module 530 is configured to obtain actual welding defect type and evaluation factor data corresponding to the defect to be diagnosed;
the fuzzy inference module 540 is configured to select a matched fuzzy diagnosis rule from a plurality of fuzzy diagnosis rules as a sample rule based on the evaluation factor data, and perform fuzzy inference on the sample rule to obtain an actual credibility of each sample rule;
and a result diagnosis module 550, configured to determine whether the sample rule with the actual reliability greater than the preset threshold exists, if yes, obtain a target diagnosis result according to a defect cause of the sample rule with the actual reliability greater than the preset threshold, and if not, obtain a target diagnosis result according to the neural network model.
What needs to be explained here is: the corresponding system 500 provided in the foregoing embodiments may implement the technical solutions described in the foregoing method embodiments, and the specific implementation principles of the foregoing modules or units may be referred to the corresponding content in the foregoing method embodiments, which is not repeated herein.
In addition, the present invention also provides an embodiment of a welding defect cause diagnosis system, the system comprising:
A database module: and the functions of searching and inquiring the data of the materials, the process and the defect identification characteristics required by welding are performed, so that the storage cost of the data and the time cost of inquiring the data are saved. Meanwhile, the authority is given to a system manager so that the system manager can increase, decrease and maintain data in the system;
defect identification module: the user can upload the X-ray film in the system, and the system can identify the X-ray film by adopting a convolution neural network reconstructed by super resolution so as to acquire relevant characteristics of welding defects, such as defect types, areas, length-width ratios and the like;
defect cause diagnosis module: combining the data in the database and the defect identification characteristics, performing cause diagnosis on the welding defect, and determining the welding defect cause with the highest confidence by using the welding defect cause diagnosis method;
and a process optimization module: the system designed in the method comprises two optimization modes, wherein the mode I starts from the defect cause diagnosis level, and after the defect causes are obtained, a process optimization scheme is given according to the defect causes to obtain a defect-free welding line; starting from the weld performance, performing multi-objective optimization on the weld performance by utilizing a multi-objective particle swarm optimization (MOPSO), and regulating and controlling the technological parameters to obtain weld performance indexes meeting industrial requirements;
The user authority management module: for the security of the database, the user rights are classified into a system administrator, a system operator, and a general user. A system administrator can use a user authority management module to assign relevant authorities to all users in the system; the system operator can increase, decrease and maintain the database; the ordinary user can only search and review the database, and cannot perform other operations on the database.
Further, referring to fig. 6, fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present application. Based on the above-mentioned welding defect cause diagnosis method, the present application also provides a welding defect cause diagnosis device 600, that is, the above-mentioned electronic device, where the welding defect cause diagnosis device 600 may be a mobile terminal, a desktop computer, a notebook computer, a palm computer, a server, and other computing devices. The welding defect cause diagnosis apparatus 600 includes a processor 610, a memory 620, and a display 630. Fig. 6 shows only a part of the components of the welding defect cause diagnosis device, but it should be understood that not all the shown components are required to be implemented, and more or fewer components may be implemented instead.
The memory 620 may be an internal storage unit of the welding defect cause diagnosis device 600, such as a hard disk or a memory of the welding defect cause diagnosis device 600, in some embodiments. The memory 620 may also be an external storage device of the welding defect cause diagnosis device 600 in other embodiments, for example, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash Card (Flash Card) or the like, which are provided on the welding defect cause diagnosis device 600. Further, the memory 620 may also include both an internal memory unit and an external memory device of the welding defect cause diagnosis device 600. The memory 620 is used to store application software installed in the welding defect cause diagnosis apparatus 600 and various kinds of data, such as program codes for installing the welding defect cause diagnosis apparatus 600. The memory 620 may also be used to temporarily store data that has been output or is to be output. In one embodiment, the memory 620 stores a welding defect cause diagnosis program 640, and the welding defect cause diagnosis program 640 is executable by the processor 610 to implement the welding defect cause diagnosis method according to the embodiments of the present application.
The processor 610 may be a central processing unit (Central Processing Unit, CPU), microprocessor or other data processing chip in some embodiments for executing program code or processing data stored in the memory 620, such as performing weld defect cause diagnostics methods, etc.
The display 630 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch, or the like in some embodiments. The display 630 is used to display information at the welding defect cause diagnosis device 600 and to display a visual user interface. The components 610-630 of the weld defect cause diagnostic device 600 communicate with each other over a system bus.
In one embodiment, the steps in the weld defect cause diagnosis method as described above are implemented when the processor 610 executes the weld defect cause diagnosis program 640 in the memory 620.
The invention provides a welding defect cause diagnosis method, a system and electronic equipment, wherein the method comprises the steps of establishing a fuzzy diagnosis rule and a cause diagnosis neural network database, combining the fuzzy diagnosis rule and the cause diagnosis neural network database to diagnose the welding defect, specifically, performing prior expansion based on the traditional fuzzy rule, adding a threshold value, performing fuzzy reasoning by using a sample rule selected from the fuzzy diagnosis rule, obtaining a target diagnosis result according to the sample rule if the sample rule with actual credibility being greater than a preset threshold value exists, and considering that the existing fuzzy diagnosis rule cannot process the existing fact if the sample rule does not exist, and performing complementary processing by utilizing the cause diagnosis neural network at the moment, thereby obtaining an ideal target diagnosis result. Compared with the prior art, the fuzzy rule and the neural network are combined, so that the diagnosis process has the advantages of fuzzy reasoning and uncertainty knowledge processing, the neural network self-learning and high efficiency and high precision, the reliability of the reasoning process is greatly improved, and the fuzzy neural network model is more suitable for welding the scene of multiple influencing factors compared with the traditional fuzzy neural network model.
The present invention is not limited to the above-mentioned embodiments, and any changes or substitutions that can be easily understood by those skilled in the art within the technical scope of the present invention are intended to be included in the scope of the present invention.
Claims (7)
1. A welding defect cause diagnosis method, comprising:
determining an evaluation factor and a defect cause, and establishing a plurality of fuzzy diagnosis rules for representing fuzzy relations between the evaluation factor and the defect cause based on fuzzy rules of an expansion front;
establishing a cause diagnosis neural network model, wherein input parameters of the cause diagnosis neural network model comprise welding defect types and evaluation factors, and output parameters of the cause diagnosis neural network model comprise diagnosis results;
acquiring actual welding defect type and evaluation factor data corresponding to defects to be diagnosed;
based on the evaluation factor data, selecting a matched fuzzy diagnosis rule from a plurality of fuzzy diagnosis rules as a sample rule, and performing fuzzy reasoning on the sample rule to obtain the actual credibility of each sample rule;
Judging whether the sample rule with the actual credibility larger than a preset threshold exists or not, if so, obtaining a target diagnosis result according to the defect cause of the sample rule with the actual credibility larger than the preset threshold, and if not, obtaining the target diagnosis result according to the cause diagnosis neural network model;
wherein the determining the evaluation factor and the defect cause, establishing a plurality of fuzzy diagnosis rules for representing fuzzy relations between the evaluation factor and the defect cause based on the fuzzy rules of the expansion foreterm, comprises the following steps:
establishing a welding characteristic information base, and obtaining the evaluation factors and the defect causes according to the welding characteristic information base;
establishing a plurality of fuzzy diagnosis rules based on fuzzy rules of the extension antecedents according to the evaluation factors;
each fuzzy diagnosis rule comprises a front term, a rear term, theoretical credibility and the preset threshold value, wherein the front term comprises a plurality of evaluation factors, and the rear term comprises the defect cause;
the fuzzy diagnosis rule is designed by adopting a C-F model based on the fuzzy rule, and the antecedents of the fuzzy diagnosis rule are expanded, so that the membership degree of each condition is increased;
The step of selecting a matched fuzzy diagnosis rule from a plurality of fuzzy diagnosis rules as a sample rule based on the evaluation factor data, and performing fuzzy reasoning on the sample rule to obtain the actual credibility of each sample rule, wherein the step of obtaining the actual credibility of each sample rule comprises the following steps:
selecting a matched fuzzy diagnosis rule from a plurality of diagnosis rules as the sample rule based on the evaluation factor data;
according to the actual welding defect type, obtaining a weight coefficient of each evaluation factor for the actual welding defect type based on an analytic hierarchy process;
obtaining the membership degree of each evaluation factor in the sample rule according to the evaluation factor data;
and calculating the actual credibility of each sample rule according to the weight coefficient and the membership degree.
2. The welding defect cause diagnosis method according to claim 1, wherein the obtaining a weight coefficient of each evaluation factor for the actual welding defect type based on a hierarchical analysis method according to the actual welding defect type comprises:
establishing a pair comparison matrix of the evaluation factors according to the actual welding defect type;
Consistency test is carried out on the paired comparison matrixes, and the paired comparison matrixes are adjusted according to test results;
obtaining the feature vector of the pair of adjusted comparison matrixes according to the pair of adjusted comparison matrixes;
and normalizing the feature vector to obtain a weight coefficient of each evaluation factor for the actual welding defect type.
3. The method according to claim 1, wherein the determining whether the sample rule having the actual reliability greater than the preset threshold exists, if so, obtaining a target diagnosis result according to a defect cause of the sample rule having the actual reliability greater than the preset threshold, and if not, obtaining a target diagnosis result according to the cause diagnosis neural network model, includes:
sequencing the sample rule according to the numerical value of the actual credibility to obtain a sample rule sequence;
judging whether the actual credibility of the sample rule is larger than the corresponding preset threshold value, if so, selecting the defect causes of the set number of the sample rules with the highest actual credibility according to the sample rule sequence to obtain the target diagnosis result, and if not, inputting the evaluation factor data and the actual welding defect type into the cause diagnosis neural network model to obtain the target diagnosis result.
4. The welding defect cause diagnosis method according to claim 1, wherein the establishing a cause diagnosis neural network model includes:
establishing an initial BP neural network model, wherein the input parameters of the initial BP neural network model comprise a plurality of evaluation factors, and the output parameters of the initial BP neural network model comprise the diagnosis results;
acquiring initial parameters of the initial BP neural network model, and optimizing through a PSO-BP algorithm to obtain optimized parameters;
and optimizing the initial BP neural network model according to the optimization parameters to obtain the causal diagnosis neural network model.
5. The welding defect cause diagnosis method according to claim 4, wherein the obtaining initial parameters of the initial BP neural network model and optimizing by a PSO-BP algorithm to obtain optimized parameters comprises:
acquiring initial parameters of the initial BP neural network model, wherein the initial parameters comprise weights, thresholds and loss functions of neurons;
establishing a solution space and a particle population according to the weight and the threshold of the neuron, and initializing the position of the particle population in the solution space;
Establishing an adaptability function according to the loss function;
optimizing the position of the particle population in the solution space according to the fitness function;
and acquiring the position of the optimized particle population in the solution space to obtain the optimization parameters.
6. A welding defect cause diagnosis system, comprising:
the rule establishing module is used for determining an evaluation factor and a defect cause, and establishing a plurality of fuzzy diagnosis rules for representing fuzzy relations between the evaluation factor and the defect cause based on fuzzy rules of an expansion front term;
the network establishment module is used for establishing a cause diagnosis neural network model, wherein the input parameters of the cause diagnosis neural network model comprise welding defect types and evaluation factors, and the output parameters of the cause diagnosis neural network model comprise diagnosis results;
the data acquisition module is used for acquiring the actual welding defect type and evaluation factor data corresponding to the defect to be diagnosed;
the fuzzy reasoning module is used for selecting a matched fuzzy diagnosis rule from a plurality of fuzzy diagnosis rules as a sample rule based on the evaluation factor data, and carrying out fuzzy reasoning on the sample rule to obtain the actual credibility of each sample rule;
The result diagnosis module is used for judging whether the sample rule with the actual credibility being larger than a preset threshold exists or not, if yes, obtaining a target diagnosis result according to the defect cause of the sample rule with the actual credibility being larger than the preset threshold, and if not, obtaining the target diagnosis result according to the neural network model;
wherein the determining the evaluation factor and the defect cause, establishing a plurality of fuzzy diagnosis rules for representing fuzzy relations between the evaluation factor and the defect cause based on the fuzzy rules of the expansion foreterm, comprises the following steps:
establishing a welding characteristic information base, and obtaining the evaluation factors and the defect causes according to the welding characteristic information base;
establishing a plurality of fuzzy diagnosis rules based on fuzzy rules of the extension antecedents according to the evaluation factors;
each fuzzy diagnosis rule comprises a front term, a rear term, theoretical credibility and the preset threshold value, wherein the front term comprises a plurality of evaluation factors, and the rear term comprises the defect cause;
the fuzzy diagnosis rule is designed by adopting a C-F model based on the fuzzy rule, and the antecedents of the fuzzy diagnosis rule are expanded, so that the membership degree of each condition is increased;
The step of selecting a matched fuzzy diagnosis rule from a plurality of fuzzy diagnosis rules as a sample rule based on the evaluation factor data, and performing fuzzy reasoning on the sample rule to obtain the actual credibility of each sample rule, wherein the step of obtaining the actual credibility of each sample rule comprises the following steps:
selecting a matched fuzzy diagnosis rule from a plurality of diagnosis rules as the sample rule based on the evaluation factor data;
according to the actual welding defect type, obtaining a weight coefficient of each evaluation factor for the actual welding defect type based on an analytic hierarchy process;
obtaining the membership degree of each evaluation factor in the sample rule according to the evaluation factor data;
and calculating the actual credibility of each sample rule according to the weight coefficient and the membership degree.
7. An electronic device comprising a memory and a processor, wherein,
the memory is used for storing programs;
the processor, coupled to the memory, is configured to execute the program stored in the memory to implement the steps in the welding defect cause diagnosis method according to any one of the preceding claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211323505.2A CN115563549B (en) | 2022-10-27 | 2022-10-27 | Welding defect cause diagnosis method and system and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211323505.2A CN115563549B (en) | 2022-10-27 | 2022-10-27 | Welding defect cause diagnosis method and system and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115563549A CN115563549A (en) | 2023-01-03 |
| CN115563549B true CN115563549B (en) | 2023-10-20 |
Family
ID=84769177
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211323505.2A Active CN115563549B (en) | 2022-10-27 | 2022-10-27 | Welding defect cause diagnosis method and system and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115563549B (en) |
Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5544256A (en) * | 1993-10-22 | 1996-08-06 | International Business Machines Corporation | Automated defect classification system |
| US5591355A (en) * | 1994-02-25 | 1997-01-07 | Miyachi Technos Corporation | Method for controlling resistance welding using fuzzy reasoning |
| KR20040088896A (en) * | 2003-04-14 | 2004-10-20 | 기아자동차주식회사 | Arc welding network system |
| WO2005071582A2 (en) * | 2004-01-27 | 2005-08-04 | British Telecommunications Public Limited Company | Detection of abnormal behaviour in dynamic systems |
| CN102436524A (en) * | 2011-10-19 | 2012-05-02 | 清华大学 | Fuzzy reasoning method for soft fault diagnosis for analog circuit |
| CN103091112A (en) * | 2013-01-31 | 2013-05-08 | 林惠堂 | Method and device of car emission fault detection and diagnosis based on fuzzy reasoning and self-learning |
| CN105787563A (en) * | 2014-12-18 | 2016-07-20 | 中国科学院沈阳自动化研究所 | Self-learning mechanism-base fast matching fuzzy reasoning method |
| CN106909727A (en) * | 2017-02-20 | 2017-06-30 | 武汉理工大学 | Laser welding temperature field Finite Element Method based on BP neural network and Genetic Algorithms |
| CN106919982A (en) * | 2017-03-20 | 2017-07-04 | 中国科学院沈阳自动化研究所 | A kind of method for diagnosing faults towards semiconductor manufacturing equipment |
| CN106961249A (en) * | 2017-03-17 | 2017-07-18 | 广西大学 | A kind of diagnosing failure of photovoltaic array and method for early warning |
| CN108205727A (en) * | 2016-12-20 | 2018-06-26 | 中国科学院沈阳自动化研究所 | A kind of digitlization plant process decision-making technique based on decision tree and expert system |
| CN109491816A (en) * | 2018-10-19 | 2019-03-19 | 中国船舶重工集团公司第七六研究所 | Knowledge based engineering method for diagnosing faults |
| CN112068542A (en) * | 2020-06-30 | 2020-12-11 | 武汉乐庭软件技术有限公司 | Automatic driving behavior planning method based on fuzzy control |
| CN112101597A (en) * | 2020-10-14 | 2020-12-18 | 辽宁电能发展股份有限公司 | Electric vehicle leasing operation platform vehicle fault pre-judging system, method and device |
| CN113139514A (en) * | 2021-05-14 | 2021-07-20 | 安徽三禾一信息科技有限公司 | Hydraulic system fault diagnosis method based on fuzzy neural network |
| CN114184367A (en) * | 2021-11-29 | 2022-03-15 | 北京唐智科技发展有限公司 | Fault diagnosis method, device and equipment for rotary mechanical equipment and readable storage medium |
| CN114361536A (en) * | 2022-03-18 | 2022-04-15 | 北汽福田汽车股份有限公司 | Fault processing method and device for fuel cell system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5421913B2 (en) * | 2007-08-03 | 2014-02-19 | スマートシグナル・コーポレーション | Fuzzy classification method for fault pattern matching cross-reference for related applications |
-
2022
- 2022-10-27 CN CN202211323505.2A patent/CN115563549B/en active Active
Patent Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5544256A (en) * | 1993-10-22 | 1996-08-06 | International Business Machines Corporation | Automated defect classification system |
| US5591355A (en) * | 1994-02-25 | 1997-01-07 | Miyachi Technos Corporation | Method for controlling resistance welding using fuzzy reasoning |
| KR20040088896A (en) * | 2003-04-14 | 2004-10-20 | 기아자동차주식회사 | Arc welding network system |
| WO2005071582A2 (en) * | 2004-01-27 | 2005-08-04 | British Telecommunications Public Limited Company | Detection of abnormal behaviour in dynamic systems |
| CN102436524A (en) * | 2011-10-19 | 2012-05-02 | 清华大学 | Fuzzy reasoning method for soft fault diagnosis for analog circuit |
| CN103091112A (en) * | 2013-01-31 | 2013-05-08 | 林惠堂 | Method and device of car emission fault detection and diagnosis based on fuzzy reasoning and self-learning |
| CN105787563A (en) * | 2014-12-18 | 2016-07-20 | 中国科学院沈阳自动化研究所 | Self-learning mechanism-base fast matching fuzzy reasoning method |
| CN108205727A (en) * | 2016-12-20 | 2018-06-26 | 中国科学院沈阳自动化研究所 | A kind of digitlization plant process decision-making technique based on decision tree and expert system |
| CN106909727A (en) * | 2017-02-20 | 2017-06-30 | 武汉理工大学 | Laser welding temperature field Finite Element Method based on BP neural network and Genetic Algorithms |
| CN106961249A (en) * | 2017-03-17 | 2017-07-18 | 广西大学 | A kind of diagnosing failure of photovoltaic array and method for early warning |
| CN106919982A (en) * | 2017-03-20 | 2017-07-04 | 中国科学院沈阳自动化研究所 | A kind of method for diagnosing faults towards semiconductor manufacturing equipment |
| CN109491816A (en) * | 2018-10-19 | 2019-03-19 | 中国船舶重工集团公司第七六研究所 | Knowledge based engineering method for diagnosing faults |
| CN112068542A (en) * | 2020-06-30 | 2020-12-11 | 武汉乐庭软件技术有限公司 | Automatic driving behavior planning method based on fuzzy control |
| CN112101597A (en) * | 2020-10-14 | 2020-12-18 | 辽宁电能发展股份有限公司 | Electric vehicle leasing operation platform vehicle fault pre-judging system, method and device |
| CN113139514A (en) * | 2021-05-14 | 2021-07-20 | 安徽三禾一信息科技有限公司 | Hydraulic system fault diagnosis method based on fuzzy neural network |
| CN114184367A (en) * | 2021-11-29 | 2022-03-15 | 北京唐智科技发展有限公司 | Fault diagnosis method, device and equipment for rotary mechanical equipment and readable storage medium |
| CN114361536A (en) * | 2022-03-18 | 2022-04-15 | 北汽福田汽车股份有限公司 | Fault processing method and device for fuel cell system |
Non-Patent Citations (4)
| Title |
|---|
| Essa Alghannam 等.A Novel Method of Using Vision System and Fuzzy Logic for Quality Estimation of Resistance Spot Welding.《symmetry》.2019,1-20. * |
| 吕现彪.模糊关联规则挖掘算法的研究.《中国优秀硕士学位论文全文数据库 信息科技辑》.2017,(第02期),I138-2485. * |
| 梁硼 等.X射线焊缝图像缺陷自动提取与识别技术研究.《中国优秀硕士学位论文全文数据库 信息科技辑》.2013,(第04期),I138-932. * |
| 高昶霖 等.基于动态权重的自适应 PSO-BP 神经网络焊接缺陷成因诊断.《焊接学报》.2022,第43卷(第1期),98-106. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115563549A (en) | 2023-01-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111814956B (en) | Multi-task learning air quality prediction method based on multi-dimensional secondary feature extraction | |
| WO2021004324A1 (en) | Resource data processing method and apparatus, and computer device and storage medium | |
| CN111898867B (en) | Airplane final assembly production line productivity prediction method based on deep neural network | |
| CN116307215A (en) | Load prediction method, device, equipment and storage medium of power system | |
| WO2023226259A1 (en) | Hyper-parameter search method and apparatus, computer device and storage medium | |
| CN116109004A (en) | Insulator leakage current prediction method, device, equipment and medium | |
| CN116542701A (en) | Carbon price prediction method and system based on CNN-LSTM combination model | |
| CN111709585A (en) | Air conditioner load prediction method and device and storage medium | |
| CN115563549B (en) | Welding defect cause diagnosis method and system and electronic equipment | |
| CN113959071B (en) | Centralized water chilling unit air conditioning system operation control optimization method based on machine learning assistance | |
| CN112700112B (en) | RPA flow adjustment method and device, electronic equipment and storage medium | |
| CN109829115B (en) | Search engine keyword optimization method | |
| CN115545960B (en) | Electronic information data interaction system and method | |
| CN116861256A (en) | Furnace temperature prediction method, system, equipment and medium for solid waste incineration process | |
| CN117113086A (en) | Energy storage unit load prediction method, system, electronic equipment and medium | |
| CN114186771B (en) | Mixed regularized random configuration network industrial process operation index estimation method | |
| CN115994674A (en) | Scheduling management method based on digital twin comprehensive energy system | |
| CN114926075A (en) | Mechanical part production scheduling method based on man-hour prediction | |
| Chand et al. | An Optimal Disassembly Sequence Planning for Complex Products using Enhanced Deep Reinforcement Learning Framework | |
| Zhang et al. | Predicting Future Event via Small Data (eg, 4 Data) by ASF and Curve Fitting Methods | |
| CN114492174A (en) | Full life cycle shield tunneling parameter prediction method and device | |
| CN110942149B (en) | Feature variable selection method based on information change rate and condition mutual information | |
| CN112232557A (en) | Switch machine health degree short-term prediction method based on long-term and short-term memory network | |
| CN110807599A (en) | Method, device, server and storage medium for deciding electrochemical energy storage scheme | |
| CN112825269B (en) | Gasification process design method, device, equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |