CN117195743A - Spraying parameter optimization method for crack structure of thermal barrier coating - Google Patents
Spraying parameter optimization method for crack structure of thermal barrier coating Download PDFInfo
- Publication number
- CN117195743A CN117195743A CN202311334362.XA CN202311334362A CN117195743A CN 117195743 A CN117195743 A CN 117195743A CN 202311334362 A CN202311334362 A CN 202311334362A CN 117195743 A CN117195743 A CN 117195743A
- Authority
- CN
- China
- Prior art keywords
- thermal barrier
- coating
- barrier coating
- image
- parameters
- 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
- 238000005507 spraying Methods 0.000 title claims abstract description 66
- 239000012720 thermal barrier coating Substances 0.000 title claims abstract description 58
- 238000000034 method Methods 0.000 title claims abstract description 44
- 238000005457 optimization Methods 0.000 title claims abstract description 17
- 239000011248 coating agent Substances 0.000 claims abstract description 66
- 238000000576 coating method Methods 0.000 claims abstract description 66
- 238000012549 training Methods 0.000 claims abstract description 35
- 239000007921 spray Substances 0.000 claims abstract description 19
- 238000002474 experimental method Methods 0.000 claims abstract description 10
- 230000006870 function Effects 0.000 claims description 44
- 238000009826 distribution Methods 0.000 claims description 18
- 208000037170 Delayed Emergence from Anesthesia Diseases 0.000 claims description 8
- 230000004913 activation Effects 0.000 claims description 8
- 238000005070 sampling Methods 0.000 claims description 8
- 238000003860 storage Methods 0.000 claims description 3
- 239000013598 vector Substances 0.000 claims description 3
- 230000008485 antagonism Effects 0.000 claims description 2
- 239000010410 layer Substances 0.000 description 32
- 230000000694 effects Effects 0.000 description 6
- 230000009471 action Effects 0.000 description 4
- 239000000919 ceramic Substances 0.000 description 4
- 238000013135 deep learning Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 238000013507 mapping Methods 0.000 description 3
- 238000002360 preparation method Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000004088 simulation Methods 0.000 description 3
- 238000002485 combustion reaction Methods 0.000 description 2
- 238000005328 electron beam physical vapour deposition Methods 0.000 description 2
- 238000009413 insulation Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 239000012790 adhesive layer Substances 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000004888 barrier function Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000001816 cooling Methods 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 230000007123 defense Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000001000 micrograph Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000007750 plasma spraying Methods 0.000 description 1
- 239000000843 powder Substances 0.000 description 1
- 239000000758 substrate Substances 0.000 description 1
Abstract
The invention discloses a spray parameter optimization method of a crack structure of a thermal barrier coating, which relates to the technical field of thermal barrier coating spray, and comprises the following steps: obtaining a real coating microstructure image of the thermal barrier coating containing surface cracks, training an improved condition generation type countermeasure network model according to a plurality of groups of spraying parameters and the real coating microstructure image, adopting a loss function based on Wasserstein distance and gradient punishment items, inputting a group of spraying parameters into a generator of the improved condition generation type countermeasure network model, obtaining coating performance information extracted from the generated image, determining coating performance information with optimal service life according to a service life experiment of the prepared thermal barrier coating containing surface cracks, and determining optimal coating performance information extracted from the generated image, thereby determining optimal spraying parameters of the thermal barrier coating crack structure. According to the method, the optimal spraying parameters are inverted by utilizing the image generated by the depth network, so that dependence on related professional knowledge is reduced.
Description
Technical Field
The invention relates to the technical field of thermal barrier coating spraying, in particular to a spraying parameter optimization method of a thermal barrier coating crack structure.
Background
The gas turbine is strategic high-technology equipment related to national energy and national defense safety, and the industrial level of one country is reflected in a concentrated way. The thermal barrier coating can effectively improve the working temperature of the hot end component of the gas turbine, thereby improving the combustion efficiency and reducing the running cost. The thermal barrier coating structure mainly comprises a substrate, an adhesive layer and a ceramic layer. Wherein the ceramic layer mainly plays a role of heat insulation and is the core of the thermal barrier coating. Methods for preparing the ceramic layer mainly comprise methods such as Atmospheric Plasma Spraying (APS), electron beam physical vapor deposition (EB-PVD) and the like. Wherein, the preparation cost of APS is lower, which is the first choice process for preparing ceramic layer. With the development of advanced aero-engine and gas turbine technologies in recent years, the working temperatures of hot end components such as a combustion chamber, a high-temperature blade and the like are continuously improved, and the requirements on the heat insulation effect and the service life of the thermal barrier coating are higher, so that the conventional APS coating cannot meet the requirements. Thermal barrier coatings (DVCs) containing surface cracks have been proposed for this purpose to produce high thermal barrier and high strain tolerance thermal barrier coatings. It was found that surface crack preparation can be achieved by controlling key spray parameters during APS spraying to vary the in-plane equibiaxial tensile stress level produced by rapid cooling of the molten powder. The DVCs coating life is 3-4 times longer than the conventional APS coating life at the same service temperature. How to determine the optimal spray parameters, establishing the relationship between the spray parameters and the coating microstructure is therefore a key issue.
The traditional way of studying the relationship between the spray parameters and the microstructure of the coating mainly comprises: orthogonal experimentation, theoretical modeling, and numerical modeling. However, the orthogonal test requires a series of orthogonal tests, which is costly and cannot traverse all values. The physical mechanism in the theoretical modeling and the numerical simulation needs to be accurately mastered, but the control mechanism behind the theoretical modeling and the numerical simulation is very complex in the face of the combined action of a plurality of spraying parameters, difficult to determine, difficult to directly establish the mapping relation from the preparation parameters to the structure, and meanwhile, the calculation cost of the two methods is high, and no accurate theoretical modeling and numerical simulation method for the problem exists at present.
Disclosure of Invention
The present invention aims to solve at least one of the technical problems existing in the prior art. Therefore, the first aspect of the invention provides a spraying parameter optimization method for a crack structure of a thermal barrier coating, which comprises the following steps:
acquiring a real coating microstructure image of the thermal barrier coating containing the surface cracks, which is prepared according to preset spraying parameters;
generating an countermeasure network model according to a plurality of groups of spraying parameters and improved conditions of real coating microstructure image training, wherein a loss function based on Wasserstein distance and gradient penalty term is adopted during training;
inputting a group of spraying parameters into a trained generator of an improved condition generation type countermeasure network model to acquire coating performance information extracted from a generated image;
and determining optimal coating performance information of service life according to a service life experiment of the prepared thermal barrier coating containing the surface cracks, and reversely optimizing spraying parameters of a thermal barrier coating crack structure according to the optimal coating performance information.
Further, training an improved condition-generating countermeasure network model based on the plurality of sets of spray parameters and the image of the actual coating microstructure, comprising
Training an improved condition generation type countermeasure network model generator by adopting an m-dimensional random number one-dimensional tensor and an n-dimensional spraying parameter condition one-dimensional tensor to acquire a two-dimensional tensor of a generated image;
training an improved condition generation type countermeasure network model discriminator by adopting a two-dimensional tensor of a generated image and a real coating microstructure image and a conditional one-dimensional tensor of n-dimensional spraying parameters, and acquiring a probability one-dimensional tensor;
during training, parameters of the discriminator are optimized and updated by adopting a loss function based on Wasserstein distance and gradient penalty term, and then parameters of the generator are optimized and updated by adopting the loss function of the generator, until the loss functions of the generator and the discriminator are minimum and reach balance, and training is stopped.
Further, a loss function based on Wasserstein distance and gradient penalty term, comprising:
wherein, the method comprises the steps of, wherein,representing a discriminator loss function; />Representation->At->Obeys->The expectation of the function under the distribution condition;representation->At->Obeys->The expectation of the function under the distribution condition; />Representation->At->Obeys->The expectation of the function under the distribution condition; />Representing a condition parameter; />Representing a distribution of random number vectors in the generator; />Representing the generated image samples; />Representing the probability that the image output by the discriminator is true, and the probability is between 0 and 1; />Representing the distribution of real samples; />Representing a real image sample; />Is a weight coefficient of the gradient penalty term; />Expressed in real sample->And generate sample->Randomly interpolating and sampling the obtained images; />Representation->Is a random distribution of (a); />Representing a 2-norm; />Representing the calculated gradient; />Obeys [0,1 ]]Evenly distributed.
Further, a loss function of the generator, comprising:wherein->Representing the generator loss function.
Further, after the image of the true coating microstructure is acquired, the method further includes:
acquiring a plurality of intercepted pictures in a real coating microstructure image;
and marking a plurality of cut pictures by using a category label or a numerical label.
Further, the generator of the improved condition generating type countermeasure network model comprises 1 fully connected layer, 2 deconvolution layers, 6 residual blocks, 2 up-sampling layers and a LeakyReLU activation function layer which are connected in sequence, and the discriminator of the improved condition generating type countermeasure network model comprises 1 fully connected layer, 7 convolution layers, 1 residual block, 4 down-sampling layers, a LeakyReLU activation function layer and 2 fully connected layers which are connected in sequence.
Another aspect of the present invention provides a spray parameter optimization apparatus for a thermal barrier coating crack structure, comprising:
the image acquisition module is used for acquiring a real coating microstructure image of the thermal barrier coating with the surface cracks, which is prepared according to preset spraying parameters;
the training module is used for training the improved condition generation type countermeasure network model according to a plurality of groups of spraying parameters and the real coating microstructure image, and a loss function based on Wasserstein distance and gradient penalty items is adopted during training;
the coating performance generation module is used for inputting a group of spraying parameters in a trained improved condition generation type antagonism network model generator to acquire coating performance information extracted from a generated image;
the reverse optimization module is used for determining coating performance information with optimal service life according to service life experiments of the prepared thermal barrier coating containing the surface cracks, and reversely optimizing spraying parameters of a thermal barrier coating crack structure according to the optimal coating performance information.
In another aspect, the invention further provides an electronic device, which comprises a processor and a memory, wherein at least one instruction, at least one section of program, a code set or an instruction set is stored in the memory, and the at least one instruction, the at least one section of program, the code set or the instruction set is loaded and executed by the processor to realize the spraying parameter optimization method of the thermal barrier coating crack structure according to the first aspect.
In another aspect, the present invention also provides a computer readable storage medium, where at least one instruction, at least one program, a code set, or an instruction set is stored, where the at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by a processor to implement the method for optimizing a spraying parameter of the crack structure of the thermal barrier coating according to the first aspect.
Compared with the prior art, the spraying parameter optimization method for the crack structure of the thermal barrier coating has the following beneficial effects:
according to the method, the deep learning and image generation technology is utilized, the coating performance information with the optimal service life is determined only through analysis of the service life experiment for generating the image coating performance information on the premise that a physical rule is not required to be known or assumed, and the physical phenomenon can be predicted by utilizing the mapping relation between the spraying parameters determined by the deep learning and the coating crack structure, so that the prediction from the spraying parameters to the coating microstructure image end-to-end is realized, and the dependence on related professional knowledge is reduced.
Drawings
In order to more clearly illustrate the technical solution of the present invention, the following description will make a brief introduction to the drawings used in the description of the embodiments or the prior art. It should be apparent that the drawings in the following description are only some embodiments of the present invention, and that other drawings can be obtained from these drawings without inventive effort to those of ordinary skill in the art.
FIG. 1 is a flow chart of a spray parameter optimization method for a thermal barrier coating crack structure provided by an embodiment of the invention;
FIG. 2 is a generator model diagram of a method for optimizing spraying parameters of a crack structure of a thermal barrier coating, which is provided by an embodiment of the invention;
FIG. 3 is a model diagram of a discriminator of a spray parameter optimization method for a thermal barrier coating crack structure provided by an embodiment of the invention;
FIG. 4 is a graph of the predicted effect of a generator of a method for optimizing the spraying parameters of a crack structure of a thermal barrier coating according to an embodiment of the present invention;
fig. 5 is a structural diagram of a spraying parameter optimizing device for a crack structure of a thermal barrier coating, which is provided by the embodiment of the invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The present specification provides method operational steps as described in the examples or flowcharts, but may include more or fewer operational steps based on conventional or non-inventive labor. When implemented in a real system or server product, the methods illustrated in the embodiments or figures may be performed sequentially or in parallel (e.g., in a parallel processor or multithreaded environment).
Fig. 1 is a flowchart of a method for optimizing spraying parameters of a thermal barrier coating crack structure according to an embodiment of the present invention, as shown in fig. 1, and the method for optimizing spraying parameters of a thermal barrier coating crack structure includes:
step 101, obtaining a real coating microstructure image of the thermal barrier coating with surface cracks, which is prepared according to preset spraying parameters;
in step 101, thermal barrier coating structure images containing surface cracks under different spraying parameters are prepared by controlling the spraying parameters, and corresponding microstructure electron microscope images are shot by SEM scanning electron microscope.
Step 102, training improved conditions according to a plurality of groups of spraying parameters and a real coating microstructure image to generate an countermeasure network model, wherein a loss function based on Wasserstein distance and gradient penalty items is adopted during training;
in step 102, inputting any group of spraying parameters into the trained network model, comparing the microstructure image predicted by the network model with the real microstructure image to judge the predicted effect of the network model, otherwise, optimizing the configuration parameters of the network model until the optimal effect is reached.
Step 103, inputting a group of spraying parameters into a trained improved condition generation type countermeasure network model generator to obtain coating performance information extracted from a generated image;
in step 103, coating property information, such as crack density, etc., is extracted from the predicted image.
And 104, determining coating performance information with optimal service life according to a service life experiment of the prepared thermal barrier coating with the surface cracks, and reversely optimizing spraying parameters of a thermal barrier coating crack structure according to the optimal coating performance information.
In step 104, service life experiments are carried out on the prepared surface crack-containing coating, crack density with optimal service life is determined, and spraying parameters are reversely optimized according to the optimal crack density.
In summary, the method utilizes the deep learning and image generation technology, determines the coating performance information with optimal service life by analyzing the service life experiment for generating the image coating performance information on the premise of no need of knowing or assuming a physical rule, and utilizes the mapping relation between the spraying parameters determined by the deep learning and the coating crack structure to realize the prediction of the physical phenomenon, thereby realizing the prediction from the spraying parameters to the image end of the coating microstructure, reducing the dependence on related professional knowledge, guiding the industrial production practice and having stronger practical application value.
In one possible embodiment, training the improved condition-generating countermeasure network model from a plurality of sets of spray parameters and a real coating microstructure image, comprises:
training an improved condition generation type countermeasure network model generator by adopting an m-dimensional random number one-dimensional tensor and an n-dimensional spraying parameter condition one-dimensional tensor to acquire a two-dimensional tensor of a generated image;
training an improved condition generation type countermeasure network model discriminator by adopting a two-dimensional tensor of a generated image and a real coating microstructure image and a conditional one-dimensional tensor of n-dimensional spraying parameters, and acquiring a probability one-dimensional tensor;
during training, parameters of the discriminator are optimized and updated by adopting a loss function based on Wasserstein distance and gradient penalty term, and then parameters of the generator are optimized and updated by adopting the loss function of the generator, until the loss functions of the generator and the discriminator are minimum and reach balance, and training is stopped.
In the embodiment provided by the invention, the network model selects a condition generation type countermeasure network model CWGAN-GP which introduces Wasserstein distance and gradient penalty terms, and the network model comprises two parts, namely a generator and a discriminator. The generator consists of a linear layer, a convolution layer, a residual block layer, an up-sampling layer and a LeakyReLU activation function layer, wherein the input is a random number one-dimensional tensor of 100 dimensions and a conditional one-dimensional tensor of n dimensions, n represents the number of spraying parameters, and the output is a two-dimensional tensor of a generated image. The discriminator consists of a linear layer, a convolution layer, a residual block layer, a downsampling layer and a LeakyReLU activation function layer, inputs are a real image, a generated image two-dimensional tensor and an n-dimensional conditional one-dimensional tensor, outputs are probability one-dimensional tensors, the probability one-dimensional tensors are used for judging the probability that the image is true, the discriminator only exists in a training stage, the discriminator is discarded after training is finished, and a generator is reserved.
In one possible implementation, the loss function based on the wasperstein distance and gradient penalty term, comprises:
wherein->Representing a discriminator loss function; />Representation->At->Obeys->The expectation of the function under the distribution condition;representation->At->Obeys->The expectation of the function under the distribution condition; />Representation->At->Obeys->The expectation of the function under the distribution condition; />Representing a condition parameter; />Representing a distribution of random number vectors in the generator; />Representing the generated image samples; />Representing the probability that the image output by the discriminator is true, and the probability is between 0 and 1; />Representing the distribution of real samples; />Representing a real image sample; />Is a weight coefficient of the gradient penalty term; />Expressed in real sample->And generate sample->Randomly interpolating and sampling the obtained images; />Representation->Is a random distribution of (a); />Representing a 2-norm; />Representing the calculated gradient; />Obeys [0,1 ]]Evenly distributed.
In one possible implementation, the loss function of the generator comprises:
wherein->Representing the generator loss function.
In one possible embodiment, after the acquisition of the image of the actual coating microstructure, it further comprises:
acquiring a plurality of intercepted pictures in a real coating microstructure image;
and marking a plurality of cut pictures by using a category label or a numerical label.
In the embodiment provided by the invention, the obtained image is subjected to data enhancement processing, namely, a plurality of smaller images are intercepted from a larger original image so as to expand the scale of a data set, and the data set is marked by adopting two modes of category labels and numerical labels, wherein the category labels are used for classifying a group of data into one category, and the marking mode only can generate the images of the existing categories. The numerical value label refers to taking the actual numerical value as the label, and normalization processing is needed to be carried out on the data, so that the method can generate the existing data and can also predict other data.
In one possible implementation, as shown in fig. 2 and 3, the generator of the improved conditional generation type countermeasure network model includes 1 fully connected layer, 2 deconvolution layers, 6 residual blocks, 2 upsampling layers, and a LeakyReLU activation function layer connected in sequence.
In the embodiment provided by the invention, in order to enable the network model to adapt to the size of the surface crack microstructure image and achieve good effect, the corresponding network structure is adjusted, and the identifier of the improved condition generation type countermeasure network model comprises 1 full connection layer, 7 convolution layers, 1 residual block, 4 downsampling layers, a LeakyReLU activation function layer and 2 full connection layers which are connected in sequence.
The embodiment provides two spraying parameters, namely a data set of preheating times and spraying speed and a prediction effect.
The dataset in this embodiment contains seven sets of data, each set of data containing 20 sheets 25601370, cutting each image into 10 1024 +.>1024, a total of 1400 images, followed by scaling the image to 512 +.>512-size images are input into the network for training.
After training, inputting a plurality of groups of spraying parameters to obtain a predicted coating crack structure diagram as shown in fig. 4.
Another aspect of the present invention provides a spraying parameter optimization device 200 for a crack structure of a thermal barrier coating, as shown in fig. 5, where the device includes:
the image acquisition module 201 is used for acquiring a real coating microstructure image of the thermal barrier coating containing the surface cracks, which is prepared according to preset spraying parameters;
the training module 202 is used for training the improved condition generation type countermeasure network model according to a plurality of groups of spraying parameters and the real coating microstructure image, and a loss function based on Wasserstein distance and gradient penalty term is adopted during training;
the coating performance generating module 203 is configured to input a set of spraying parameters in a trained generator of the improved condition generating type countermeasure network model, and obtain coating performance information extracted from the generated image;
the reverse optimization module 204 is configured to determine coating performance information with optimal service life according to a service life experiment of the prepared thermal barrier coating containing surface cracks, and reversely optimize spraying parameters of a thermal barrier coating crack structure according to the optimal coating performance information.
In another aspect, the invention further provides an electronic device, which comprises a processor and a memory, wherein at least one instruction, at least one section of program, a code set or an instruction set is stored in the memory, and the at least one instruction, the at least one section of program, the code set or the instruction set is loaded and executed by the processor to realize the spraying parameter optimization method of the thermal barrier coating crack structure according to the first aspect.
In another aspect, the present invention also provides a computer readable storage medium, where at least one instruction, at least one program, a code set, or an instruction set is stored, where the at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by a processor to implement the method for optimizing a spraying parameter of the crack structure of the thermal barrier coating according to the first aspect.
It is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
In this specification, each embodiment is described in a related manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for system embodiments, since they are substantially similar to method embodiments, the description is relatively simple, as relevant to see a section of the description of method embodiments.
The foregoing description is only of the preferred embodiments of the present invention and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention are included in the protection scope of the present invention.
Claims (9)
1. The spray coating parameter optimization method of the thermal barrier coating crack structure is characterized by comprising the following steps of:
acquiring a real coating microstructure image of the thermal barrier coating containing the surface cracks, which is prepared according to preset spraying parameters;
generating an countermeasure network model according to a plurality of groups of spraying parameters and improved conditions of real coating microstructure image training, wherein a loss function based on Wasserstein distance and gradient penalty term is adopted during training;
inputting a group of spraying parameters into a trained generator of an improved condition generation type countermeasure network model to acquire coating performance information extracted from a generated image;
and determining the optimal coating performance information of the service life according to the service life experiment of the prepared thermal barrier coating containing the surface cracks, and determining the optimal coating performance information extracted from the generated image, thereby determining the optimal spraying parameters of the crack structure of the thermal barrier coating.
2. The method for optimizing spray parameters of a thermal barrier coating crack structure according to claim 1, wherein training the improved condition generating type countermeasure network model based on a plurality of sets of spray parameters and a real coating microstructure image comprises:
training an improved condition generation type countermeasure network model generator by adopting an m-dimensional random number one-dimensional tensor and an n-dimensional spraying parameter condition one-dimensional tensor to obtain a two-dimensional tensor of a generated image, wherein m and n are natural numbers;
training an improved condition generation type countermeasure network model discriminator by adopting a two-dimensional tensor of a generated image and a real coating microstructure image and a conditional one-dimensional tensor of n-dimensional spraying parameters, and acquiring a probability one-dimensional tensor;
during training, parameters of the discriminator are optimized and updated by adopting a loss function based on Wasserstein distance and gradient penalty term, and then parameters of the generator are optimized and updated by adopting the loss function of the generator, until the loss functions of the generator and the discriminator are minimum and reach balance, and training is stopped.
3. The method for optimizing spray parameters of a thermal barrier coating crack structure of claim 1, wherein the loss function based on the wasperstein distance and gradient penalty term comprises:
wherein->Representing a discriminator loss function; />Representation->At->Obeys->The expectation of the function under the distribution condition; />Representation->At->Obeys->The expectation of the function under the distribution condition; />Representation->At->Obeys->The expectation of the function under the distribution condition; />Representing a condition parameter; />Representing a distribution of random number vectors in the generator; />Representing the generated image samples; />Representing the probability that the image output by the discriminator is true, and the probability is between 0 and 1; />Representing the distribution of real samples; />Representing a real image sample; />Is a weight coefficient of the gradient penalty term; />Expressed in real sample->And generate sample->Randomly interpolating and sampling the obtained images; />Representation->Is a random distribution of (a); />Representing a 2-norm; />Representing the calculated gradient; />Obeys [0,1 ]]Evenly distributed.
4. A method of optimizing spray parameters of a thermal barrier coating crack structure as recited in claim 3, wherein the generator loss function comprises:wherein->Representing the generator loss function.
5. The method for optimizing spray parameters of a crack structure of a thermal barrier coating of claim 1, further comprising, after acquiring the image of the true coating microstructure:
acquiring a plurality of intercepted pictures in a real coating microstructure image;
and marking a plurality of cut pictures by using a category label or a numerical label.
6. The method for optimizing spray parameters of a thermal barrier coating crack structure according to claim 1, wherein the generator of the improved condition generating type countermeasure network model comprises 1 fully connected layer, 2 deconvolution layers, 6 residual blocks, 2 up-sampling layers and a LeakyReLU activation function layer which are connected in sequence, and the discriminator of the improved condition generating type countermeasure network model comprises 1 fully connected layer, 7 convolution layers, 1 residual block, 4 down-sampling layers, a LeakyReLU activation function layer and 2 fully connected layers which are connected in sequence.
7. The utility model provides a spraying parameter optimization device of thermal barrier coating crack structure which characterized in that includes:
the image acquisition module is used for acquiring a real coating microstructure image of the thermal barrier coating with the surface cracks, which is prepared according to preset spraying parameters;
the training module is used for training the improved condition generation type countermeasure network model according to a plurality of groups of spraying parameters and the real coating microstructure image, and a loss function based on Wasserstein distance and gradient penalty items is adopted during training;
the coating performance generation module is used for inputting a group of spraying parameters in a trained improved condition generation type antagonism network model generator to acquire coating performance information extracted from a generated image;
the reverse optimization module is used for determining coating performance information with optimal service life according to service life experiments of the prepared thermal barrier coating containing the surface cracks, and reversely optimizing spraying parameters of a thermal barrier coating crack structure according to the optimal coating performance information.
8. An electronic device comprising a processor and a memory, wherein the memory stores at least one instruction, at least one program, a set of codes, or a set of instructions, the at least one instruction, the at least one program, the set of codes, or the set of instructions being loaded and executed by the processor to implement the method of optimizing the spray parameters of the thermal barrier coating crack structure of claims 1-6.
9. A computer readable storage medium having stored therein at least one instruction, at least one program, code set, or instruction set, the at least one instruction, the at least one program, the code set, or instruction set being loaded and executed by a processor to implement the method of optimizing a spray parameter of a thermal barrier coating crack structure as recited in claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311334362.XA CN117195743A (en) | 2023-10-16 | 2023-10-16 | Spraying parameter optimization method for crack structure of thermal barrier coating |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311334362.XA CN117195743A (en) | 2023-10-16 | 2023-10-16 | Spraying parameter optimization method for crack structure of thermal barrier coating |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117195743A true CN117195743A (en) | 2023-12-08 |
Family
ID=89005434
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311334362.XA Pending CN117195743A (en) | 2023-10-16 | 2023-10-16 | Spraying parameter optimization method for crack structure of thermal barrier coating |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117195743A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109584337A (en) * | 2018-11-09 | 2019-04-05 | 暨南大学 | A kind of image generating method generating confrontation network based on condition capsule |
| CN111582348A (en) * | 2020-04-29 | 2020-08-25 | 武汉轻工大学 | Method, device, equipment and storage medium for training condition generating type countermeasure network |
| CN112329322A (en) * | 2020-10-09 | 2021-02-05 | 清华大学 | Thermal spraying method and device based on convolutional neural network |
| CN112765904A (en) * | 2020-12-22 | 2021-05-07 | 苏州先机动力科技有限公司 | Method and equipment for measuring and calculating service life of thermal barrier coating |
| IT202000000664A1 (en) * | 2020-01-15 | 2021-07-15 | Digital Design S R L | GENERATIVE SYSTEM FOR THE CREATION OF DIGITAL IMAGES FOR PRINTING ON DESIGN SURFACES |
| CN114154405A (en) * | 2021-11-19 | 2022-03-08 | 东风汽车集团股份有限公司 | Motor fault data enhancement method for generating countermeasure network based on conditional Wasserstein |
| CN115238581A (en) * | 2022-07-25 | 2022-10-25 | 苏州大学 | Method for predicting HVOF coating performance optimization spraying parameters based on HML |
-
2023
- 2023-10-16 CN CN202311334362.XA patent/CN117195743A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109584337A (en) * | 2018-11-09 | 2019-04-05 | 暨南大学 | A kind of image generating method generating confrontation network based on condition capsule |
| IT202000000664A1 (en) * | 2020-01-15 | 2021-07-15 | Digital Design S R L | GENERATIVE SYSTEM FOR THE CREATION OF DIGITAL IMAGES FOR PRINTING ON DESIGN SURFACES |
| CN111582348A (en) * | 2020-04-29 | 2020-08-25 | 武汉轻工大学 | Method, device, equipment and storage medium for training condition generating type countermeasure network |
| CN112329322A (en) * | 2020-10-09 | 2021-02-05 | 清华大学 | Thermal spraying method and device based on convolutional neural network |
| CN112765904A (en) * | 2020-12-22 | 2021-05-07 | 苏州先机动力科技有限公司 | Method and equipment for measuring and calculating service life of thermal barrier coating |
| CN114154405A (en) * | 2021-11-19 | 2022-03-08 | 东风汽车集团股份有限公司 | Motor fault data enhancement method for generating countermeasure network based on conditional Wasserstein |
| CN115238581A (en) * | 2022-07-25 | 2022-10-25 | 苏州大学 | Method for predicting HVOF coating performance optimization spraying parameters based on HML |
Non-Patent Citations (1)
| Title |
|---|
| MAHENDRU, P ET AL.: "Artificial Intelligence Models for Analyzing Thermally Sprayed Functional Coatings", 《JOURNAL OF THERMAL SPRAY TECHNOLOGY》, no. 32, 28 March 2023 (2023-03-28), pages 388 - 400 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110797096B (en) | Deep learning-based high-temperature mechanical property prediction method for heat-resistant alloy | |
| CN105628511B (en) | A kind of high temperature alloy creep life prediction technique | |
| Gupta et al. | A diffusion-based oxide layer growth model using real interface roughness in thermal barrier coatings for lifetime assessment | |
| CN106897520B (en) | Heat transfer system reliability analysis method containing fuzzy parameters | |
| CN110895814A (en) | Intelligent segmentation method for aero-engine hole detection image damage based on context coding network | |
| CN111898237A (en) | Parallel simulated annealing rapid optimization method for inversion measurement of multiple thermophysical parameters of material | |
| Garikapati et al. | K-means partitioning approach to predict the error observations in small datasets | |
| CN108459993B (en) | Complex high-dimensional system optimization method based on rapid peak-tracking sampling | |
| CN116822567A (en) | Optimization method for evaporation waveguide prediction model parameters | |
| CN109736720A (en) | A kind of deep-sea connector sealing structure optimization method based on improvement Kriging model | |
| Xu et al. | An intelligent optimization algorithm for blocking flow-shop scheduling based on differential evolution | |
| CN117195743A (en) | Spraying parameter optimization method for crack structure of thermal barrier coating | |
| CN109885875A (en) | A kind of turbo blade probability damage tolerance analysis method | |
| CN116825243B (en) | Multi-source data-based thermal barrier coating service life prediction method and system | |
| Gao et al. | A 6-sigma robust optimization method for stamping forming of automobile covering parts based on residual error and radial basis interpolation | |
| CN106909747B (en) | Fuzzy parameter membership function identification method in heat convection diffusion system | |
| CN116342542A (en) | Lightweight neural network-based steel product surface defect detection method | |
| CN115661071A (en) | Composite material processing surface defect detection and evaluation method based on deep learning | |
| Kim et al. | Prediction on the Distributions of the Strength and Toughness of Thick Steel Plates Based on Bayesian Neural Network | |
| CN115050434A (en) | High-flux screening method for ultrahigh-temperature thermal barrier coating material | |
| CN114996880A (en) | Composite armor structure optimization method based on ANSYS secondary development | |
| CN109145518B (en) | Method for constructing reliability decision graph model of large-scale complex equipment | |
| CN109390946B (en) | Optimal probability load flow rapid calculation method based on multi-parameter planning theory | |
| CN113033042A (en) | Method for fitting natural pore information of continuous fiber toughened ceramic matrix composite | |
| CN112326540A (en) | Nuclear zirconium-4 alloy corrosion resistance prediction method based on BP network model, electronic 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 |