CN110274960B - Steel pipe microscopic structure evaluation method and device based on nonlinear ultrasound - Google Patents
Steel pipe microscopic structure evaluation method and device based on nonlinear ultrasound Download PDFInfo
- Publication number
- CN110274960B CN110274960B CN201910712674.7A CN201910712674A CN110274960B CN 110274960 B CN110274960 B CN 110274960B CN 201910712674 A CN201910712674 A CN 201910712674A CN 110274960 B CN110274960 B CN 110274960B
- Authority
- CN
- China
- Prior art keywords
- neural network
- steel pipe
- ultrasonic
- signal
- evaluated
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N29/00—Investigating or analysing materials by the use of ultrasonic, sonic or infrasonic waves; Visualisation of the interior of objects by transmitting ultrasonic or sonic waves through the object
- G01N29/04—Analysing solids
- G01N29/07—Analysing solids by measuring propagation velocity or propagation time of acoustic waves
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N29/00—Investigating or analysing materials by the use of ultrasonic, sonic or infrasonic waves; Visualisation of the interior of objects by transmitting ultrasonic or sonic waves through the object
- G01N29/44—Processing the detected response signal, e.g. electronic circuits specially adapted therefor
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2291/00—Indexing codes associated with group G01N29/00
- G01N2291/02—Indexing codes associated with the analysed material
- G01N2291/023—Solids
- G01N2291/0234—Metals, e.g. steel
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2291/00—Indexing codes associated with group G01N29/00
- G01N2291/02—Indexing codes associated with the analysed material
- G01N2291/028—Material parameters
- G01N2291/0289—Internal structure, e.g. defects, grain size, texture
Landscapes
- Physics & Mathematics (AREA)
- Biochemistry (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Acoustics & Sound (AREA)
- Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Investigating Or Analyzing Materials By The Use Of Ultrasonic Waves (AREA)
Abstract
The invention discloses a steel pipe microscopic structure evaluation method and a steel pipe microscopic structure evaluation device based on nonlinear ultrasound, wherein the method comprises the following steps: detecting a steel pipe to be evaluated by using ultrasonic transmitting equipment, and obtaining an ultrasonic signal and a sound velocity; and identifying the ultrasonic signal and the sound velocity of the steel pipe to be evaluated by using a pre-trained target BP neural network, and acquiring the microstructure parameters of the steel pipe to be evaluated. By applying the embodiment of the invention, the BP neural network is used for evaluating the steel pipe microstructure, compared with the defects that heavy equipment is used and the measurement position is limited in the prior art, the acquisition position of the ultrasonic signal is not limited, and the ultrasonic equipment has smaller volume and is lighter, so that the embodiment of the invention can evaluate the steel pipe microstructure more conveniently and is convenient for steel pipe detection.
Description
Technical Field
The invention relates to a steel pipe evaluation method and a steel pipe evaluation device, in particular to a steel pipe microscopic structure evaluation method and a steel pipe microscopic structure evaluation device based on nonlinear ultrasound.
Background
Steel pipes, particularly P91 steel, are a typical martensitic heat resistant steel, which is widely used in power plant boiler manufacturing, and therefore, the health status of steel pipes needs to be evaluated. In a nonlinear ultrasonic detection method (CN108107111A) for heat-resistant steel components, the inventor first constructed a quantitative model between nonlinear ultrasonic parameters and performance degradation parameters, and then performed corresponding detection based on the model; in a nondestructive testing and evaluation method for residual life of a plate-shaped metal member/material (cn201510465209.x), the inventor first obtains a relation curve between the residual life time fraction of the metal member/material and an ultrasonic Lamb wave second harmonic normalized value, and performs life evaluation on the basis of the relation curve; in "non-linear ultrasonic mixing method for detecting structural fatigue crack direction (CN 201810646120.7)", the inventors detected the direction of fatigue crack by using non-linear ultrasound; in "a nonlinear torsional mode ultrasonic guided wave method for evaluating micro-damage of metal round tube" (CN201810436079.0), "the inventor used nonlinear ultrasound to evaluate micro-damage of metal round tube.
However, all of the above techniques can be applied only in evaluating the properties, cracks and damages of the material. The microstructure change of the steel pipe includes various aspects, such as: coarsening of carbides, reduction of dislocation density, growth of crystal grains, etc., which also have a serious influence on the health of the steel pipe. In order to solve the problems, the conventional microstructure detection means is an improved indentation method, but the improved indentation method is heavy, so that the measurement position of the improved indentation method is limited, and the application of the improved indentation method is greatly limited.
Therefore, the technical problem that steel pipe detection is inconvenient exists in the prior art.
Disclosure of Invention
The invention aims to solve the technical problem of providing a steel pipe microstructure evaluation method and device based on nonlinear ultrasound so as to solve the technical problem of inconvenience in steel pipe detection in the prior art.
The invention solves the technical problems through the following technical scheme:
the embodiment of the invention provides a steel pipe microscopic structure evaluation method based on nonlinear ultrasound, which comprises the following steps:
detecting a steel pipe to be evaluated by using ultrasonic transmitting equipment, and obtaining an ultrasonic signal and a sound velocity;
and identifying the ultrasonic signal and the sound velocity of the steel pipe to be evaluated by using a pre-trained target BP neural network, and acquiring the microstructure parameters of the steel pipe to be evaluated.
Optionally, the detecting the steel pipe to be evaluated by using the ultrasonic wave emitting device and obtaining the ultrasonic signal and the sound velocity includes:
detecting a steel pipe to be evaluated by using ultrasonic transmitting equipment, and collecting ultrasonic echoes;
and filtering the ultrasonic echo to obtain an ultrasonic signal.
Optionally, the training process of the target BP neural network includes:
detecting an ultrasonic signal of each pipeline sample by using ultrasonic transmitting equipment;
acquiring microstructure parameters of the pipeline sample;
constructing a sample set according to the ultrasonic signals and the microscopic tissue parameters, and dividing the sample set into a training set and a testing set;
training a pre-constructed BP neural network by using the training set until convergence; testing the converged BP neural network by using a test set, judging whether the accuracy of the converged BP neural network is greater than or equal to a preset threshold value, and if so, taking the converged BP neural network as a target BP neural network; if not, adjusting the weight and the hyper-parameters of the parameters in the BP neural network, and returning to execute the BP neural network which is constructed in advance by using the training set until convergence.
Optionally, the ultrasonic signal includes: one or a combination of a first order nonlinear signal, a second order nonlinear signal, a third order nonlinear signal, a fourth order nonlinear signal, and ultrasonic sound speed.
Optionally, the microstructure parameters include: one or a combination of carbide size, grain size, and dislocation density.
The embodiment of the invention provides a steel pipe microscopic structure evaluation device based on nonlinear ultrasound, which comprises:
the acquisition module is used for detecting the steel pipe to be evaluated by using ultrasonic transmitting equipment and acquiring an ultrasonic signal and a sound velocity;
and the identification module is used for identifying the ultrasonic signal and the sound velocity of the steel pipe to be evaluated by using a pre-trained target BP neural network to acquire the microstructure parameters of the steel pipe to be evaluated.
Optionally, the obtaining module is configured to:
detecting a steel pipe to be evaluated by using ultrasonic transmitting equipment, and collecting ultrasonic echoes;
and filtering the ultrasonic echo to obtain an ultrasonic signal.
Optionally, the training process of the target BP neural network includes:
detecting an ultrasonic signal of each pipeline sample by using ultrasonic transmitting equipment;
acquiring microstructure parameters of the pipeline sample;
constructing a sample set according to the ultrasonic signals and the microscopic tissue parameters, and dividing the sample set into a training set and a testing set;
training a pre-constructed BP neural network by using the training set until convergence; testing the converged BP neural network by using the test set, judging whether the accuracy of the converged BP neural network is greater than or equal to a preset threshold value, and if so, taking the converged BP neural network as a target BP neural network; if not, adjusting the weight and the hyper-parameters of the parameters in the BP neural network, and returning to execute the BP neural network which is constructed in advance by using the training set until convergence.
Optionally, the ultrasonic signal includes: one or a combination of a first order nonlinear signal, a second order nonlinear signal, a third order nonlinear signal, a fourth order nonlinear signal, and ultrasonic sound speed.
Optionally, the microstructure parameters include: one or a combination of carbide size, grain size, and dislocation density.
Compared with the prior art, the invention has the following advantages:
by applying the embodiment of the invention, the BP neural network is used for evaluating the steel pipe microstructure, compared with the defects that heavy equipment is used and the measurement position is limited in the prior art, the acquisition position of the ultrasonic signal is not limited, and the ultrasonic equipment has smaller volume and is lighter, so that the embodiment of the invention can evaluate the steel pipe microstructure more conveniently and is convenient for steel pipe detection.
Drawings
FIG. 1 is a schematic flow chart of a steel pipe microstructure evaluation method based on nonlinear ultrasound according to an embodiment of the present invention;
FIG. 2 is a schematic structural diagram of a steel pipe microstructure evaluation system based on nonlinear ultrasound according to an embodiment of the present invention;
FIG. 3 is a schematic structural diagram of a BP neural network used in a steel tube microstructure evaluation method based on nonlinear ultrasound according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of a steel pipe microstructure evaluation apparatus based on nonlinear ultrasound according to an embodiment of the present invention.
Detailed Description
The following examples are given for the detailed implementation and the specific operation procedures, but the scope of the present invention is not limited to the following examples.
The embodiment of the invention provides a steel pipe microstructure evaluation method and device based on nonlinear ultrasound, and firstly introduces the steel pipe microstructure evaluation method based on nonlinear ultrasound provided by the embodiment of the invention.
Fig. 1 is a schematic flow chart of a steel pipe microstructure evaluation method based on nonlinear ultrasound according to an embodiment of the present invention, and as shown in fig. 1, the method includes:
s101: and detecting the steel pipe to be evaluated by using ultrasonic wave transmitting equipment, and obtaining an ultrasonic signal and a sound velocity.
Fig. 2 is a schematic structural diagram of a steel pipe microstructure evaluation system based on nonlinear ultrasound according to an embodiment of the present invention, as shown in fig. 2,
the steel pipe microscopic structure evaluation system mainly comprises a central controller, an ultrasonic generator, a power amplifier, a signal transmitter, a signal receiver, a filtering module, a signal processing module, a god network module, a storage module and a display, wherein,
taking a P91 steel pipe as an example, a central controller sends an instruction to an ultrasonic generator, and the ultrasonic generator sends ultrasonic waves which are amplified by a power amplifier; the amplified ultrasonic wave reaches a signal transmitter, and then the signal enters a P91 steel pipeline and is received by a signal receiver; the received signals reach a signal processing module after passing through a filtering module, and the signal processing module transforms the ultrasonic signals to obtain frequency domain signals; the central controller receives the frequency domain signal sent by the model processing module, and then processes the frequency domain signal to obtain: five indexes of a first-order nonlinear signal, a second-order nonlinear signal, a third-order nonlinear signal, a fourth-order nonlinear signal and ultrasonic sound velocity; the central controller transmits five indexes of a first-order nonlinear signal, a second-order nonlinear signal, a third-order nonlinear signal, a fourth-order nonlinear signal and ultrasonic sound velocity to the neural network module, and the neural network module calculates three indexes of grain size, carbide size and dislocation density of P91 steel and transmits the three indexes back to the central controller; the central controller stores and displays the three indexes on a display screen.
Specifically, an ultrasonic wave emitting device can be used for detecting a steel pipe to be evaluated, such as a P91 steel pipe, and collecting an ultrasonic echo of the P91 steel pipe;
carrying out time domain to frequency domain conversion on the ultrasonic echo signal, and when the signal emission frequency is F, respectively selecting corresponding amplitude values at 1F, 2F, 3F and 4F, wherein the four amplitude values are four nonlinear signals;
carrying out filtering processing on the ultrasonic echo to obtain an ultrasonic signal, wherein the ultrasonic signal comprises: one or a combination of a first order nonlinear signal, a second order nonlinear signal, a third order nonlinear signal, a fourth order nonlinear signal, and ultrasonic sound speed.
The distance S of the probe, the time of transmitting the signal by the transmitting probe is t1, the time of receiving the signal by the receiving probe is t2, and the sound velocity is as follows: v is S/(t2-t 1).
S102: and identifying the ultrasonic signal and the sound velocity of the steel pipe to be evaluated by using a pre-trained target BP neural network, and acquiring the microstructure parameters of the steel pipe to be evaluated.
Specifically, the microstructure parameters of the pipeline sample can be obtained by the following method:
after normalizing at 1060 ℃, P91 steel pipes can be respectively treated according to the following tempering process conditions:
tempering at 700 deg.C for 1 hr, 2 hr, 3 hr, 4 hr, 5 hr, 6 hr, 7 hr, 8 hr, 9 hr, 10 hr;
tempering at 720 ℃ for 1 hour, 2 hours, 3 hours, 4 hours, 5 hours, 6 hours, 7 hours, 8 hours, 9 hours, 10 hours;
tempering at 740 ℃ for 1 hour, 2 hours, 3 hours, 4 hours, 5 hours, 6 hours, 7 hours, 8 hours, 9 hours and 10 hours;
tempering at 760 ℃ for 1 hour, 2 hours, 3 hours, 4 hours, 5 hours, 6 hours, 7 hours, 8 hours, 9 hours, 10 hours;
tempering at 780 ℃ for 1 hour, 2 hours, 3 hours, 4 hours, 5 hours, 6 hours, 7 hours, 8 hours, 9 hours, 10 hours;
tempering at 800 deg.C for 1 hr, 2 hr, 3 hr, 4 hr, 5 hr, 6 hr, 7 hr, 8 hr, 9 hr, 10 hr;
the total number of 60 groups of P91 steel pipes with different grain sizes, dislocation densities and carbide sizes were obtained.
Then, the 60 groups of steel pipe samples are used for 500h, 1000h, 1500h, 2000h, 2500h and 3000h under the use conditions of 650 ℃/70MPa, 620 ℃/90MPa and 600 ℃/110MPa respectively. 1080 groups of P91 steel pipes with different microstructure parameters are obtained. Then, the dislocation density and the size of the precipitated phase were measured by a transmission Electron microscope, and the crystal grain size was measured by EBSD (Electron back scattered Diffraction), thereby obtaining microstructure data of 1080 sets of sample samples.
And carrying out nonlinear ultrasonic detection on the 1080 samples to obtain 1080 groups of nonlinear coefficients and sound velocity.
And then, constructing a sample set according to the ultrasonic signals and the microscopic structure parameters, and dividing the sample set into a training set and a testing set.
And forming 1080 sample data by the microstructure data of the corresponding steel pipe, the corresponding nonlinear ultrasonic signal and the sound velocity, and then dividing a sample set formed by the 1080 sample data into a training set and a testing set.
Next, fig. 3 is a schematic structural diagram of a BP neural network used in the steel tube microstructure evaluation method based on nonlinear ultrasound according to an embodiment of the present invention, and the BP neural network shown in fig. 3 is constructed. As shown in the figure 3 of the drawings,
the input layer has 4 input nodes, the hidden layer is one layer, the hidden layer has 5 nodes, and the output layer has 3 output nodes. In the embodiment of the invention, the hidden layer node uses a sigmoid function as a transfer function of a neural network; the training precision of the neural network is 0.00005, and the maximum training frequency is 10000.
Then, training a pre-constructed BP (Back Propagation) neural network by using the training set until convergence; testing the converged BP neural network by using the test set, judging whether the accuracy of the converged BP neural network is greater than or equal to a preset threshold value, and if so, taking the converged BP neural network as a target BP neural network; if not, adjusting the weight and the hyper-parameter of the parameter in the BP neural network, and returning to execute the step of training the pre-constructed BP neural network by using the training set until convergence.
And then, identifying the ultrasonic signal obtained in the step S101 by using a target BP neural network, and further obtaining the microstructure parameters of the steel pipe to be evaluated. Typically, the microstructural parameters include: one or a combination of carbide size, grain size, and dislocation density.
In practical application, when the number of iterations of the pre-constructed BP neural network reaches the set maximum training number, or when the error between the microstructure of the steel pipe predicted by the BP neural network and the real microstructure of the steel pipe is smaller than a set value, the BP neural network can be judged to be converged.
By applying the embodiment shown in the figure 1 of the invention, the BP neural network is used for evaluating the steel pipe microstructure, compared with the defects that the prior art uses heavy equipment and has limited measuring position, the acquisition position of the ultrasonic signal is not limited, and the ultrasonic equipment has smaller volume and is lighter, so that the embodiment of the invention can evaluate the steel pipe microstructure more conveniently, and is convenient for steel pipe detection. And then still improved the detection efficiency of steel pipe.
In addition, the on-site coating metallographic phase is an effective method for observing the microstructure of the P91 steel, but the coating metallographic phase method has low magnification, less obtainable structural information and greatly reduced data reliability,
table 1 is a table comparing the evaluation results of the examples of the present invention with those of the prior art.
TABLE 1
| Technique of | Grain size | Size of precipitated phase | Dislocation density |
| Metallographic method | Coarse | Coarse | Is not applicable to |
| Hardness method | Is not applicable to | Is not applicable to | Is not applicable to |
| This patent | Accurate and accurate | Accurate and accurate | Accurate and accurate |
As shown in Table 1, compared with the metallographic method and the hardness method which are widely adopted in the power industry at present, the detection process of the technology is not interfered by human factors, and the repeatability and the detection precision are high.
The embodiment of the invention has the advantages of simple operation, short data acquisition time and strong repeatability, and meanwhile, the equipment can detect microstructures at multiple positions, thereby overcoming the defects of an improved indentation method.
Corresponding to the embodiment of the invention shown in fig. 1, the embodiment of the invention also provides a steel pipe microstructure evaluation device based on nonlinear ultrasound.
Fig. 4 is a schematic structural diagram of a steel pipe microstructure evaluation apparatus based on nonlinear ultrasound according to an embodiment of the present invention, and as shown in fig. 4, the apparatus includes:
the acquisition module 401 is configured to detect a steel pipe to be evaluated by using ultrasonic wave transmitting equipment, and obtain an ultrasonic signal and a sound velocity;
the identification module 402 is configured to identify the ultrasonic signal and the sound velocity of the steel pipe to be evaluated by using a pre-trained target BP neural network, and acquire a microstructure parameter of the steel pipe to be evaluated.
By applying the embodiment shown in the figure 1 of the invention, the BP neural network is used for evaluating the steel pipe microstructure, compared with the defects that the prior art uses heavy equipment and has limited measuring position, the acquisition position of the ultrasonic signal is not limited, and the ultrasonic equipment has smaller volume and is lighter, so that the embodiment of the invention can evaluate the steel pipe microstructure more conveniently, and is convenient for steel pipe detection.
In a specific implementation manner of the embodiment of the present invention, the obtaining module 401 is configured to:
detecting a steel pipe to be evaluated by using ultrasonic transmitting equipment, and collecting ultrasonic echoes;
and filtering the ultrasonic echo to obtain an ultrasonic signal.
In a specific implementation manner of the embodiment of the present invention, the training process of the target BP neural network includes:
detecting an ultrasonic signal of each pipeline sample by using ultrasonic transmitting equipment;
acquiring microstructure parameters of the pipeline sample;
constructing a sample set according to the ultrasonic signals and the microscopic tissue parameters, and dividing the sample set into a training set and a testing set;
training a pre-constructed BP neural network by using the training set until convergence; testing the converged BP neural network by using a test set, judging whether the accuracy of the converged BP neural network is greater than or equal to a preset threshold value, and if so, taking the converged BP neural network as a target BP neural network; if not, adjusting the weight and the hyper-parameters of the parameters in the BP neural network, and returning to execute the BP neural network which is constructed in advance by using the training set until convergence.
In a specific implementation of the embodiments of the present invention, the ultrasound signal includes: one or a combination of a first order nonlinear signal, a second order nonlinear signal, a third order nonlinear signal, a fourth order nonlinear signal, and ultrasonic sound speed.
In a specific implementation of the embodiment of the present invention, the microstructure parameters include: one or a combination of carbide size, grain size, and dislocation density.
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 and improvements made within the spirit and principle of the present invention are intended to be included within the scope of the present invention.
Claims (2)
1. A steel pipe microstructure evaluation method based on nonlinear ultrasound is characterized by comprising the following steps:
detecting a steel pipe to be evaluated by using ultrasonic transmitting equipment, and obtaining an ultrasonic signal and a sound velocity; the ultrasonic signal includes: a first order nonlinear signal, a second order nonlinear signal, a third order nonlinear signal, a fourth order nonlinear signal;
identifying an ultrasonic signal and a sound velocity of a steel pipe to be evaluated by using a pre-trained target BP neural network, and acquiring a microstructure parameter of the steel pipe to be evaluated;
the method for detecting the steel pipe to be evaluated by utilizing the ultrasonic transmitting equipment and obtaining the ultrasonic signal and the sound velocity comprises the following steps:
detecting a steel pipe to be evaluated by using ultrasonic transmitting equipment, and collecting ultrasonic echoes;
filtering the ultrasonic echo to obtain an ultrasonic signal;
the training process of the target BP neural network comprises the following steps:
for each pipeline sample, detecting an ultrasonic signal of the pipeline sample by using ultrasonic transmitting equipment;
acquiring microstructure parameters of the pipeline sample;
constructing a sample set according to the ultrasonic signals and the microscopic tissue parameters, and dividing the sample set into a training set and a testing set;
training a pre-constructed BP neural network by using the training set until convergence; testing the converged BP neural network by using a test set, judging whether the accuracy of the converged BP neural network is greater than or equal to a preset threshold value, and if so, taking the converged BP neural network as a target BP neural network; if not, adjusting the weight and the hyper-parameters of the parameters in the BP neural network, and returning to execute the BP neural network which is trained and pre-constructed by using the training set until convergence;
the input layer of the BP neural network is provided with 4 input nodes, the hidden layer is a layer 1, the hidden layer is provided with 5 nodes, and the output layer is provided with 3 output nodes; the hidden layer node uses a sigmoid function as a transfer function of the neural network;
the microstructural parameters include: one or a combination of carbide size, grain size, and dislocation density.
2. A steel pipe microscopic structure evaluation device based on nonlinear ultrasound is characterized by comprising:
the acquisition module is used for detecting the steel pipe to be evaluated by using ultrasonic transmitting equipment and acquiring an ultrasonic signal and a sound velocity; the ultrasound signal includes: a first order nonlinear signal, a second order nonlinear signal, a third order nonlinear signal, a fourth order nonlinear signal;
the identification module is used for identifying the ultrasonic signal and the sound velocity of the steel pipe to be evaluated by using a pre-trained target BP neural network to acquire the microstructure parameters of the steel pipe to be evaluated;
an acquisition module to:
detecting a steel pipe to be evaluated by using ultrasonic transmitting equipment, and collecting ultrasonic echoes;
filtering the ultrasonic echo to obtain an ultrasonic signal;
the training process of the target BP neural network comprises the following steps:
for each pipeline sample, detecting an ultrasonic signal of the pipeline sample by using ultrasonic transmitting equipment;
acquiring microstructure parameters of the pipeline sample;
constructing a sample set according to the ultrasonic signals and the microscopic tissue parameters, and dividing the sample set into a training set and a testing set;
training a pre-constructed BP neural network by using the training set until convergence; testing the converged BP neural network by using the test set, judging whether the accuracy of the converged BP neural network is greater than or equal to a preset threshold value, and if so, taking the converged BP neural network as a target BP neural network; if not, adjusting the weight and the hyper-parameters of the parameters in the BP neural network, and returning to execute the BP neural network which is trained and pre-constructed by using the training set until convergence;
the input layer of the BP neural network is provided with 4 input nodes, the hidden layer is a layer 1, the hidden layer is provided with 5 nodes, and the output layer is provided with 3 output nodes; the hidden layer node uses a sigmoid function as a transfer function of the neural network;
the microstructural parameters include: one or a combination of carbide size, grain size, and dislocation density.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910712674.7A CN110274960B (en) | 2019-08-02 | 2019-08-02 | Steel pipe microscopic structure evaluation method and device based on nonlinear ultrasound |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910712674.7A CN110274960B (en) | 2019-08-02 | 2019-08-02 | Steel pipe microscopic structure evaluation method and device based on nonlinear ultrasound |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110274960A CN110274960A (en) | 2019-09-24 |
| CN110274960B true CN110274960B (en) | 2022-07-26 |
Family
ID=67965920
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910712674.7A Active CN110274960B (en) | 2019-08-02 | 2019-08-02 | Steel pipe microscopic structure evaluation method and device based on nonlinear ultrasound |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110274960B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110702792B (en) * | 2019-09-29 | 2023-02-10 | 中国航发北京航空材料研究院 | Alloy tissue ultrasonic detection classification method based on deep learning |
| CN110865124B (en) * | 2019-11-27 | 2022-09-27 | 华东理工大学 | Nonlinear ultrasonic guided wave detection system and method based on linear power amplifier |
| CN111208196A (en) * | 2019-11-28 | 2020-05-29 | 郑州国电机械设计研究所有限公司 | Nonlinear ultrasonic detection method for early fatigue damage of high-strength bolt |
| CN111141821A (en) * | 2020-01-06 | 2020-05-12 | 大连理工大学 | No. 20 steel high-temperature tissue degradation nondestructive evaluation method based on critical refraction longitudinal wave technology |
| CN115791972B (en) * | 2023-01-18 | 2023-06-06 | 河北铁达科技有限公司 | Method, terminal and storage medium for evaluating health state of steel structure upright post |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101995436A (en) * | 2010-10-26 | 2011-03-30 | 江苏大学 | Neural network based drawing part crack monitoring method |
| 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 |
| CN104359977A (en) * | 2014-10-22 | 2015-02-18 | 北京理工大学 | Acoustic surface wave high-order nonlinear parameter representation method for bending fatigue state of metal plate |
| CN104636760A (en) * | 2015-03-11 | 2015-05-20 | 西安科技大学 | Positioning method for welding seam |
| CN105067703A (en) * | 2015-07-31 | 2015-11-18 | 华东理工大学 | Plate-like metal member/material remaining life non-destructive detection and evaluation method |
| CN105334269A (en) * | 2015-10-19 | 2016-02-17 | 江苏大学 | Pipeline defect type determination method based on neural network and guided wave characteristic database |
| CN108037190A (en) * | 2017-11-03 | 2018-05-15 | 大连理工大学 | A kind of high temperature furnace pipe damage rank intelligence assessment method based on neutral net |
| CN108490081A (en) * | 2018-04-03 | 2018-09-04 | 南昌航空大学 | A kind of determination method of alloy grain size based on multi-parameter and determining system |
| CN108896649A (en) * | 2018-04-28 | 2018-11-27 | 南京航空航天大学 | Method for estimating yield strength of ferromagnetic material based on electromagnetic ultrasound |
| CN108956787A (en) * | 2018-06-13 | 2018-12-07 | 西安理工大学 | A kind of rail failure detection method neural network based |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4244334B2 (en) * | 2004-08-27 | 2009-03-25 | 有限会社超音波材料診断研究所 | Ultrasonic material evaluation system |
| CN103323525A (en) * | 2013-05-21 | 2013-09-25 | 广东电网公司电力科学研究院 | Nonlinear ultrasound system used for detecting super austenitic stainless steel inter-granular corrosion |
-
2019
- 2019-08-02 CN CN201910712674.7A patent/CN110274960B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101995436A (en) * | 2010-10-26 | 2011-03-30 | 江苏大学 | Neural network based drawing part crack monitoring method |
| 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 |
| CN104359977A (en) * | 2014-10-22 | 2015-02-18 | 北京理工大学 | Acoustic surface wave high-order nonlinear parameter representation method for bending fatigue state of metal plate |
| CN104636760A (en) * | 2015-03-11 | 2015-05-20 | 西安科技大学 | Positioning method for welding seam |
| CN105067703A (en) * | 2015-07-31 | 2015-11-18 | 华东理工大学 | Plate-like metal member/material remaining life non-destructive detection and evaluation method |
| CN105334269A (en) * | 2015-10-19 | 2016-02-17 | 江苏大学 | Pipeline defect type determination method based on neural network and guided wave characteristic database |
| CN108037190A (en) * | 2017-11-03 | 2018-05-15 | 大连理工大学 | A kind of high temperature furnace pipe damage rank intelligence assessment method based on neutral net |
| CN108490081A (en) * | 2018-04-03 | 2018-09-04 | 南昌航空大学 | A kind of determination method of alloy grain size based on multi-parameter and determining system |
| CN108896649A (en) * | 2018-04-28 | 2018-11-27 | 南京航空航天大学 | Method for estimating yield strength of ferromagnetic material based on electromagnetic ultrasound |
| CN108956787A (en) * | 2018-06-13 | 2018-12-07 | 西安理工大学 | A kind of rail failure detection method neural network based |
Non-Patent Citations (1)
| Title |
|---|
| Effect of welding parameters on tensile strength of ultrasonic spot welded joints of aluminum to steel – By experimentation and artificial neural network;Dewang Zhao;《Journal of Manufacturing Processes》;20171231;第30卷;第63-74页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110274960A (en) | 2019-09-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110274960B (en) | Steel pipe microscopic structure evaluation method and device based on nonlinear ultrasound | |
| Konstantinidis et al. | The temperature stability of guided wave structural health monitoring systems | |
| CN102636303B (en) | Method for measuring residual stress of thin plating layer based on surface ultrasonic waves | |
| EP3483602B1 (en) | Classification of ultrasonic indications using pattern recognition | |
| CN108283011A (en) | The variation in metal structure is detected and is monitored using multimode acoustical signal | |
| Guan et al. | Quantitative ultrasonic testing for near-surface defects of large ring forgings using feature extraction and GA-SVM | |
| CN104034287B (en) | A kind of elastic anisotropy metallic matrix thermal barrier coating thickness ultrasonic measurement method | |
| Liu et al. | Guided waves based diagnostic imaging of circumferential cracks in small-diameter pipe | |
| Ma et al. | The reflection of guided waves from simple dents in pipes | |
| CN109142532A (en) | A kind of lossless detection method and device of the damage of high martensitic chromium heat resisting steel connector creep hole | |
| CN106018553A (en) | Device for evaluating and optimizing heat treatment technology based on nonlinear ultrasound | |
| CN112154324B (en) | Using multimode acoustic signals to detect, monitor and determine the location of changes in metal structures | |
| CN112771360A (en) | Signal processing | |
| CN110243521A (en) | A kind of sheet stress measurement method and sheet stress measuring system | |
| CN114330435A (en) | Composite material defect detection method based on dynamic weight wavelet coefficient deep residual network | |
| CN113484336B (en) | Subsurface crack size measuring method based on surface wave and BP neural network | |
| Sarris et al. | Ultrasonic methods for the detection of near surface fatigue damage | |
| CN105403621A (en) | Method for initiatively monitoring structural damages of metal pipe | |
| Wu et al. | Data-driven time–frequency analysis of nonlinear Lamb waves for characterization of grain size distribution | |
| CN103954628A (en) | Ensemble empirical mode decomposition (EEMD) and approximate entropy combined steel tube damage monitoring method | |
| Zhang et al. | Grain size characterization of Ti-6Al-4V titanium alloy based on laser ultrasonic random forest regression | |
| CN113189199B (en) | Nonlinear ultrasonic detection grading evaluation method for heat-resistant steel part performance degradation | |
| Khlybov et al. | Assessment of the degree of damage in structural materials using the parameters of structural acoustic noise | |
| US20120053895A1 (en) | Method and system for evaluating the condition of a collection of similar elongated hollow objects | |
| Hensman et al. | Detecting and identifying artificial acoustic emission signals in an industrial fatigue environment |
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 |