CN116825243A - Multi-source data-based thermal barrier coating service life prediction method and system - Google Patents
Multi-source data-based thermal barrier coating service life prediction method and system Download PDFInfo
- Publication number
- CN116825243A CN116825243A CN202310523994.4A CN202310523994A CN116825243A CN 116825243 A CN116825243 A CN 116825243A CN 202310523994 A CN202310523994 A CN 202310523994A CN 116825243 A CN116825243 A CN 116825243A
- Authority
- CN
- China
- Prior art keywords
- source data
- barrier coating
- thermal barrier
- service life
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 239000012720 thermal barrier coating Substances 0.000 title claims abstract description 102
- 238000000034 method Methods 0.000 title claims abstract description 67
- 239000013598 vector Substances 0.000 claims abstract description 114
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 50
- 238000005516 engineering process Methods 0.000 claims abstract description 31
- 238000001514 detection method Methods 0.000 claims abstract description 25
- 238000012545 processing Methods 0.000 claims abstract description 25
- 238000000605 extraction Methods 0.000 claims abstract description 22
- 230000008569 process Effects 0.000 claims abstract description 20
- 238000010801 machine learning Methods 0.000 claims abstract description 19
- 230000008859 change Effects 0.000 claims abstract description 14
- 238000007781 pre-processing Methods 0.000 claims abstract description 13
- 238000010606 normalization Methods 0.000 claims abstract description 10
- 238000004140 cleaning Methods 0.000 claims abstract description 8
- 241000254173 Coleoptera Species 0.000 claims description 30
- 210000003608 fece Anatomy 0.000 claims description 30
- 238000000513 principal component analysis Methods 0.000 claims description 17
- 238000004458 analytical method Methods 0.000 claims description 16
- 239000011159 matrix material Substances 0.000 claims description 12
- 230000004927 fusion Effects 0.000 claims description 11
- 238000003064 k means clustering Methods 0.000 claims description 11
- 230000002159 abnormal effect Effects 0.000 claims description 10
- 230000006870 function Effects 0.000 claims description 10
- 238000012549 training Methods 0.000 claims description 7
- 238000013528 artificial neural network Methods 0.000 claims description 6
- 238000005311 autocorrelation function Methods 0.000 claims description 6
- 238000005457 optimization Methods 0.000 claims description 6
- 230000006641 stabilisation Effects 0.000 claims description 5
- 238000011105 stabilization Methods 0.000 claims description 5
- 238000009499 grossing Methods 0.000 claims description 4
- 230000009467 reduction Effects 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 238000001328 terahertz time-domain spectroscopy Methods 0.000 claims description 2
- 238000009659 non-destructive testing Methods 0.000 description 7
- 238000009826 distribution Methods 0.000 description 5
- 239000000919 ceramic Substances 0.000 description 4
- 238000001914 filtration Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 239000011248 coating agent Substances 0.000 description 3
- 238000000576 coating method Methods 0.000 description 3
- 230000006872 improvement Effects 0.000 description 3
- 239000000463 material Substances 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 238000002759 z-score normalization Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000005336 cracking Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000977 initiatory effect Effects 0.000 description 2
- 239000002184 metal Substances 0.000 description 2
- 230000005855 radiation Effects 0.000 description 2
- 239000000758 substrate Substances 0.000 description 2
- 229910004298 SiO 2 Inorganic materials 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 230000003321 amplification Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 230000007797 corrosion Effects 0.000 description 1
- 238000005260 corrosion Methods 0.000 description 1
- 230000006378 damage Effects 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000003628 erosive effect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000011049 filling Methods 0.000 description 1
- 238000003331 infrared imaging Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000003199 nucleic acid amplification method Methods 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 230000001681 protective effect Effects 0.000 description 1
- 238000004451 qualitative analysis Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000004901 spalling Methods 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000012731 temporal analysis Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000000700 time series analysis Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C60/00—Computational materials science, i.e. ICT specially adapted for investigating the physical or chemical properties of materials or phenomena associated with their design, synthesis, processing, characterisation or utilisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
- G06F18/15—Statistical pre-processing, e.g. techniques for normalisation or restoring missing data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/10—Numerical modelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2113/00—Details relating to the application field
- G06F2113/26—Composites
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/02—Reliability analysis or reliability optimisation; Failure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/08—Thermal analysis or thermal optimisation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Mathematical Physics (AREA)
- Medical Informatics (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Investigating Or Analyzing Materials Using Thermal Means (AREA)
Abstract
The invention discloses a service life prediction method and a service life prediction system for a thermal barrier coating based on multi-source data, wherein the method comprises the following steps: detecting the thermal barrier coating based on a nondestructive detection technology to obtain full life cycle multisource data representing the growth of microcracks and the change state of TGO of the thermal barrier coating; preprocessing multi-source data by adopting a data cleaning, missing value processing and anomaly detection method; performing feature extraction on the preprocessed multi-source data by adopting a method of combining machine learning with signal processing to obtain multi-source data vectors; splicing the multi-source data vectors to obtain a feature vector group, and carrying out normalization processing; based on a machine learning algorithm and the normalized feature vector group, a thermal barrier coating service life prediction model is constructed, and the thermal barrier coating service life prediction model is used for predicting the thermal barrier coating service life. The invention can flexibly and accurately process signal data and accurately predict the service life of the thermal barrier coating.
Description
Technical Field
The invention belongs to the technical field of nondestructive testing and evaluation, and particularly relates to a service life prediction method and a service life prediction system of a thermal barrier coating based on multi-source data.
Background
The development of the aero-engine is an important mark for measuring the national science and technology level, and the service life of the hot end part is an important factor for influencing the performance of the aero-engine. With the improvement of the pushing ratio and the inlet temperature, the high-temperature resistant requirement of the engine blade is higher, and the thermal barrier coating is needed to be a protective material for reducing the working temperature of the blade and prolonging the service life of the hot end component. However, the engine blade is subjected to high temperatures resulting in cracking of the coating, modification of the micro-porous structure, CMAS (main component CaO, mgO, al 2 O 3 And SiO 2 CMAS) corrosion, hard particle erosion, etc., are critical factors affecting the service life of the coating. In the process from service start to failure destruction of a thermal barrier coating in a severe environment for a long time, the thermal barrier coating has a plurality of influencing factors and complex interaction, such as: thickness, porosity, TGO layer thickness, number of cracks, density and location, and the like. At present, no single nondestructive testing technology can be used for evaluating the effective service life of the metallographic phase of the thermal barrier coating at different service stages. Therefore, prediction of service life of the thermal barrier coating becomes one of key problems to be solved in the aspect of guaranteeing service safety of the aero-engine blade. Based on the method, a non-contact, safe and reliable thermal barrier coating nondestructive testing method with high detection precision and high response speed is needed, and a multi-source data mode is formed by combining the advantages of multiple nondestructive testing technologies, so that the thermal barrier coatings in different working stages are effectively tested, and the thermal barrier coating is realizedAnd (5) accurately predicting the service life of the layer.
At present, the method for evaluating the service life of the thermal barrier coating by evaluating the internal microstructure of the thermal barrier coating at home and abroad comprises the following steps: eddy current detection, ultrasonic detection, impedance detection, metallographic detection, and the like. However, as the service environment of the thermal barrier coating becomes more severe, a single detection method has a defect, and the prediction of service life is difficult. Therefore, the service life state of the thermal barrier coating can be reflected more accurately by combining the multi-source data, and meanwhile, the limitation of single data can be reduced by multi-source data fusion, so that the comprehensive performance of the thermal barrier coating can be qualitatively and quantitatively analyzed in multiple angles, multiple aspects and multiple dimensions, and the purpose of accurately predicting the service life can be achieved.
Disclosure of Invention
The invention aims to solve the defects of the prior art, and provides a thermal barrier coating service life prediction method and a thermal barrier coating service life prediction system based on multi-source data, which are innovatively designed on the basis of a traditional single detection method and utilize a plurality of detection technologies: terahertz technology, acoustic emission technology and thermal infrared technology, and multi-source data are acquired. Through fusion of multi-source data, correlation and complementarity among different source data can be fully utilized. And the machine learning is applied to the detection of the thermal barrier coating, so that the method is an effective method, and can flexibly and accurately process signal data and accurately predict the service life of the thermal barrier coating. To perfect and solve the problems existing in the prior art.
In order to achieve the above object, the present invention provides the following solutions:
a service life prediction method of a thermal barrier coating based on multi-source data comprises the following steps:
s1, detecting a thermal barrier coating based on a nondestructive detection technology to obtain full life cycle multi-source data representing the growth of microcracks and the change state of TGO of the thermal barrier coating;
s2, preprocessing the multi-source data by adopting a method of data cleaning, missing value processing and anomaly detection;
s3, performing feature extraction on the preprocessed multi-source data by adopting a method of combining machine learning with signal processing to obtain multi-source data vectors;
s4, splicing the multi-source data vectors to obtain a feature vector group, and carrying out normalization processing;
s5, constructing a thermal barrier coating service life prediction model based on a machine learning algorithm and the normalized feature vector group, wherein the thermal barrier coating service life prediction model is used for predicting the thermal barrier coating service life.
Preferably, the multi-source data includes first source data, second source data and third source data;
the first source data is terahertz time-domain spectroscopy;
the second source data is the acoustic emission time number;
the third source data is thermal infrared radiant energy.
Preferably, in the step S3, the step of reducing the dimension of the high-dimensional data in the multi-source data, and performing feature extraction after performing principal component analysis includes:
performing sparse representation on the first source data subjected to principal component analysis by adopting an orthogonal matching pursuit algorithm, removing invalid and redundant features, and extracting sparse coefficient vectors;
adopting a K-means clustering algorithm to identify and filter abnormal events and interference signals of the second source data subjected to principal component analysis, and extracting a clustering center vector;
adopting an autoregressive model to carry out smoothing treatment and predictive analysis on the third source data subjected to principal component analysis, and extracting an autoregressive coefficient vector;
the sparse coefficient vector, the cluster center vector, and the autoregressive coefficient vector form the multi-source data vector.
Preferably, the application process of the orthogonal matching pursuit algorithm is as follows:
s11, based on the orthogonal matching pursuit algorithm, obtaining a base vector of the first source data, and calculating an inner product of the first source data and the base vector;
s21, calculating projection of the first source data in a base vector set based on the inner product;
s31, calculating residual errors based on the first source data and the projection;
s41, calculating projection of the residual error in the base vector set to obtain a base vector with the maximum projection;
s51, merging the base vector with the largest projection with a base vector index set, and updating the base vector index set;
s61, if the size of the base vector index set reaches a preset value or the norm of the residual reaches a preset range, terminating an algorithm; otherwise, the process returns to step S21 to continue the iteration.
Preferably, the application process of the K-means clustering algorithm is as follows:
converting the second source data into a feature matrix;
calculating a distance matrix between data points based on the feature vectors in the feature matrix;
and inputting the distance matrix into the K-means clustering algorithm, and executing clustering.
Preferably, the application process of the autoregressive model is as follows:
removing noise and abnormal values of the third source data;
carrying out time sequence stabilization on the non-stable time sequence of the third source data with noise and abnormal values removed by adopting differential and logarithmic transformation to obtain a stable time sequence;
adopting an autocorrelation function to realize the autocorrelation analysis of the data in the stable time sequence;
constructing the autoregressive model based on the stationary time series analyzed by autocorrelation;
residual analysis is carried out on the autoregressive model, and whether preset conditions are met or not is judged;
and based on the autoregressive model meeting the preset conditions, realizing predictive analysis of the third source data.
Preferably, the method for constructing the thermal barrier coating service life prediction model comprises the following steps:
taking the normalized characteristic vector group as input and taking the service life stage of the thermal barrier coating as output;
and combining the input and the output into a data set, substituting the data set into a deep extreme learning machine for training, and constructing the thermal barrier coating service life prediction model.
Preferably, the weight and the threshold of the depth extreme learning machine are optimized by adopting a dung beetle optimization algorithm, and the optimization process is as follows:
s12, initializing a dung beetle population;
s22, calculating errors of the neural networks corresponding to the dung beetles on a training set in the data set, and taking the errors as fitness function values of the dung beetles;
s32, updating the position of the dung beetle based on the fitness function value and neighborhood information of the dung beetle;
s42, obtaining an optimal solution of the dung beetle population based on the fitness function value of the dung beetle at the updated position;
s52, presetting an algorithm stopping condition, wherein when the algorithm stopping condition is met, the algorithm is terminated, the optimal solution of the current dung beetle population is obtained, and otherwise, the step S32 is returned.
The invention also provides a thermal barrier coating service life prediction system based on multi-source data, which comprises the following steps: multisource data acquisition module, data preprocessing module, feature extraction module, data fusion module and prediction module
The multisource data acquisition module is used for detecting the thermal barrier coating based on a nondestructive detection technology to obtain full life cycle multisource data representing the growth of microcracks and the change state of TGO of the thermal barrier coating;
the data preprocessing module is used for preprocessing the multi-source data by adopting methods of data cleaning, missing value processing and anomaly detection;
the feature extraction module is used for extracting features of the preprocessed multi-source data by adopting a method of combining machine learning with signal processing to obtain multi-source data vectors;
the data fusion module is used for splicing the multi-source data vectors to obtain a feature vector group and carrying out normalization processing;
the prediction module is used for constructing a thermal barrier coating service life prediction model based on a machine learning algorithm and the normalized feature vector group, and realizing thermal barrier coating service life prediction based on multi-source data.
Compared with the prior art, the invention has the beneficial effects that: the invention provides a method for predicting service life of a thermal barrier coating of an aeroengine blade based on multi-source data. The thermal barrier coating is subjected to nondestructive quantitative detection by using a terahertz technology, an acoustic emission technology and a thermal infrared imaging technology, multiple signal multi-source data used for representing service life are obtained, and machine learning is applied to a regression model of service life state of the thermal barrier coating, which is established based on data fusion and a distribution state stage of microcrack growth and thermal growth oxide layers. The multi-source data can improve the prediction accuracy and stability, and expand the application range of the thermal barrier coating life prediction.
The service life prediction method of the thermal barrier coating provided by the invention belongs to non-contact nondestructive testing, is carried out by adopting multi-source data and combining a machine learning algorithm to carry out data fusion and neural network regression, can comprehensively and accurately predict, and is suitable for predicting the service life of the thermal barrier coating of the blade of the aeroengine.
Drawings
In order to more clearly illustrate the technical solutions of the present invention, the drawings that are needed in the embodiments are briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings can be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flowchart of a method for predicting service life of a thermal barrier coating based on multi-source data according to an embodiment of the invention;
FIG. 2 is a schematic diagram of a typical structure of a thermal barrier coating being tested in accordance with an embodiment of the present invention;
FIG. 3 is a schematic representation of a microcrack and TGO variation in accordance with an embodiment of the present invention;
FIG. 4 is a schematic flow diagram of an embodiment of the present invention;
FIG. 5 is a flowchart of DBO-DELM prediction according to an embodiment of the present 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.
In order that the above-recited objects, features and advantages of the present invention will become more readily apparent, a more particular description of the invention will be rendered by reference to the appended drawings and appended detailed description.
Example 1
As shown in FIG. 1, the method for predicting the service life of the thermal barrier coating based on multi-source data comprises the following steps:
s1, detecting a thermal barrier coating based on a nondestructive detection technology to obtain full life cycle multi-source data representing the growth of microcracks and the change state of TGO of the thermal barrier coating;
specifically, as shown in FIG. 2, the thermal barrier coating comprises; metal substrate, tie layer, thermally grown oxide layer and top ceramic layer. The ceramic layer is fixed on the metal substrate through a bonding layer; a thermally grown oxide layer is between the ceramic layer and the bonding layer.
As shown in fig. 4, the non-destructive testing technique includes: terahertz technology, acoustic emission technology, thermal infrared technology; the terahertz technology is a novel nondestructive testing technology, and is used for detecting the microcrack change of the thermal barrier coating due to the advantages of safety, high resolution, non-contact nondestructive performance and the like of the terahertz technology; the acoustic emission technology is a phenomenon that a material is cracked or deformed and energy is released in the form of elastic stress wave, and the acoustic emission signal of the material is received and analyzed to judge the structural integrity and is used for detecting the growth and change state distribution of the thermal barrier coating TGO; the thermal infrared technology is used for detecting the complete failure process of the thermal barrier coating, namely, the extracted complete failure characteristic data parameters are fused into a time sequence.
Specifically, the multi-source data comprises first source data, second source data and third source data;
the first source data is a terahertz time-domain spectrum (terahertz signal);
the second source data is the acoustic emission time number;
the third source data is thermal infrared radiant energy.
S2, preprocessing multi-source data by adopting a data cleaning, missing value processing and anomaly detection method;
specifically, as shown in fig. 4, the data cleaning includes filtering and denoising, a filter is used to remove high-frequency noise and low-frequency noise of the first and third source data, and a reduction amplification factor is adopted for the second source data, so that low-pass filtering is added to reduce noise interference; and removing the repeated data by using a data de-duplication algorithm; processing the missing value; filling the missing value by using an interpolation method and an average value method. Abnormality detection; using a statistical method, outliers in the data are identified based on the probability distribution and the median absolute deviation method.
S3, performing feature extraction on the preprocessed multi-source data by adopting a method of combining machine learning with signal processing to obtain multi-source data vectors;
specifically, in S3, the dimension of the high-dimensional data in the multi-source data is reduced, and after the principal component analysis, the feature extraction is performed, where the high-dimensional data refers to the multi-dimensional data, and the number of attributes of the data set is large. In this embodiment, there are many data points of the terahertz time-domain data, and the dimension of one group of terahertz time-domain data is up to 4000 dimensions, that is, the number of data attribute points is large, so that dimension reduction processing is required. The number of acoustic emission events and the thermal infrared radiation energy are the same, and the number of data points is more. I.e. to reduce the dimensions of the multi-source high-dimensional data.
Particularly, taking the data of each main component with the accumulated contribution rate reaching 95% for subsequent analysis;
the first source data (terahertz signals) subjected to principal component analysis are subjected to sparse representation by adopting an orthogonal matching pursuit algorithm (OMP algorithm), namely, the original data are represented as a linear combination form of basis vector combination. Retaining the characteristics which have obvious influence on the life prediction of the thermal barrier coating, removing invalid and redundant characteristics, and extracting sparse coefficient vectors; the OMP algorithm can automatically select the base vector which can most represent the signal, and gradually eliminates the base vector which has smaller influence on the signal, thereby achieving the purpose of sparse representation. A more compact, representative representation of features can be obtained, which helps to improve the accuracy and robustness of subsequent machine learning algorithms.
Adopting a K-means clustering algorithm to identify and filter abnormal events and interference signals of the second source data subjected to principal component analysis, and extracting a clustering center vector;
adopting an autoregressive model to carry out smoothing treatment and predictive analysis on the third source data subjected to principal component analysis, and extracting an autoregressive coefficient vector;
the sparse coefficient vector, the cluster center vector and the autoregressive coefficient vector form a multi-source data vector.
Specifically, the application process of the orthogonal matching pursuit algorithm is as follows:
s11, based on an orthogonal matching pursuit algorithm, obtaining a base vector of the first source data, and calculating an inner product a of the first source data and the base vector k ：Wherein x is an input terahertz time-domain signal, < >>Is the kth base vector in the set of base vectors;
s21, calculating projection of the first source data in the base vector set based on the inner product
S31, based on the first source dataAnd projecting, calculating residual r:
s41, calculating projection of residual errors in a base vector set to obtain a base vector k with the maximum projection * ：
S51, merging the base vector with the largest projection with the base vector index set, and updating the base vector index set S: s=s { k } U * }
S61, if the size of the base vector index set reaches a preset value or the norm of the residual reaches a preset range, terminating the algorithm; otherwise, the process returns to step S21 to continue the iteration. In particular, the use of OMP algorithms requires preset parameters such as sparsity and maximum number of iterations. The sparsity represents the number of selected basis vectors, and the maximum iteration number represents the termination condition of the algorithm.
Specifically, the application process of the K-means clustering algorithm is as follows:
converting the second source data (acoustic emission time number) into a feature matrix;
calculating a distance matrix between the data points based on the feature vectors in the feature matrix;
and inputting the distance matrix into a K-means clustering algorithm, and executing clustering.
Specifically, the application process of the autoregressive model is as follows:
removing noise and abnormal values of the third source data;
carrying out time sequence stabilization on the non-stable time sequence of the third source data with noise and abnormal values removed by adopting differential and logarithmic transformation to obtain a stable time sequence;
the autocorrelation function is adopted to realize the autocorrelation analysis of the data in the stable time sequence, the autocorrelation function can be used for describing the self correlation in the time sequence data, and the calculation formula of the autocorrelation function can be used:
where c is the lag phase number, n is the time series length, x t For thermal infrared data at time t,the value of the autocorrelation function is between-1 and 1, which is the mean of the time series, reflecting the correlation between adjacent values in the time series.
Constructing an autoregressive model (AR model) based on the autocorrelation-analyzed stationary time sequence; the AR model is a common time series analysis method that can be used to predict future values. The calculation formula of the AR model is as follows:
wherein, p is the order number,e is a regression coefficient i Is an error term. The coefficients of the AR model are estimated by the least squares method.
Carrying out residual analysis on the autoregressive model, and judging whether preset conditions are met or not; the preset condition is whether the assumption of randomness, stationarity and the like is satisfied.
Based on the autoregressive model meeting the preset conditions, predictive analysis of third source data (thermal infrared radiation energy) is realized.
S4, splicing the multi-source data vectors to obtain a feature vector group, and carrying out normalization processing;
specifically, three sets of vectors (namely a sparse coefficient vector, a clustering center vector and an autoregressive coefficient vector) are spliced to form a high-dimension feature vector set, and then normalization processing is carried out to avoid that the dimensions of different features influence the performance of the model. And finally, taking the processed characteristic vector group as input, and carrying out life prediction by using a neural network model. Sparse representation of first source data(sparse coefficient vector) isThe second source data clustering result (clustering center vector) isThe third source data time series AR model parameters (autoregressive coefficient vector) are +.>Splicing to form a feature vector->Can be expressed as x= [ x ] 1 ；x 2 ；x 3 ]Wherein [;]representing a stitching operation. And carrying out normalization treatment by using a Z-score normalization method, namely subtracting the mean value of each characteristic value from each characteristic value, and dividing the mean value by the standard deviation. Let the ith eigenvalue be x i Mean value of mu i Standard deviation is sigma i The Z-score normalization of the eigenvalue results in:
after treatment, the mean value of all the characteristic values is 0, the standard deviation is 1, and the generalization capability and stability of the prediction model are improved.
S5, constructing a thermal barrier coating service life prediction model based on a machine learning algorithm and the normalized feature vector group, wherein the thermal barrier coating service life prediction model is used for predicting the thermal barrier coating service life.
As shown in FIG. 3, the service life prediction of the thermal barrier coating is specifically divided into five stages: initial stabilization phase: the surface of the thermal barrier coating can go through an initial stabilization stage in the initial service period, and the thermal barrier coating performance is basically stable at the stage without obvious degradation signs; a first change phase: namely, the surface crack initiation stage; second change phase: the thermal barrier coating gradually reduces in the growth stage, which is the initiation of interfacial cracks and the expansion of internal cracks caused by the burning of TGO; third variation phase: TGOs exhibit cracking and debonding with increased microcrack growth, i.e., accelerated failure phases. The performance of the thermal barrier coating is drastically reduced at this stage, and failure is likely to occur; fourth phase of change: the spalling of the ceramic layer results in failure of the thermal barrier coating, and the coating structure is completely damaged at this stage, and cannot be used continuously, and needs to be replaced or repaired.
Specifically, the method for constructing the thermal barrier coating service life prediction model comprises the following steps:
taking the normalized characteristic vector group (namely fusion of multi-source data representing the microcrack growth and the thermal growth oxide layer distribution state) as input and taking the service life stage of the thermal barrier coating as output;
and combining the input and the output into a data set, substituting the data set into a deep extreme learning machine ((Deep Extreme Learning Machine, DELM)) for training, and constructing a thermal barrier coating service life prediction model.
In particular, as shown in fig. 5, the weight and the threshold of the deep extreme learning machine are optimized by adopting a dung beetle optimization algorithm (Dung Beetle Optimizer, DBO), so that the DELM prediction accuracy is improved. The optimization process (DBO-DELM flow) is:
s12, initializing a dung beetle population: firstly, randomly initializing a certain number of dung beetles, wherein each dung beetle represents a parameter vector of a neural network.
S22, calculating a fitness function: calculating errors of the neural network corresponding to each dung beetle on a training set (the data set of the input and output combination comprises the training set and a testing set), and taking the errors as fitness function values of the dung beetles;
s32, updating the position of the dung beetles: updating the position of the dung beetle based on the fitness function value and neighborhood information of the dung beetle;
s42, selecting a global optimal solution: acquiring an optimal solution of the current dung beetle population based on the fitness function value of the dung beetle at the updated position;
s52, judging a stop condition: and presetting an algorithm stopping condition, and stopping the algorithm when the stopping condition is met (the iteration times are reached), wherein the algorithm is stopped, and the optimal solution of the current dung beetle population, namely the weight and the threshold value are used as optimal parameters of a thermal barrier coating service life prediction model of the deep extreme learning machine. Otherwise, the process returns to step S32.
Example two
The invention also provides a thermal barrier coating service life prediction system based on multi-source data, which is characterized by comprising the following steps: multisource data acquisition module, data preprocessing module, feature extraction module, data fusion module and prediction module
The multi-source data acquisition module is used for detecting the thermal barrier coating based on a nondestructive detection technology to obtain full life cycle multi-source data representing the growth of microcracks and the change state of TGO of the thermal barrier coating; and detecting the full life cycle of the thermal barrier coating by using a terahertz technology, an acoustic emission technology and a thermal infrared technology, and obtaining multi-source data representing the life. The terahertz technology characterizes the microcrack change of the thermal barrier coating, the acoustic emission technology characterizes the TGO growth distribution state, and the thermal infrared technology is used for the whole service process of the thermal barrier coating, so that the qualitative analysis of failure trend and process can be performed;
the data preprocessing module is used for preprocessing the multi-source data by adopting the methods of data cleaning, missing value processing and anomaly detection; the data quality and the integrity are verified, and the effectiveness of subsequent analysis is ensured;
the feature extraction module is used for carrying out feature extraction on the preprocessed multi-source data by adopting a method of combining machine learning with signal processing to obtain multi-source data vectors; extracting the obvious characteristics with representativeness and distinguishing degree from the multi-source data by using a method combining machine learning with signal processing, namely PCA (principal component analysis), OMP signal sparseness, K-means (K-means clustering algorithm) and AR model, and filtering redundant information;
the data fusion module is used for splicing the multi-source data vectors to obtain a feature vector group and performing Z-score normalization processing; the normalization process provides a reliable data base for modeling, and improves the generalization capability and robustness of the model.
The prediction module is used for constructing a thermal barrier coating service life prediction model based on a machine learning algorithm and the normalized feature vector group, and realizing thermal barrier coating service life prediction based on multi-source data. The fused multi-source data (the characteristic vector group after normalization treatment) is used for representing the service life stage change of different thermal barrier coatings, and a DBO-optimized DELM neural network is utilized to establish a multi-source data-driven thermal barrier coating service life prediction model for accurately predicting the service life stage of the thermal barrier coating.
In particular, in the feature extraction module, the feature extraction is performed after the dimension of the high-dimensional data in the multi-source data is reduced and the principal component analysis is performed, including: the device comprises a sparse coefficient vector extraction unit, a clustering center vector extraction unit, an autoregressive coefficient vector extraction unit and a multi-source data vector unit;
the sparse coefficient vector extraction unit is used for carrying out sparse representation on the first source data subjected to principal component analysis by adopting an orthogonal matching pursuit algorithm, removing invalid and redundant features and extracting sparse coefficient vectors;
the clustering center vector extraction unit is used for identifying and filtering abnormal events and interference signals of the second source data subjected to the principal component analysis by adopting a K-means clustering algorithm to extract a clustering center vector;
the autoregressive coefficient vector extraction unit is used for carrying out smoothing treatment and predictive analysis on the third source data subjected to principal component analysis by adopting an autoregressive model, and extracting autoregressive coefficient vectors;
and the multi-source data vector unit is used for forming a multi-source data vector from the sparse coefficient vector, the clustering center vector and the autoregressive coefficient vector.
The above embodiments are merely illustrative of the preferred embodiments of the present invention, and the scope of the present invention is not limited thereto, but various modifications and improvements made by those skilled in the art to which the present invention pertains are made without departing from the spirit of the present invention, and all modifications and improvements fall within the scope of the present invention as defined in the appended claims.
Claims (9)
1. The service life prediction method of the thermal barrier coating based on the multi-source data is characterized by comprising the following steps of:
s1, detecting a thermal barrier coating based on a nondestructive detection technology to obtain full life cycle multi-source data representing the growth of microcracks and the change state of TGO of the thermal barrier coating;
s2, preprocessing the multi-source data by adopting a method of data cleaning, missing value processing and anomaly detection;
s3, performing feature extraction on the preprocessed multi-source data by adopting a method of combining machine learning with signal processing to obtain multi-source data vectors;
s4, splicing the multi-source data vectors to obtain a feature vector group, and carrying out normalization processing;
s5, constructing a thermal barrier coating service life prediction model based on a machine learning algorithm and the normalized feature vector group, wherein the thermal barrier coating service life prediction model is used for predicting the thermal barrier coating service life.
2. The method for predicting service life of a thermal barrier coating based on multi-source data of claim 1,
the multi-source data comprises first source data, second source data and third source data;
the first source data is terahertz time-domain spectroscopy;
the second source data is the acoustic emission time number;
the third source data is thermal infrared radiant energy.
3. The method for predicting service life of a thermal barrier coating based on multi-source data according to claim 2, wherein in S3, performing feature extraction after performing main component analysis on high-dimensional data in the multi-source data by dimension reduction, comprises:
performing sparse representation on the first source data subjected to principal component analysis by adopting an orthogonal matching pursuit algorithm, removing invalid and redundant features, and extracting sparse coefficient vectors;
adopting a K-means clustering algorithm to identify and filter abnormal events and interference signals of the second source data subjected to principal component analysis, and extracting a clustering center vector;
adopting an autoregressive model to carry out smoothing treatment and predictive analysis on the third source data subjected to principal component analysis, and extracting an autoregressive coefficient vector;
the sparse coefficient vector, the cluster center vector, and the autoregressive coefficient vector form the multi-source data vector.
4. The method for predicting service life of a thermal barrier coating based on multi-source data according to claim 3, wherein the orthogonal matching pursuit algorithm is applied as follows:
s11, based on the orthogonal matching pursuit algorithm, obtaining a base vector of the first source data, and calculating an inner product of the first source data and the base vector;
s21, calculating projection of the first source data in a base vector set based on the inner product;
s31, calculating residual errors based on the first source data and the projection;
s41, calculating projection of the residual error in the base vector set to obtain a base vector with the maximum projection;
s51, merging the base vector with the largest projection with a base vector index set, and updating the base vector index set;
s61, if the size of the base vector index set reaches a preset value or the norm of the residual reaches a preset range, terminating an algorithm; otherwise, the process returns to step S21 to continue the iteration.
5. The method for predicting service life of a thermal barrier coating based on multi-source data according to claim 3, wherein the K-means clustering algorithm is applied in the following steps:
converting the second source data into a feature matrix;
calculating a distance matrix between data points based on the feature vectors in the feature matrix;
and inputting the distance matrix into the K-means clustering algorithm, and executing clustering.
6. The method for predicting service life of a thermal barrier coating based on multi-source data according to claim 3, wherein the autoregressive model is applied as follows:
removing noise and abnormal values of the third source data;
carrying out time sequence stabilization on the non-stable time sequence of the third source data with noise and abnormal values removed by adopting differential and logarithmic transformation to obtain a stable time sequence;
adopting an autocorrelation function to realize the autocorrelation analysis of the data in the stable time sequence;
constructing the autoregressive model based on the stationary time series analyzed by autocorrelation;
residual analysis is carried out on the autoregressive model, and whether preset conditions are met or not is judged;
and based on the autoregressive model meeting the preset conditions, realizing predictive analysis of the third source data.
7. The method for predicting service life of a thermal barrier coating based on multi-source data according to claim 3, wherein the method for constructing the thermal barrier coating service life prediction model is as follows:
taking the normalized characteristic vector group as input and taking the service life stage of the thermal barrier coating as output;
and combining the input and the output into a data set, substituting the data set into a deep extreme learning machine for training, and constructing the thermal barrier coating service life prediction model.
8. The multisource data-based thermal barrier coating service life prediction method according to claim 7, wherein the weight and the threshold of the depth extreme learning machine are optimized by adopting a dung beetle optimization algorithm, and the optimization process is as follows:
s12, initializing a dung beetle population;
s22, calculating errors of the neural networks corresponding to the dung beetles on a training set in the data set, and taking the errors as fitness function values of the dung beetles;
s32, updating the position of the dung beetle based on the fitness function value and neighborhood information of the dung beetle;
s42, obtaining an optimal solution of the dung beetle population based on the fitness function value of the dung beetle at the updated position;
s52, presetting an algorithm stopping condition, wherein when the algorithm stopping condition is met, the algorithm is terminated, the optimal solution of the current dung beetle population is obtained, and otherwise, the step S32 is returned.
9. A thermal barrier coating service life prediction system based on multi-source data, comprising: multisource data acquisition module, data preprocessing module, feature extraction module, data fusion module and prediction module
The multisource data acquisition module is used for detecting the thermal barrier coating based on a nondestructive detection technology to obtain full life cycle multisource data representing the growth of microcracks and the change state of TGO of the thermal barrier coating;
the data preprocessing module is used for preprocessing the multi-source data by adopting methods of data cleaning, missing value processing and anomaly detection;
the feature extraction module is used for extracting features of the preprocessed multi-source data by adopting a method of combining machine learning with signal processing to obtain multi-source data vectors;
the data fusion module is used for splicing the multi-source data vectors to obtain a feature vector group and carrying out normalization processing;
the prediction module is used for constructing a thermal barrier coating service life prediction model based on a machine learning algorithm and the normalized feature vector group, and realizing thermal barrier coating service life prediction based on multi-source data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310523994.4A CN116825243B (en) | 2023-05-09 | 2023-05-09 | Multi-source data-based thermal barrier coating service life prediction method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310523994.4A CN116825243B (en) | 2023-05-09 | 2023-05-09 | Multi-source data-based thermal barrier coating service life prediction method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116825243A true CN116825243A (en) | 2023-09-29 |
| CN116825243B CN116825243B (en) | 2024-01-16 |
Family
ID=88126594
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310523994.4A Active CN116825243B (en) | 2023-05-09 | 2023-05-09 | Multi-source data-based thermal barrier coating service life prediction method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116825243B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117129511A (en) * | 2023-09-18 | 2023-11-28 | 安徽工程大学 | Terahertz online monitoring device and method for thermal fatigue behavior of thermal barrier coating |
| CN117272116A (en) * | 2023-10-13 | 2023-12-22 | 西安工程大学 | Transformer fault diagnosis method based on LORAS balance data set |
| CN118038147A (en) * | 2024-02-06 | 2024-05-14 | 安徽工程大学 | Mechanical property evaluation method for micron embedded thermal barrier coating based on multi-feature driving |
Citations (27)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070076845A1 (en) * | 2005-10-04 | 2007-04-05 | Siemens Power Generation, Inc. | Methodologies for non-destructive quantification of thermal barrier coating temperatures on service run parts |
| JP2007303956A (en) * | 2006-05-11 | 2007-11-22 | Tohoku Univ | Technique for noncontact and nondestructive inspection on thermal barrier coating which has deteriorated with age |
| US20110296810A1 (en) * | 2010-06-07 | 2011-12-08 | General Electric Company | Life management system and method for gas turbine thermal barrier coatings |
| CN103439413A (en) * | 2013-08-07 | 2013-12-11 | 湘潭大学 | Acoustic emission signal analytical method for automatically identifying damage mode of thermal barrier coating |
| CN104166787A (en) * | 2014-07-17 | 2014-11-26 | 南京航空航天大学 | Aero-engine remaining life prediction method based on multi-stage information fusion |
| CN107328868A (en) * | 2017-06-27 | 2017-11-07 | 福州大学 | A kind of Analysis of Acoustic Emission Signal method of quick identification ceramic coating failure type |
| US20180120246A1 (en) * | 2016-10-27 | 2018-05-03 | General Electric Company | Nondestructive inspection method for coatings and ceramic matrix composites |
| CN109459286A (en) * | 2018-12-10 | 2019-03-12 | 湘潭大学 | Real-time detection method is damaged in a kind of thermal barrier coating of turbine blade simulation test procedure |
| CN110553998A (en) * | 2019-07-31 | 2019-12-10 | 西安交通大学 | nondestructive testing method for blade test piece of aero-engine based on terahertz technology |
| CN111089877A (en) * | 2019-12-31 | 2020-05-01 | 苏州先机动力科技有限公司 | Nondestructive testing method and equipment for thermal barrier coating |
| CN111239019A (en) * | 2020-01-22 | 2020-06-05 | 华东理工大学 | Thermal barrier coating micropore structural characteristic characterization method based on terahertz spectrum technology |
| CN111258297A (en) * | 2020-01-17 | 2020-06-09 | 北京大学 | Equipment health index construction and service life prediction method based on data fusion network |
| WO2020133639A1 (en) * | 2018-12-29 | 2020-07-02 | 东北大学 | Intelligent analysis system for magnetic flux leakage detection data in pipeline |
| CN111523197A (en) * | 2018-12-26 | 2020-08-11 | 通用电气公司 | Model for predicting faults on a component |
| CN112697629A (en) * | 2020-12-08 | 2021-04-23 | 武汉理工大学 | Device and method for measuring reliability of thermal barrier coating of piston |
| CN112730454A (en) * | 2020-12-23 | 2021-04-30 | 中国人民解放军空军工程大学 | Intelligent damage detection method for composite material based on fusion of optics, infrared thermal waves and ultrasonic waves |
| 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 |
| CN113560955A (en) * | 2021-06-09 | 2021-10-29 | 西安电子科技大学 | Method, system and application for predicting remaining service life of numerical control machine tool cutter |
| CN113850022A (en) * | 2021-09-26 | 2021-12-28 | 中国人民解放军陆军装甲兵学院 | Spraying layer contact fatigue life prediction method based on acoustic-thermal signals |
| CN114201911A (en) * | 2021-11-16 | 2022-03-18 | 华南理工大学 | Rubber material fatigue life prediction method based on extreme learning machine |
| US20220146462A1 (en) * | 2019-07-05 | 2022-05-12 | Chongqing University Of Posts And Telecommunications | Method for predicting remaining life of numerical control machine tool |
| CN114861827A (en) * | 2022-05-27 | 2022-08-05 | 太原理工大学 | Coal mining machine prediction diagnosis and health management method based on multi-source data fusion |
| CN114925614A (en) * | 2022-05-27 | 2022-08-19 | 太原理工大学 | Method for predicting residual life of coal mining machine |
| WO2022268043A1 (en) * | 2021-06-22 | 2022-12-29 | 重庆邮电大学工业互联网研究院 | Method for predicting residual life of numerical control machine tool based on hybrid neural model |
| CN115561201A (en) * | 2022-10-10 | 2023-01-03 | 安徽工程大学 | Terahertz evaluation device for structural integrity of thermal barrier coating and application method thereof |
| CN115630284A (en) * | 2022-10-19 | 2023-01-20 | 山东科技大学 | Air compressor fault analysis and service life prediction system considering service period |
| CN115980130A (en) * | 2022-12-28 | 2023-04-18 | 湘潭大学 | Method and system for quantitatively detecting defects in thermal barrier coating |
-
2023
- 2023-05-09 CN CN202310523994.4A patent/CN116825243B/en active Active
Patent Citations (27)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070076845A1 (en) * | 2005-10-04 | 2007-04-05 | Siemens Power Generation, Inc. | Methodologies for non-destructive quantification of thermal barrier coating temperatures on service run parts |
| JP2007303956A (en) * | 2006-05-11 | 2007-11-22 | Tohoku Univ | Technique for noncontact and nondestructive inspection on thermal barrier coating which has deteriorated with age |
| US20110296810A1 (en) * | 2010-06-07 | 2011-12-08 | General Electric Company | Life management system and method for gas turbine thermal barrier coatings |
| CN103439413A (en) * | 2013-08-07 | 2013-12-11 | 湘潭大学 | Acoustic emission signal analytical method for automatically identifying damage mode of thermal barrier coating |
| CN104166787A (en) * | 2014-07-17 | 2014-11-26 | 南京航空航天大学 | Aero-engine remaining life prediction method based on multi-stage information fusion |
| US20180120246A1 (en) * | 2016-10-27 | 2018-05-03 | General Electric Company | Nondestructive inspection method for coatings and ceramic matrix composites |
| CN107328868A (en) * | 2017-06-27 | 2017-11-07 | 福州大学 | A kind of Analysis of Acoustic Emission Signal method of quick identification ceramic coating failure type |
| CN109459286A (en) * | 2018-12-10 | 2019-03-12 | 湘潭大学 | Real-time detection method is damaged in a kind of thermal barrier coating of turbine blade simulation test procedure |
| CN111523197A (en) * | 2018-12-26 | 2020-08-11 | 通用电气公司 | Model for predicting faults on a component |
| WO2020133639A1 (en) * | 2018-12-29 | 2020-07-02 | 东北大学 | Intelligent analysis system for magnetic flux leakage detection data in pipeline |
| US20220146462A1 (en) * | 2019-07-05 | 2022-05-12 | Chongqing University Of Posts And Telecommunications | Method for predicting remaining life of numerical control machine tool |
| CN110553998A (en) * | 2019-07-31 | 2019-12-10 | 西安交通大学 | nondestructive testing method for blade test piece of aero-engine based on terahertz technology |
| CN111089877A (en) * | 2019-12-31 | 2020-05-01 | 苏州先机动力科技有限公司 | Nondestructive testing method and equipment for thermal barrier coating |
| CN111258297A (en) * | 2020-01-17 | 2020-06-09 | 北京大学 | Equipment health index construction and service life prediction method based on data fusion network |
| CN111239019A (en) * | 2020-01-22 | 2020-06-05 | 华东理工大学 | Thermal barrier coating micropore structural characteristic characterization method based on terahertz spectrum technology |
| CN112697629A (en) * | 2020-12-08 | 2021-04-23 | 武汉理工大学 | Device and method for measuring reliability of thermal barrier coating of piston |
| 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 |
| CN112730454A (en) * | 2020-12-23 | 2021-04-30 | 中国人民解放军空军工程大学 | Intelligent damage detection method for composite material based on fusion of optics, infrared thermal waves and ultrasonic waves |
| CN113560955A (en) * | 2021-06-09 | 2021-10-29 | 西安电子科技大学 | Method, system and application for predicting remaining service life of numerical control machine tool cutter |
| WO2022268043A1 (en) * | 2021-06-22 | 2022-12-29 | 重庆邮电大学工业互联网研究院 | Method for predicting residual life of numerical control machine tool based on hybrid neural model |
| CN113850022A (en) * | 2021-09-26 | 2021-12-28 | 中国人民解放军陆军装甲兵学院 | Spraying layer contact fatigue life prediction method based on acoustic-thermal signals |
| CN114201911A (en) * | 2021-11-16 | 2022-03-18 | 华南理工大学 | Rubber material fatigue life prediction method based on extreme learning machine |
| CN114861827A (en) * | 2022-05-27 | 2022-08-05 | 太原理工大学 | Coal mining machine prediction diagnosis and health management method based on multi-source data fusion |
| CN114925614A (en) * | 2022-05-27 | 2022-08-19 | 太原理工大学 | Method for predicting residual life of coal mining machine |
| CN115561201A (en) * | 2022-10-10 | 2023-01-03 | 安徽工程大学 | Terahertz evaluation device for structural integrity of thermal barrier coating and application method thereof |
| CN115630284A (en) * | 2022-10-19 | 2023-01-20 | 山东科技大学 | Air compressor fault analysis and service life prediction system considering service period |
| CN115980130A (en) * | 2022-12-28 | 2023-04-18 | 湘潭大学 | Method and system for quantitatively detecting defects in thermal barrier coating |
Non-Patent Citations (11)
| Title |
|---|
| BAOHAN YUAN 等: "Nondestructive Evaluation of Thermal Barrier Coatings Thickness Using Terahertz Technique Combined with PCA–GA–ELM Algorithm", COATINGS, pages 1 - 12 * |
| DONGDONG YE 等: "Characterization of thermal barrier coatings microstructural features using terahertz spectroscopy", SURFACE AND COATINGS TECHNOLOGY, vol. 394, pages 1 - 10 * |
| RUIPENG GAO 等: "Intelligent Life Prediction of Thermal Barrier Coating for Aero Engine Blades", COATINGS, pages 1 - 14 * |
| WEI-BANG CHEN 等: "A Fully Automated Approach for Classification of Microstructures in Thermal Barrier Coating Microscopic Images", 2016 IEEE INTERNATIONAL SYMPOSIUM ON MULTIMEDIA (ISM), pages 411 - 412 * |
| XIN WU 等: "Damage Identification of Low Emissivity Coating Based on Convolution Neural Network", IEEE ACCESS, pages 156792 - 156800 * |
| 何箐;吴鹏;汪瑞军;王伟平;: "模拟服役环境下热障涂层损伤趋势的红外原位检测技术", 中国表面工程, no. 04 * |
| 叶东东 等: "热障涂层太赫兹无损检测技术研究进展", 表面技术, vol. 49, no. 10, pages 126 - 137 * |
| 康海松: "基于声发射信号分析的热障涂层损伤模式识别研究", 中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑, vol. 2015, no. 3, pages 031 - 51 * |
| 张秋雁;杨忠;姜遇红;张启伦;卢凯文;张辉斌;: "基于多源信息融合的飞行器部件剩余寿命预测", 机械制造与自动化, no. 01, pages 82 - 86 * |
| 李军;李世昌;: "基于K-SVD-OMP的稀疏表示方法在电力负荷预测中的应用", 电机与控制学报, no. 09 * |
| 胡龙飞;傅攀;林志斌;张思聪;赵蕾;: "基于QPSO-SVR和声发射信号的机械密封寿命预测", 润滑与密封, no. 04, pages 40 - 45 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117129511A (en) * | 2023-09-18 | 2023-11-28 | 安徽工程大学 | Terahertz online monitoring device and method for thermal fatigue behavior of thermal barrier coating |
| CN117129511B (en) * | 2023-09-18 | 2024-04-12 | 安徽工程大学 | Terahertz online monitoring device and method for thermal fatigue behavior of thermal barrier coating |
| CN117272116A (en) * | 2023-10-13 | 2023-12-22 | 西安工程大学 | Transformer fault diagnosis method based on LORAS balance data set |
| CN117272116B (en) * | 2023-10-13 | 2024-05-17 | 西安工程大学 | Transformer fault diagnosis method based on LORAS balance data set |
| CN118038147A (en) * | 2024-02-06 | 2024-05-14 | 安徽工程大学 | Mechanical property evaluation method for micron embedded thermal barrier coating based on multi-feature driving |
| CN118038147B (en) * | 2024-02-06 | 2024-09-10 | 安徽工程大学 | Mechanical property evaluation method for micron embedded thermal barrier coating based on multi-feature driving |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116825243B (en) | 2024-01-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116825243B (en) | Multi-source data-based thermal barrier coating service life prediction method and system | |
| CN107562696B (en) | On-line detection and control method for tire product quality | |
| CN111914875A (en) | Fault early warning method of rotating machinery based on Bayesian LSTM model | |
| CN110779745B (en) | Heat exchanger early fault diagnosis method based on BP neural network | |
| CN115824519B (en) | Comprehensive diagnosis method for valve leakage faults based on multi-sensor information fusion | |
| CN114487129B (en) | Flexible material damage identification method based on acoustic emission technology | |
| CN109710955B (en) | Rolling bearing fault diagnosis and health assessment method based on LCD-recursion quantitative analysis | |
| CN108897354B (en) | Aluminum smelting process hearth temperature prediction method based on deep belief network | |
| CN109409271B (en) | Ferromagnetic material hardness prediction algorithm based on BP neural network improved algorithm | |
| CN112364567A (en) | Residual life prediction method based on consistency check of similarity of degraded tracks | |
| CN109828548B (en) | Performance degradation characteristic evaluation method based on time series change mutation point detection | |
| Zhang et al. | Remaining Useful Life Prediction of Rolling Bearings Using Electrostatic Monitoring Based on Two‐Stage Information Fusion Stochastic Filtering | |
| CN108804740A (en) | Long distance pipeline pressure monitoring method based on integrated improvement ICA-KRR algorithms | |
| CN117809164A (en) | Substation equipment fault detection method and system based on multi-mode fusion | |
| CN116008404A (en) | Heat supply pipeline damage identification and risk assessment method and system based on laser ultrasound | |
| CN115375026A (en) | Method for predicting service life of aircraft engine in multiple fault modes | |
| CN115014789A (en) | CNN-GCN-based dual-sensor aeroengine case fault source acoustic emission positioning method | |
| Qin et al. | A novel physically interpretable end-to-end network for stress monitoring in laser shock peening | |
| CN107505337B (en) | Microwave reflection signal characteristic database for composite insulator defect detection | |
| CN116227172A (en) | Rolling bearing performance degradation evaluation method based on convolutional neural network | |
| CN113266449B (en) | Method and system for predicting air leakage situation in front of aftertreatment system | |
| CN113688895A (en) | Method and system for detecting abnormal firing zone of ceramic roller kiln based on simplified KECA | |
| CN114414524A (en) | Method for rapidly detecting properties of aviation kerosene | |
| CN114493234A (en) | Method for identifying key pressure control points of water supply pipe network | |
| CN114509506A (en) | Online crack evaluation method based on guided wave time-frequency spectrum difference and convolutional neural network set |
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 |