CN113705289A - Method for improving nondestructive testing precision based on machine learning - Google Patents
Method for improving nondestructive testing precision based on machine learning Download PDFInfo
- Publication number
- CN113705289A CN113705289A CN202110091636.1A CN202110091636A CN113705289A CN 113705289 A CN113705289 A CN 113705289A CN 202110091636 A CN202110091636 A CN 202110091636A CN 113705289 A CN113705289 A CN 113705289A
- Authority
- CN
- China
- Prior art keywords
- data
- machine learning
- detection
- improving
- point
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000010801 machine learning Methods 0.000 title claims abstract description 33
- 238000000034 method Methods 0.000 title claims abstract description 27
- 238000009659 non-destructive testing Methods 0.000 title claims abstract description 17
- 238000001514 detection method Methods 0.000 claims abstract description 83
- 238000012549 training Methods 0.000 claims abstract description 21
- 238000013473 artificial intelligence Methods 0.000 claims abstract description 14
- 238000004458 analytical method Methods 0.000 claims abstract description 9
- 238000012795 verification Methods 0.000 claims abstract description 6
- 230000001502 supplementing effect Effects 0.000 claims abstract description 4
- 230000005284 excitation Effects 0.000 claims description 6
- 230000008859 change Effects 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 5
- 238000007405 data analysis Methods 0.000 claims description 4
- 238000013500 data storage Methods 0.000 claims description 3
- 230000000644 propagated effect Effects 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 abstract description 2
- 230000007547 defect Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000002950 deficient Effects 0.000 description 2
- 230000001902 propagating effect Effects 0.000 description 2
- 239000013589 supplement Substances 0.000 description 2
- 230000007704 transition Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/02—Preprocessing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/14—Fourier, Walsh or analogous domain transformations, e.g. Laplace, Hilbert, Karhunen-Loeve, transforms
- G06F17/141—Discrete Fourier transforms
- G06F17/142—Fast Fourier transforms, e.g. using a Cooley-Tukey type algorithm
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- 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
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Analysis (AREA)
- Evolutionary Computation (AREA)
- Mathematical Optimization (AREA)
- Computing Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computational Mathematics (AREA)
- Discrete Mathematics (AREA)
- Medical Informatics (AREA)
- Signal Processing (AREA)
- Algebra (AREA)
- Databases & Information Systems (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Investigating Or Analyzing Materials By The Use Of Ultrasonic Waves (AREA)
Abstract
The invention discloses a method for improving nondestructive testing precision based on machine learning, and particularly relates to the technical field of machine learning, wherein the specific method comprises the following steps: (1) and establishing a system: establishing a machine learning modeling system; (2) and establishing a model: establishing an artificial intelligence basic model, and collecting detection data as learning data; (3) and machine learning training: importing the detection data into a machine learning modeling system for learning training; (4) optimizing a detection model: the detection model is optimized by analyzing various different detection data and verifying the analysis results, making examples of the data and the verification results and supplementing the examples to a training set. The method has the advantages of more analysis target parameters, accurate judgment, high automation degree, improvement of the utilization of the target parameters compared with the traditional technology, adoption of artificial intelligent judgment, abandonment of traditional human experience intervention, and effective improvement of the precision of the detection result.
Description
Technical Field
The invention relates to the technical field of machine learning, in particular to a method for improving nondestructive testing precision based on machine learning.
Background
The existing engineering nondestructive testing technology utilizes a signal excitation device and a signal receiving device to carry out data acquisition work, then utilizes programmed software to analyze data to obtain one or two required target parameters, and a tester judges according to a judgment standard and by combining experience to give a result. The results obtained by the traditional data analysis and analysis method have certain defects such as: the software is programmed only for obtaining one or two target parameters, except the target parameters, other available parameters can not participate in the judgment of the result; some analysis results need to be judged by combining with experience, and the judgment result cannot be accurate and self-enabled.
Therefore, a method for improving the nondestructive testing precision based on machine learning is provided to solve the renting problem of the water purifier.
Disclosure of Invention
In order to overcome the above-mentioned drawbacks of the prior art, embodiments of the present invention provide a method for improving the accuracy of nondestructive testing based on machine learning, so as to solve the problems mentioned in the above background art.
In order to achieve the purpose, the invention provides the following technical scheme:
a method for improving nondestructive testing precision based on machine learning comprises the following specific steps:
(1) and establishing a system: establishing a machine learning modeling system;
(2) and establishing a model: establishing an artificial intelligence basic model, and collecting detection data as learning data;
(3) and machine learning training: importing the detection data into a machine learning modeling system for learning training;
(4) optimizing a detection model: analyzing various different detection data, verifying the analysis result, making the data and the verification result into examples, and supplementing the examples to a training set so as to optimize a detection model;
(5) and intelligently detecting: and importing the detection data to be analyzed into the trained artificial intelligence basic model, and generating a detection result by the artificial intelligence basic model.
On the basis of the technical scheme, the detected object is collected when the detection data are collected, the detection point is marked on the detected object, and the signal excitation device is used for exciting the signal so that the signal is transmitted inside the detected object.
On the basis of the technical scheme, the signal receiving device is used for collecting the detection data, collecting the signals propagated in the structure point by point along the detection point of the mark, and fully analyzing the signals.
On the basis of the technical scheme, the detected object is analyzed when the detection data are collected, the state of each detection point is marked, and the state of each detection point is matched with the acquired signal to serve as learning data.
On the basis of the above technical solution, the signal detected at each measurement point in the detection data collection process includes 11 parameters: t _ FFT [0], T _ FFT [1], T _ FFT [2 ]: 3-order extreme points before the amplitude of the signal after Fourier transform; t _ MEM [0], T _ MEM [1], and T _ MEM [2 ]: the extreme point of the first 3 orders of the maximum entropy processing; r _ FFT [1], R _ FFT [2 ]: after Fourier change, the second order amplitude and the third order amplitude are relative to the relative amplitude of the first order extreme point; t _ fsths: the first wavelength is half wavelength; VRF (km/s): the standard wave speed; v _ VPB (km/s): and (6) testing the wave speed.
On the basis of the technical scheme, the machine learning modeling system comprises a data module, a training learning module, a summing module, a modeling module and a display module.
On the basis of the technical scheme, the data module comprises a data acquisition unit, a data storage unit and a data analysis unit, and the display module comprises a display unit.
The invention has the technical effects and advantages that:
compared with the prior art, the method obtains and analyzes data through a data module, trains machines corresponding to the detection data through a training learning module, analyzes various different detection data, verifies the analysis result, makes an example of the data and the verification result, supplements the example to a training set, further optimizes a detection model, introduces the detection data to be analyzed into an artificial intelligence basic model after training, generates the detection result through the artificial intelligence basic model, collects the detected object when collecting the detection data, marks a detection point on the detected object, uses a signal excitation device to excite a signal to propagate inside the detected object, uses a signal receiving device to collect the signals after propagating inside a structure point by point along the marked detection point when collecting the detection data, fully analyzes the signal, and analyzes the detected object when collecting the detection data, marking the state of each detection point, matching the state of each detection point with the collected signals to be used as learning data, wherein the signals detected by each detection point in the process of collecting detection data comprise 11 parameters: t _ FFT [0], T _ FFT [1], T _ FFT [2 ]: 3-order extreme points before the amplitude of the signal after Fourier transform; t _ MEM [0], T _ MEM [1], and T _ MEM [2 ]: the extreme point of the first 3 orders of the maximum entropy processing; r _ FFT [1], R _ FFT [2 ]: after Fourier change, the second order amplitude and the third order amplitude are relative to the relative amplitude of the first order extreme point; t _ fsths: the first wavelength is half wavelength; VRF (km/s): the standard wave speed; v _ VPB (km/s): the method is characterized in that the wave velocity is tested, the state class of the detection points is divided into 3 types including SOUND SOUND, defective DEFECT and transition section UNCERTAIN, and the state class of the detection points is connected with T _ FFT [0], T _ FFT [1], T _ FFT [2], T _ MEM [0], T _ MEM [1], T _ MEM [2], R _ FFT [1], R _ FFT [2], T _ FstHfS, VRF and V _ VPB in a dividing mode.
Drawings
FIG. 1 is a flow chart of a method for improving nondestructive testing accuracy based on machine learning according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention is further described with reference to the following drawings and detailed description:
referring to fig. 1, a method for improving the accuracy of nondestructive testing based on machine learning according to an embodiment of the present invention includes the following steps:
s101, establishing a system: establishing a machine learning modeling system;
s103, model establishment: establishing an artificial intelligence basic model, and collecting detection data as learning data;
s105, machine learning training: importing the detection data into a machine learning modeling system for learning training;
s107, optimizing a detection model: analyzing various different detection data, verifying the analysis result, making the data and the verification result into examples, and supplementing the examples to a training set so as to optimize a detection model;
s109, intelligent detection: and importing the detection data to be analyzed into the trained artificial intelligence basic model, and generating a detection result by the artificial intelligence basic model.
Furthermore, the detected object is collected when the detection data is collected, the detection point is marked on the detected object, and the signal excitation device is used for exciting the signal so that the signal is transmitted in the detected object.
Furthermore, a signal receiving device is used for collecting detection data, signals propagated in the structure are collected point by point along the marked detection points, and the signals are fully analyzed.
Furthermore, the detected object is analyzed when the detection data is collected, the state of each detection point is marked, and the state of each detection point is matched with the collected signal to be used as learning data.
Further, the signal detected at each station in the collection of detection data includes 11 parameters: t _ FFT [0], T _ FFT [1], T _ FFT [2 ]: 3-order extreme points before the amplitude of the signal after Fourier transform; t _ MEM [0], T _ MEM [1], and T _ MEM [2 ]: the extreme point of the first 3 orders of the maximum entropy processing; r _ FFT [1], R _ FFT [2 ]: after Fourier change, the second order amplitude and the third order amplitude are relative to the relative amplitude of the first order extreme point; t _ fsths: the first wavelength is half wavelength; VRF (km/s): the standard wave speed; v _ VPB (km/s): and (6) testing the wave speed.
Further, the machine learning modeling system comprises a data module 1, a training learning module 2, a summing module 3, a modeling module 4 and a display module 5.
Further, the data module 1 includes a data obtaining unit 6, a data storage unit 7 and a data analysis unit 8, and the display module 5 includes a display unit 9.
The working principle of the invention is as follows: the method of the invention obtains and analyzes data through a data module 1, trains machines corresponding to the detection data through a training learning module 2, analyzes various different detection data, verifies the analysis result, makes an example of the data and the verification result and supplements the example to a training set so as to optimize a detection model, introduces the detection data to be analyzed into an artificial intelligence basic model after training, generates the detection result by the artificial intelligence basic model, collects the detected object when collecting the detection data, marks a detection point on the detected object, uses a signal excitation device to excite a signal to propagate the signal inside the detected object, uses a signal receiving device when collecting the detection data, collects the signal after propagating inside a structure point by point along the marked detection point, fully analyzes the signal, and analyzes the detected object when collecting the detection data, marking the state of each detection point, matching the state of each detection point with the collected signals to be used as learning data, wherein the signals detected by each detection point in the process of collecting detection data comprise 11 parameters: t _ FFT [0], T _ FFT [1], T _ FFT [2 ]: 3-order extreme points before the amplitude of the signal after Fourier transform; t _ MEM [0], T _ MEM [1], and T _ MEM [2 ]: the extreme point of the first 3 orders of the maximum entropy processing; r _ FFT [1], R _ FFT [2 ]: after Fourier change, the second order amplitude and the third order amplitude are relative to the relative amplitude of the first order extreme point; t _ fsths: the first wavelength is half wavelength; VRF (km/s): the standard wave speed; v _ VPB (km/s): the method is characterized in that the wave velocity is tested, the state class of the detection points is divided into 3 types including SOUND SOUND, defective DEFECT and transition section UNCERTAIN, and the state class of the detection points is connected with T _ FFT [0], T _ FFT [1], T _ FFT [2], T _ MEM [0], T _ MEM [1], T _ MEM [2], R _ FFT [1], R _ FFT [2], T _ FstHfS, VRF and V _ VPB in a dividing mode.
The points to be finally explained are: first, in the description of the present application, it should be noted that, unless otherwise specified and limited, the terms "mounted," "connected," and "connected" should be understood broadly, and may be a mechanical connection or an electrical connection, or a communication between two elements, and may be a direct connection, and "upper," "lower," "left," and "right" are only used to indicate a relative positional relationship, and when the absolute position of the object to be described is changed, the relative positional relationship may be changed;
secondly, the method comprises the following steps: in the drawings of the disclosed embodiments of the invention, only the structures related to the disclosed embodiments are referred to, other structures can refer to common designs, and the same embodiment and different embodiments of the invention can be combined with each other without conflict;
and finally: the above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that are within the spirit and principle of the present invention are intended to be included in the scope of the present invention.
Claims (7)
1. A method for improving nondestructive testing precision based on machine learning is characterized in that: the specific method comprises the following steps:
(1) and establishing a system: establishing a machine learning modeling system;
(2) and establishing a model: establishing an artificial intelligence basic model, and collecting detection data as learning data;
(3) and machine learning training: importing the detection data into a machine learning modeling system for learning training;
(4) optimizing a detection model: analyzing various different detection data, verifying the analysis result, making the data and the verification result into examples, and supplementing the examples to a training set so as to optimize a detection model;
(5) and intelligently detecting: and importing the detection data to be analyzed into the trained artificial intelligence basic model, and generating a detection result by the artificial intelligence basic model.
2. The method for improving the accuracy of nondestructive testing based on machine learning of claim 1, wherein: and collecting the detected object when detecting data, marking a detection point on the detected object, and exciting a signal by using a signal excitation device to enable the signal to be transmitted in the detected object.
3. The method for improving the accuracy of nondestructive testing based on machine learning according to claim 2, wherein: and when the detection data is collected, a signal receiving device is used for collecting the signals propagated in the structure point by point along the marked detection point and fully analyzing the signals.
4. The method for improving the accuracy of nondestructive testing based on machine learning of claim 3, wherein: and analyzing the detected object when the detection data is collected, marking the state of each detection point, and matching the state of each detection point with the acquired signal to obtain learning data.
5. The method for improving the accuracy of nondestructive testing based on machine learning of claim 4, wherein: the signal detected at each measuring point in the collection of detection data comprises 11 parameters: t _ FFT [0], T _ FFT [1], T _ FFT [2 ]: 3-order extreme points before the amplitude of the signal after Fourier transform; t _ MEM [0], T _ MEM [1], and T _ MEM [2 ]: the extreme point of the first 3 orders of the maximum entropy processing; r _ FFT [1], R _ FFT [2 ]: after Fourier change, the second order amplitude and the third order amplitude are relative to the relative amplitude of the first order extreme point; t _ fsths: the first wavelength is half wavelength; VRF (km/s): the standard wave speed; v _ VPB (km/s): and (6) testing the wave speed.
6. The method for improving the accuracy of nondestructive testing based on machine learning of claim 5, wherein: the machine learning modeling system comprises a data module (1), a training learning module (2), a summing module (3), a modeling module (4) and a display module (5).
7. The method for improving the accuracy of nondestructive testing based on machine learning of claim 6, wherein: the data module (1) comprises a data acquisition unit (6), a data storage unit (7) and a data analysis unit (8), and the display module (5) comprises a display unit (9).
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110091636.1A CN113705289A (en) | 2021-01-23 | 2021-01-23 | Method for improving nondestructive testing precision based on machine learning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110091636.1A CN113705289A (en) | 2021-01-23 | 2021-01-23 | Method for improving nondestructive testing precision based on machine learning |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113705289A true CN113705289A (en) | 2021-11-26 |
Family
ID=78647689
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110091636.1A Pending CN113705289A (en) | 2021-01-23 | 2021-01-23 | Method for improving nondestructive testing precision based on machine learning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113705289A (en) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108491931A (en) * | 2018-03-29 | 2018-09-04 | 四川升拓检测技术股份有限公司 | A method of non-destructive testing precision is improved based on machine learning |
CN110082429A (en) * | 2019-04-19 | 2019-08-02 | 四川升拓检测技术股份有限公司 | A kind of auxiliary judgement method of the tunnel-liner non-destructive testing of combination machine learning |
CN112213396A (en) * | 2020-09-18 | 2021-01-12 | 同济大学 | Nondestructive testing method for ballastless track plate of track traffic |
-
2021
- 2021-01-23 CN CN202110091636.1A patent/CN113705289A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108491931A (en) * | 2018-03-29 | 2018-09-04 | 四川升拓检测技术股份有限公司 | A method of non-destructive testing precision is improved based on machine learning |
CN110082429A (en) * | 2019-04-19 | 2019-08-02 | 四川升拓检测技术股份有限公司 | A kind of auxiliary judgement method of the tunnel-liner non-destructive testing of combination machine learning |
CN112213396A (en) * | 2020-09-18 | 2021-01-12 | 同济大学 | Nondestructive testing method for ballastless track plate of track traffic |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102288883B (en) | Oscillation wave partial discharge identifying and positioning method for asynchronous double-end power cable | |
CN107144635B (en) | A kind of detection method of wood materials ultrasonic non-destructive classified detection system | |
CN109190272B (en) | Concrete structure defect detection method based on elastic waves and machine learning | |
CN113324648B (en) | Portable high-speed railway wheel rail vibration space-time synchronization test method and system | |
CN104063873A (en) | Shaft sleeve part surface defect on-line detection method based on compressed sensing | |
CN112710871B (en) | Test method and device for positioning speed measurement system host | |
CN106274979A (en) | Rail wear automatic detection device | |
CN113705412A (en) | Deep learning-based multi-source data fusion track state detection method and device | |
CN108844856B (en) | Sleeve grouting defect nondestructive testing method based on impact elastic wave and machine learning | |
CN110441388A (en) | A kind of jacket platform underwater portion non-destructive testing method of detection | |
CN103712561A (en) | Testing method and testing device of cooling assembly with jump testing function and rotating-speed testing function | |
CN112722010A (en) | Rail corrugation acoustic diagnosis system for rail transit | |
CN114113332A (en) | Prefabricated column sleeve grouting nondestructive testing method based on elastic waves and machine learning | |
CN103995147B (en) | A kind of Data Post Processing System being applicable to acoustic Doppler velocimetry and application | |
CN114055249B (en) | Bore hole surface roughness monitoring method and system and storage medium | |
CN104132949B (en) | A kind of device and method of radial ply tyre moulds surface defect Non-Destructive Testing | |
CN113705289A (en) | Method for improving nondestructive testing precision based on machine learning | |
CN109766794B (en) | Automatic real-time road detection method and system thereof | |
CN103868689B (en) | Vibration frequency analysis-based gear defect rapid detection system and method | |
TWM538518U (en) | Simplified road flatness inspection device | |
CN102980536A (en) | Brake pad steel backing planeness detecting system and method thereof | |
CN112049787A (en) | Non-contact water pump fault detection method and system | |
CN109444622B (en) | Automatic fault detection method, system and device for shaft angle sending system | |
CN110065522B (en) | Steel rail stress trend change detection method | |
CN105975710A (en) | Bad data set detection and recognition method for synchronous generator on-line parameter identification |
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 |