CN114799610A - Welding quality real-time detection method and system based on inverse Fourier transform and self-encoder - Google Patents
Welding quality real-time detection method and system based on inverse Fourier transform and self-encoder Download PDFInfo
- Publication number
- CN114799610A CN114799610A CN202210720862.6A CN202210720862A CN114799610A CN 114799610 A CN114799610 A CN 114799610A CN 202210720862 A CN202210720862 A CN 202210720862A CN 114799610 A CN114799610 A CN 114799610A
- Authority
- CN
- China
- Prior art keywords
- data
- frequency
- welding quality
- window
- fourier transform
- 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
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23K—SOLDERING OR UNSOLDERING; WELDING; CLADDING OR PLATING BY SOLDERING OR WELDING; CUTTING BY APPLYING HEAT LOCALLY, e.g. FLAME CUTTING; WORKING BY LASER BEAM
- B23K31/00—Processes relevant to this subclass, specially adapted for particular articles or purposes, but not covered by only one of the preceding main groups
- B23K31/12—Processes relevant to this subclass, specially adapted for particular articles or purposes, but not covered by only one of the preceding main groups relating to investigating the properties, e.g. the weldability, of materials
- B23K31/125—Weld quality monitoring
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23K—SOLDERING OR UNSOLDERING; WELDING; CLADDING OR PLATING BY SOLDERING OR WELDING; CUTTING BY APPLYING HEAT LOCALLY, e.g. FLAME CUTTING; WORKING BY LASER BEAM
- B23K37/00—Auxiliary devices or processes, not specially adapted to a procedure covered by only one of the preceding main groups
-
- 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
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Mechanical Engineering (AREA)
- Theoretical Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Optics & Photonics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Quality & Reliability (AREA)
- Investigating Or Analyzing Materials By The Use Of Ultrasonic Waves (AREA)
Abstract
The invention discloses a real-time welding quality detection method and a real-time welding quality detection system based on inverse Fourier transform and a self-encoder, wherein the detection method comprises the following steps: collecting time sequence data such as high-frequency current, voltage and the like in a normal welding process, acquiring the period and the main frequency of a high-frequency time sequence, and further obtaining noise data of the high-frequency time sequence through Fourier transform and inverse Fourier transform; and finally, reconstructing noise data by using a self-encoder model and obtaining an abnormal threshold value to meet the requirement of identifying welding quality defects. Compared with the method for constructing the identification model by using data such as images, sounds, spectrums and the like to diagnose the welding quality, the method only needs to use non-invasive data acquisition equipment to acquire high-frequency time sequence data in the welding process, the data is easy to acquire, the detection cost is low, meanwhile, a large amount of abnormal labels are not needed, the defect that the welding defects are difficult to label is effectively overcome, and the model is constructed on the basis of the time sequence data closely related to the welding quality, so that the model is more robust and has good robustness.
Description
Technical Field
The invention relates to a welding quality real-time detection method and system based on inverse Fourier transform and a self-encoder, and belongs to the technical field of automatic welding.
Background
With the rapid development of industries such as automobiles, aerospace, construction and transportation in recent years, the process and quality requirements for industrial equipment are higher and higher, and the welding quality detection technology is widely applied in a plurality of fields in recent years. The welding quality can be divided into direct welding quality and indirect welding quality, and the service performance of a general welding joint mainly has mechanical performance, internal and external defects, geometrical size of a welded product and the like, namely the direct welding quality. The indirect welding quality is a relevant factor which can be detected by a sensor with sense or characteristic of a welder during the welding process and indirectly determines the direct welding quality. Although such indirect weld quality does not directly indicate the performance of the weld joint, it is largely reflected in the presence of weld quality problems during the welding process.
At present, deep learning is combined with data such as visual images, arc spectrums and arc sounds to detect welding quality, a certain effect can be obtained in a laboratory environment, but in an actual use scene, the data such as the welding images, the spectrums and the arc sounds are difficult to collect, the influence of the environment is large, the defect types are difficult to define, a large amount of data labeling is needed, the time and the economic cost are high, and the practical application value is not high due to the fact that a large amount of defect labeling data are difficult to obtain. And the time sequence data such as current, voltage in the welding process not only contain power performance information, also contain a large amount of welding quality's information, the collection cost is lower simultaneously, be difficult for receiving external environment factor to influence, excavate the corresponding relation of high frequency time sequence data and welding quality defect through the analysis, can realize effective discernment and the detection of typical welding quality defect, have the lower characteristics of detection cost concurrently simultaneously, have higher practical value.
Disclosure of Invention
The purpose of the invention is as follows: the invention provides a welding quality real-time detection method and system based on inverse Fourier transform and a self-encoder, which aims at solving the technical problems that data such as welding images, spectrums, sounds and the like are difficult to acquire, defect data are difficult to label, model robustness is insufficient and the like.
The technical scheme is as follows: in order to achieve the aim, the invention provides a welding quality real-time detection method based on inverse Fourier transform and a self-encoder, which comprises the following steps:
step 1: collecting high-frequency time sequence data (such as current, voltage and the like) in a normal welding process, and acquiring a period Window and a main frequency F of a high-frequency time sequence through frequency identification;
step 2: obtaining a frequency spectrum of the high-frequency time sequence through Fourier transform, and filtering out frequencies except for F to obtain main frequency data M of the high-frequency time sequence;
and step 3: obtaining time domain data L corresponding to the main frequency data M through inverse Fourier transform, and subtracting the time domain data L by using the original time sequence to obtain Noise data Noise;
and 4, step 4: carrying out sliding Window construction on Noise data Noise by taking Window as a Window to generate a data sample set;
and 5: constructing an Autoencoder framework, performing model training by using a data sample set, and determining relevant parameters of a model;
step 6: for new high-frequency time sequence data, sliding windows are carried out according to Window, and Noise data Noise of each Window is obtained by means of Fourier transformation and inverse Fourier transformation ’ And inputting the data into a trained Autoencoder model for prediction, further acquiring a reconstruction error between reconstructed data and real data, and comparing the reconstruction error with a set threshold value K to realize real-time detection of welding quality defects.
Further, since the main frequency signal cannot be visually seen by directly performing fourier transform on the high-frequency time series data under the real condition, the following frequency identification method is proposed:
step 1.1: setting the size of a periodic window as Gap, performing sliding window on the time sequence by using the periodic window, calculating a peak value peak in each periodic window and a distance PeakInterval between adjacent peak values, and finally calculating a Score corresponding to the periodic window according to the following formula:
wherein the content of the first and second substances,
the standard deviation of the peak value distance PeakInterval is shown, the formula means that the more stable the distance between the peak values is, the longer the period length is, and the more possible the period at the moment is the period of the original sequence;
step 1.2: and traversing the periodic windows with different sizes, calculating a Score corresponding to each periodic Window, and selecting the periodic Window corresponding to the minimum Score as a periodic Window, wherein the dominant frequency F = 1/Window.
Further, the data sample set is randomly divided into a training set and a test set according to a set proportion, wherein the training set is used for training the Autoencoder model, and the test set is used for determining the threshold K, and specifically includes: and (3) transmitting the test set samples into a trained Autoencoder model for reconstruction, calculating the reconstruction errors of the reconstructed data and the sample data, and determining a threshold K according to the reconstruction errors of all the test set samples.
Further, the reconstruction error is 3/4 bits of the difference between the reconstructed data and the real data.
In addition, the invention also provides a welding quality real-time detection system based on the inverse Fourier transform and the self-encoder, which comprises a data acquisition module and a data processing module, wherein the data acquisition module comprises but is not limited to a current sensor, a voltage sensor and the like, and the data processing module carries out real-time detection on welding quality defects according to the high-frequency welding time sequence data acquired by the data acquisition module by using the welding quality real-time detection method.
Has the advantages that: compared with the prior art, the welding quality real-time detection method and system based on the inverse Fourier transform and the self-encoder provided by the invention have the following advantages:
1. compared with the method for constructing the identification model by using data such as images, sounds, spectrums and the like to diagnose the welding quality, the method only needs to use non-invasive data acquisition equipment to acquire high-frequency time sequence data in the welding process, and has the advantages of easy data acquisition and low detection cost.
2. An accurate and efficient frequency automatic identification method is designed aiming at the defect that the time sequence dominant frequency is difficult to identify through Fourier transform, meanwhile, filtering and inverse Fourier transform are combined to obtain stable noise data, finally, an Autoencoder model is used for reconstructing the noise data and obtaining an abnormal threshold value, the requirement for identifying welding quality defects is met, a large number of abnormal labels are not needed, and the defect that welding defect data are difficult to label is effectively overcome.
3. In the actual production process, time series data such as current, voltage and the like are relatively standardized and are not easily influenced by environmental factors, and a model is constructed on the basis of the time series data closely related to welding quality, so that the model is more robust and has good robustness.
Drawings
FIG. 1 is a general flowchart of a real-time welding quality detection method according to an embodiment of the present invention;
FIG. 2 is a graph of normal welding current collected in an embodiment of the present invention, wherein the abscissa is the collection time and the ordinate is the collected current data;
fig. 3 is a graph of single-sided amplitude spectrum of current obtained by fourier transform in an embodiment of the present invention, where the abscissa is current frequency and the ordinate is normalized current amplitude;
fig. 4 is a structural diagram of an Autoencoder model constructed in the embodiment of the present invention.
Detailed Description
The following description of the preferred embodiments of the present invention with reference to the accompanying drawings will more clearly and completely illustrate the technical solutions of the present invention.
Fig. 1 shows a real-time welding quality detection method based on inverse fourier transform and self-encoder, which includes the following steps:
step 1: collecting high-frequency current data in a normal welding process, and acquiring a period Window and a main frequency F of a current sequence through frequency identification;
as can be seen from fig. 2 and 3, the main frequency signal of the current sequence cannot be accurately extracted by using fourier transform, so the following frequency identification methods are proposed:
step 1.1: setting the size of a periodic window as Gap, performing non-coincident sliding on a high-frequency current sequence by using the periodic window, calculating a peak value peak in each periodic window and a distance PeakInterval between adjacent peak values, and finally calculating a Score corresponding to the periodic window according to the following formula:
wherein the content of the first and second substances,
the standard deviation of the peak value distance PeakInterval is shown, the formula means that the more stable the distance between the peak values is, the longer the period length is, and the more possible the period at the moment is the period of the original sequence;
step 1.2: and traversing the periodic windows with different sizes, calculating the Score of each periodic Window, selecting the periodic Window corresponding to the minimum Score as the periodic Window of the current sequence, and setting the main frequency F as 1/Window.
Step 2: and after the main frequency F is obtained, converting the time domain data into a frequency domain through Fourier transform to obtain a frequency spectrum of the high-frequency current sequence, and further filtering out frequencies except the frequency F to obtain the main frequency data M of the high-frequency current sequence.
And step 3: and performing inverse Fourier transform on the main frequency data M to obtain time domain data L again, and subtracting the time domain data L from the original current sequence to obtain Noise data Noise (here, if welding abnormity does not occur, normal welding current Noise is considered to be stable).
And 4, step 4: and performing sliding Window sampling on the Noise data Noise by using a Window, and performing sliding Window sampling on all generated samples according to a ratio of 8: 2 are randomly divided into a training set Train and a Test set Test.
And 5: constructing an Autoencoder framework, transmitting Train data into an Autoencoder model for training, and determining relevant parameters of the model;
as shown in fig. 4, the constructed Autoencoder model includes an Encoder (Encoder) and a Decoder (Decoder), and the abstract feature (Z) is extracted from the input high-frequency time series data (X) by the Encoder, and then the high-frequency time series data is restored by the Decoder (b)
）。
Step 6: and (3) transmitting the Test data into a trained Autoencoder model for data reconstruction, calculating 3/4 digits of difference values of the reconstructed data and the sample data for each Test sample, namely the reconstruction error, and finally taking the mean value of the reconstruction errors of all the Test samples as a threshold K.
And 7: collecting new high-frequency current data in real time, performing sliding Window according to Window, and obtaining Noise data Noise of each Window by utilizing Fourier transform and inverse Fourier transform ’ Inputting the data into a trained Autoencoder model for prediction, further acquiring 3/4 digits of difference values of reconstructed data and real data of each window, and comparing the 3/4 digits with a set threshold value K to realize real-time detection of welding quality defects.
In addition, the invention also provides a welding quality real-time detection system based on the inverse Fourier transform and the self-encoder, which comprises a data acquisition module and a data processing module, wherein the data acquisition module adopts a high-precision current sensor, and the data processing module carries out real-time detection on welding quality defects according to the high-frequency current data acquired by the data acquisition module by using the welding quality real-time detection method.
Compared with the method for constructing the identification model by using data such as images, sounds, spectra and the like to diagnose the welding quality, the method only needs non-invasive data acquisition equipment to acquire high-frequency time sequence data in the welding process, is easy to acquire data, has low detection cost, does not need a large amount of abnormal labels, effectively overcomes the defect that the welding quality defect is difficult to label, constructs the model based on the time sequence data closely related to the welding quality, and has more robust model and good robustness.
The above detailed description merely describes preferred embodiments of the present invention and does not limit the scope of the invention. Without departing from the spirit and scope of the present invention, it should be understood that various changes, substitutions and alterations can be made herein by those skilled in the art from the following detailed description and drawings.
Claims (5)
1. A welding quality real-time detection method based on inverse Fourier transform and a self-encoder is characterized by comprising the following steps:
step 1: collecting high-frequency time sequence data in a normal welding process, and acquiring a period Window and a main frequency F of a high-frequency time sequence through frequency identification;
step 2: obtaining a frequency spectrum of the high-frequency time sequence through Fourier transform, and filtering out frequencies except for F to obtain main frequency data M of the high-frequency time sequence;
and step 3: obtaining time domain data L corresponding to the main frequency data M through inverse Fourier transform, and subtracting the time domain data L by using the original time sequence to obtain Noise data Noise;
and 4, step 4: carrying out sliding Window construction on Noise data Noise by taking Window as a Window to generate a data sample set;
and 5: constructing an Autoencoder frame, performing model training by using a data sample set, and determining model parameters;
step 6: for new high-frequency time sequence data, sliding windows are carried out according to Window, and Noise data Noise of each Window is obtained by means of Fourier transformation and inverse Fourier transformation ’ And inputting the data into a trained Autoencoder model for prediction, further acquiring a reconstruction error between reconstructed data and real data, and comparing the reconstruction error with a set threshold value K to realize real-time detection of welding quality defects.
2. The method for detecting the welding quality in real time based on the inverse Fourier transform and the self-encoder according to claim 1, wherein the frequency identification method comprises the following steps:
step 1.1: setting the size of a periodic window as Gap, performing sliding window on the time sequence by using the periodic window, calculating a peak value peak in each periodic window and a distance PeakInterval between adjacent peak values, and finally calculating a Score corresponding to the periodic window according to the following formula:
wherein the content of the first and second substances,
represents the standard deviation of the peak distance peakterval;
step 1.2: and traversing the periodic windows with different sizes, calculating a Score corresponding to each periodic Window, and selecting the periodic Window corresponding to the minimum Score as a periodic Window, wherein the dominant frequency F = 1/Window.
3. The welding quality real-time detection method based on the inverse fourier transform and the self-encoder as claimed in claim 1, wherein the data sample set is randomly divided into a training set and a test set according to a set proportion, wherein the training set is used for training an Autoencoder model, and the test set is used for determining a threshold K, specifically comprising: and (3) transmitting the test set samples into a trained Autoencoder model for reconstruction, calculating the reconstruction errors of the reconstructed data and the sample data, and determining a threshold K according to the reconstruction errors of all the test set samples.
4. The method for detecting the welding quality in real time based on the inverse Fourier transform and the self-encoder is characterized in that the reconstruction error is 3/4 bits of the difference value of the reconstructed data and the real data.
5. A welding quality real-time detection system based on inverse Fourier transform and a self-encoder is characterized by comprising a data acquisition module and a data processing module, wherein the data processing module carries out real-time detection on welding quality defects according to high-frequency welding time sequence data acquired by the data acquisition module by using the detection method of any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210720862.6A CN114799610B (en) | 2022-06-24 | 2022-06-24 | Welding quality real-time detection method and system based on inverse Fourier transform and self-encoder |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210720862.6A CN114799610B (en) | 2022-06-24 | 2022-06-24 | Welding quality real-time detection method and system based on inverse Fourier transform and self-encoder |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114799610A true CN114799610A (en) | 2022-07-29 |
| CN114799610B CN114799610B (en) | 2022-10-04 |
Family
ID=82521518
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210720862.6A Active CN114799610B (en) | 2022-06-24 | 2022-06-24 | Welding quality real-time detection method and system based on inverse Fourier transform and self-encoder |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114799610B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115106615A (en) * | 2022-08-30 | 2022-09-27 | 苏芯物联技术(南京)有限公司 | Welding deviation real-time detection method and system based on intelligent working condition identification |
| CN115255567A (en) * | 2022-09-26 | 2022-11-01 | 苏芯物联技术(南京)有限公司 | Welding deviation real-time detection method and system based on frequency domain and time-frequency domain characteristics |
| CN116204830A (en) * | 2023-04-28 | 2023-06-02 | 苏芯物联技术(南京)有限公司 | Welding abnormality real-time detection method based on path aggregation network |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107832777A (en) * | 2017-10-12 | 2018-03-23 | 吉林化工学院 | A kind of electrical energy power quality disturbance recognition methods using the quick S-transformation feature extraction of time domain data compression multiresolution |
| CN113312996A (en) * | 2021-05-19 | 2021-08-27 | 哈尔滨工程大学 | Detection and identification method for aliasing short-wave communication signals |
| CN113515684A (en) * | 2020-04-09 | 2021-10-19 | 阿里巴巴集团控股有限公司 | Abnormal data detection method and device |
| CN113695713A (en) * | 2021-09-17 | 2021-11-26 | 蕴硕物联技术(上海)有限公司 | Online monitoring method and device for welding quality of inner container of water heater |
| CN113870260A (en) * | 2021-12-02 | 2021-12-31 | 苏芯物联技术(南京)有限公司 | Welding defect real-time detection method and system based on high-frequency time sequence data |
| CN113878214A (en) * | 2021-12-08 | 2022-01-04 | 苏芯物联技术(南京)有限公司 | Welding quality real-time detection method and system based on LSTM and residual distribution |
| CN114168586A (en) * | 2022-02-10 | 2022-03-11 | 北京宝兰德软件股份有限公司 | Abnormal point detection method and device |
| CN114265882A (en) * | 2021-12-24 | 2022-04-01 | 中冶赛迪重庆信息技术有限公司 | Method, system, device and medium for detecting time sequence signal point abnormity |
| CN114417699A (en) * | 2021-12-10 | 2022-04-29 | 烟台杰瑞石油服务集团股份有限公司 | Pump valve fault detection method |
-
2022
- 2022-06-24 CN CN202210720862.6A patent/CN114799610B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107832777A (en) * | 2017-10-12 | 2018-03-23 | 吉林化工学院 | A kind of electrical energy power quality disturbance recognition methods using the quick S-transformation feature extraction of time domain data compression multiresolution |
| CN113515684A (en) * | 2020-04-09 | 2021-10-19 | 阿里巴巴集团控股有限公司 | Abnormal data detection method and device |
| CN113312996A (en) * | 2021-05-19 | 2021-08-27 | 哈尔滨工程大学 | Detection and identification method for aliasing short-wave communication signals |
| CN113695713A (en) * | 2021-09-17 | 2021-11-26 | 蕴硕物联技术(上海)有限公司 | Online monitoring method and device for welding quality of inner container of water heater |
| CN113870260A (en) * | 2021-12-02 | 2021-12-31 | 苏芯物联技术(南京)有限公司 | Welding defect real-time detection method and system based on high-frequency time sequence data |
| CN113878214A (en) * | 2021-12-08 | 2022-01-04 | 苏芯物联技术(南京)有限公司 | Welding quality real-time detection method and system based on LSTM and residual distribution |
| CN114417699A (en) * | 2021-12-10 | 2022-04-29 | 烟台杰瑞石油服务集团股份有限公司 | Pump valve fault detection method |
| CN114265882A (en) * | 2021-12-24 | 2022-04-01 | 中冶赛迪重庆信息技术有限公司 | Method, system, device and medium for detecting time sequence signal point abnormity |
| CN114168586A (en) * | 2022-02-10 | 2022-03-11 | 北京宝兰德软件股份有限公司 | Abnormal point detection method and device |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115106615A (en) * | 2022-08-30 | 2022-09-27 | 苏芯物联技术(南京)有限公司 | Welding deviation real-time detection method and system based on intelligent working condition identification |
| CN115255567A (en) * | 2022-09-26 | 2022-11-01 | 苏芯物联技术(南京)有限公司 | Welding deviation real-time detection method and system based on frequency domain and time-frequency domain characteristics |
| CN116204830A (en) * | 2023-04-28 | 2023-06-02 | 苏芯物联技术(南京)有限公司 | Welding abnormality real-time detection method based on path aggregation network |
| CN116204830B (en) * | 2023-04-28 | 2023-07-11 | 苏芯物联技术(南京)有限公司 | Welding abnormality real-time detection method based on path aggregation network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114799610B (en) | 2022-10-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114799610B (en) | Welding quality real-time detection method and system based on inverse Fourier transform and self-encoder | |
| CN111024728B (en) | Railway detection method and system based on computer vision and ultrasonic flaw detection | |
| CN114722883B (en) | Welding quality real-time detection method and system based on high-frequency time sequence data | |
| CN109649432B (en) | System and method for monitoring integrity of steel rail of cloud platform based on guided wave technology | |
| CN109855874B (en) | Random resonance filter for enhancing detection of weak signals in vibration assisted by sound | |
| CN102980894B (en) | Steel structure special type weld nondestructive detection system and method | |
| CN113878214B (en) | Welding quality real-time detection method and system based on LSTM and residual distribution | |
| CN105913059A (en) | Vehicle VIN code automatic identifying system and control method therefor | |
| CN111753877B (en) | Product quality detection method based on deep neural network migration learning | |
| CN113805018A (en) | Intelligent identification method for partial discharge fault type of 10kV cable of power distribution network | |
| CN114700587A (en) | Missing welding defect real-time detection method and system based on fuzzy reasoning and edge calculation | |
| CN115144259B (en) | Method and system for detecting deformation resistance of steel | |
| CN103954628A (en) | Ensemble empirical mode decomposition (EEMD) and approximate entropy combined steel tube damage monitoring method | |
| US20230084562A1 (en) | Non-destructive inspection method and system based on artificial intelligence | |
| CN115255567B (en) | Welding deviation real-time detection method and system based on frequency domain and time-frequency domain characteristics | |
| Hua et al. | Matching linear Chirplet strategy-based synchroextracting transform and its application to rotating machinery fault diagnosis | |
| CN112347903B (en) | Multi-component pipeline identification method based on heterogeneous field signals | |
| CN115015375A (en) | Method for identifying middle ring welding seam in pipeline detection | |
| CN115165885A (en) | Identification system and method based on machine vision identification and spectral measurement | |
| CN115106615A (en) | Welding deviation real-time detection method and system based on intelligent working condition identification | |
| CN113672859A (en) | Switch point machine fault acoustic diagnosis system | |
| Tang et al. | Sliding Window Dynamic Time-Series Warping-Based Ultrasonic Guided Wave Temperature Compensation and Defect Monitoring Method for Turnout Rail Foot | |
| CN111912910A (en) | Intelligent identification method for polyethylene pipeline hot-melt weld joint hybrid ultrasonic scanning defects | |
| Imaouchen et al. | Complexity based on synchrosqueezing analysis in gear diagnosis | |
| EP4261535A1 (en) | Method for automatic flawless tube detection |
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 |