CN114324584A - Steel structure detection method based on intelligent algorithm and ultrasonic phased array technology - Google Patents
Steel structure detection method based on intelligent algorithm and ultrasonic phased array technology Download PDFInfo
- Publication number
- CN114324584A CN114324584A CN202111546254.XA CN202111546254A CN114324584A CN 114324584 A CN114324584 A CN 114324584A CN 202111546254 A CN202111546254 A CN 202111546254A CN 114324584 A CN114324584 A CN 114324584A
- Authority
- CN
- China
- Prior art keywords
- steel structure
- damage
- intelligent algorithm
- phased array
- information
- 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
Landscapes
- Investigating Or Analyzing Materials By The Use Of Ultrasonic Waves (AREA)
Abstract
The invention discloses a steel structure detection method based on an intelligent algorithm and an ultrasonic phased array technology, which comprises the following steps: acquiring initial information of a steel structure; grouping the steel structural members according to the member types; establishing a steel structure analysis model by using the initial information and the construction grouping condition of the steel structure as a modeling basis and by means of structural analysis software; structural analysis software simulates damage of components at different parts of a steel structure, and further obtains damaged modal information; analyzing, inducing and sorting modal data after different parts are damaged to form a training sample; establishing a damage positioning model by adopting an intelligent algorithm; carrying out region detection on the movement of the ultrasonic focusing point; positioning the defect position; the field control computer processes data and transmits the data back to the PC terminal computer, and the functions of visual display and statistics are realized. The invention has the beneficial effects that: the damage position is accurately positioned in the steel structure detection, and the damage degree detection efficiency is high.
Description
Technical Field
The invention relates to the technical field of steel structure detection, in particular to a steel structure detection method based on an intelligent algorithm and an ultrasonic phased array technology.
Background
Compared with other structures, the steel structure has the advantages of light weight, good toughness, convenience in construction, environmental friendliness and the like, so that the steel structure is widely applied to various building structures. But with the increase of service time, the steel structure can be damaged in different degrees and different positions. Degradation of performance occurs once a steel structure in a building structure is damaged and left to develop, and even failure collapse results in an immeasurable loss. Therefore, damage detection and health condition assessment for steel structures become hot issues of intense interest at home and abroad.
At present, the steel structure is generally detected by a non-destructive detection method, such as ultrasonic detection, eddy current detection, magnetic memory detection, acoustic vibration detection, and the like. The methods can well evaluate the condition of the damaged part of the steel structure. However, the methods generally have the defects of complicated operation mode, low detection efficiency due to the fact that the detection angle cannot be arranged according to needs, high requirements on operators and the like. The ultrasonic phased array technology is different from the above technologies, can flexibly control the focusing position of an acoustic beam, improves the detection signal-to-noise ratio, the coverage rate and the sensitivity, positions and images the inside of an object to be detected on the premise of not moving or moving a transducer a little, has a relatively large detection range, and can realize high-efficiency detection in a small range.
In practical engineering, the steel structure is often huge in quantity, undamaged members are often in most parts, and the defects of large workload, low efficiency, low economy and the like are generally caused when the existing local detection method is used for detecting the members of the steel structure one by one. The development of intelligent algorithms in recent years has brought a hope of solving this problem. The intelligent algorithm has the advantages of parallelism, self-organization, self-learning, associative memory function, strong robustness, fault tolerance and the like of processing information, and is widely applied to various fields of civil engineering and the like. The structure damage positioning method applying the intelligent algorithm does not need prior knowledge of the structure dynamic characteristics, and has the advantage of non-parameter damage positioning; the method can obtain the implicit relation between input and output hidden in sample data through training and learning, can filter noise and extract the inherent characteristics of objects under the noisy condition, and is more suitable for carrying out damage positioning on a structure with a large amount of noise and measurement errors. However, due to the limitation of the intelligent algorithm, namely, the damaged part cannot be accurately positioned to a point when a positioning model is designed, and the damaged part can only be positioned to a certain area through modal parameter information.
Disclosure of Invention
Aiming at the two problems of low detection efficiency of the prior detection technology and difficulty in damage positioning in the current detection, the invention provides the steel structure detection method based on the intelligent algorithm and the ultrasonic phased array technology, which has accurate positioning and high detection efficiency.
The invention relates to a steel structure detection method based on an intelligent algorithm and an ultrasonic phased array technology, which is characterized by comprising the following steps of:
acquiring initial information of a steel structure;
grouping the steel structural members according to the member types;
establishing a steel structure analysis model by using the initial information and the construction grouping condition of the steel structure as a modeling basis and by means of a structural analysis program or structural analysis software;
simulating the damage of the components at different parts of the steel structure by a structural analysis program, and further obtaining the damaged modal information;
analyzing, inducing and sorting modal data after different parts are damaged to form a training sample;
adopting an intelligent algorithm to take the damage conditions of different components of the steel structure as input parameters and corresponding modal parameters as output parameters, and combining the different input parameters and the output parameters to form a training sample; training and learning the model constructed by the intelligent algorithm by using the combined training sample, continuously performing iterative computation if the learning result does not meet the precision requirement, and establishing a damage positioning model based on the intelligent algorithm until the model learning and training result meets the specified precision requirement; the intelligent algorithm comprises a group intelligent algorithm, a simulated annealing algorithm, a Bayesian method, a Gaussian process, a neural network, various agent models and the like;
according to the positioning information given by the intelligent algorithm damage positioning system, a PC terminal computer belonging to the ultrasonic phased array system realizes the movement of an ultrasonic focusing point for area detection by changing the setting of a multi-channel instrument;
the method comprises the following steps that a tester holds an array probe to transmit a plurality of groups of signals to a region to be tested, and a receiving end receives echo signals of the region to be tested and intercepts a plurality of groups of scattered echo signals; inverting the time sequence of each group of received scattering signals, selecting a time point with the maximum amplitude in each group, and positioning the position of a defect;
the field control computer processes the acquired data and inputs the processed data into a PC terminal computer of a control room in a long-distance transmission communication mode, thereby realizing the functions of visual display and statistics.
Further, the steel structure initial information comprises construction drawing information of the bridge, steel structure line information, steel structure material information and current state steel structure modal information.
Further, in order to improve the accuracy and the convergence speed of the intelligent algorithm for judging the damage position of the steel structure, the whole steel structure is dispersed into different sub-components according to the component type, and the sub-components can be divided into beam components, rod components, supporting components and the like according to the stress condition of the steel structure.
Furthermore, when the construction analysis software is used for modeling, a steel structure model under a non-damage condition is established according to the grouping condition of each component of the design information, the models are grouped according to the type of the sub-component, and the damage condition of each sub-component is digitized and used for later-stage statistical information to form a training sample.
Furthermore, in the structural damage part simulation process, all the possibly damaged sub-components are subjected to damage simulation one by one, and damage mode information in a single-component damage state is obtained, which is the most important and complex task in the whole detection method.
The damage introduction may be by reducing the modulus of the damage site, providing a cell gap, or even removing a portion of the cell. Specifically, damage is introduced into an original non-damage model, the part where the damage is introduced is determined according to the grouping condition of the subcomponents, each subcomponent is ensured to be introduced with damage at least once, the damage introducing mode can adopt the mode of reducing the modulus of the damaged part, setting a unit notch, even removing part of units, and a structural analysis program is utilized to output modal parameters of different parts of a steel structure in a single damage state.
Furthermore, the single damage state refers to that only a single component damage working condition needs to be simulated in the structural analysis model each time, and the intelligent algorithm randomly combines each single component damage working condition and the corresponding modal sample to obtain multi-damage working condition corresponding modal information of simultaneous damage of multiple components.
Further, the training sample has universality for current detection and subsequent detection of the steel structure to be detected, and the data is still applicable no matter how the performance of the steel structure is degraded in the subsequent service period, and belongs to a permanent structure data model.
Further, after the structural analysis model is established, the steel structure damage position needs to be inversely calculated according to the actually measured steel structure modal information, the damage positioning precision of the intelligent algorithm is judged, if the precision requirement is met, the establishment of the intelligent damage positioning model is completed, otherwise, the intelligent algorithm needs to be iteratively optimized again until the precision requirement is met; the intelligent algorithm adopts the normalized damage signal index NDSI as a network input vector to carry out specific damage positioning, the normalized damage signal index NDSI is used as the damage identification of the network input vector, and the error between the positioning damage position and the actual damage position of the intelligent algorithm is smaller than the maximum scanning range of the ultrasonic phased array technology, namely the accuracy requirement is met.
Furthermore, the multi-channel instrument is divided into an array transmitting part and an array receiving part, is controlled by a PC terminal computer, and realizes scanning of the area to be detected by controlling the position of the phased array probe.
Furthermore, the phased array probe is composed of a focused acoustic beam, the focused acoustic beam is composed of a certain number of array elements, and the PC terminal computer controls the phased array probe to scan the area to be detected by moving the array elements.
Further, the field control computer processes the acquired data as follows: the array receiving circuit converts the reflected ultrasonic signals into voltage which can be measured, data stored in a memory of the field control computer are finally obtained through data processing, and the bus is converted into an RS232 bus and an RS485 bus, so that long-distance transmission is realized and then the data are converted into RS 232. In the operation of computer software, the RS232 bus is only required to be programmed to realize bidirectional data communication.
The invention has the beneficial effects that: by fusing an intelligent algorithm positioning prediction technology and an ultrasonic phased array detection technology, the detection efficiency of the steel structure is greatly improved; the intelligent algorithm positioning system can realize the initial positioning of the damage position according to the actual modal parameters of the steel structure by using the learning and iteration functions of the algorithm and only through less manpower and material resource consumption; after the ultrasonic phased array system is preliminarily positioned, the PC terminal computer controls the phased array probe to accurately scan and detect the area to be detected, the scanning range is expanded by changing the position of the array element, the advantage of accurate detection in a small range of the ultrasonic phased array technology is fully exerted, the problems that the position of a damage is difficult to position and the detection efficiency is low in the traditional detection method are solved, and the steel structure detection technology is further improved.
Drawings
FIG. 1a is a flow chart of the present invention;
FIG. 1b is a schematic view of the detection method of the present invention;
FIG. 2 is a flow chart of the intelligent algorithm of the present invention;
FIG. 3 is a flow chart of an ultrasonic phased array technique of the present invention;
FIG. 4 is a schematic diagram of the operation of the ultrasonic phased array technique;
fig. 5 is a diagram of the detection effect of the ultrasonic phased array technology.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, 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 some, but not all, embodiments of the present invention. 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.
Fig. 1a and 1b are structural diagrams of a steel structure detection method based on an intelligent algorithm and an ultrasonic phased array technology, which include an intelligent algorithm system as shown in fig. 2 and an ultrasonic phased array technology system as shown in fig. three, and specifically include the following steps
S1, acquiring initial information of the steel structure;
s2, grouping the steel structural members according to the member types;
s3, establishing a steel structure analysis model by using the steel structure initial information and the construction grouping condition as a modeling basis and by means of a structure analysis program;
s4, simulating the damage of the components at different parts of the steel structure by the structural analysis program, and further obtaining the damaged modal information;
s5, analyzing, inducing and sorting modal data after different parts are damaged to form training samples;
the S6 intelligent algorithm takes the damage conditions of different components of the steel structure as input parameters, takes corresponding modal parameters as output parameters, and combines different input parameters and output parameters to form a training sample; training and learning the model constructed by the intelligent algorithm by using the combined training sample, continuously performing iterative computation if the learning result does not meet the precision requirement, and establishing a damage positioning model based on the intelligent algorithm until the model learning and training result meets the specified precision requirement; the intelligent algorithm comprises a group intelligent algorithm, a simulated annealing algorithm, a Bayesian method, a Gaussian process, a neural network, various agent models and the like;
s7, according to the positioning information given by the intelligent algorithm damage positioning system, the PC terminal belonging to the ultrasonic phased array system realizes the movement of the ultrasonic focusing point for area detection by changing the setting of the multi-channel instrument;
s8, a detector holds the array probe to transmit a plurality of groups of signals to a region to be detected, and a receiving end receives echo signals of the region to be detected and intercepts a plurality of groups of scattered echo signals; and reversing the time sequence of each group of received scattering signals, selecting a time point with the maximum amplitude in each group, and positioning the defect position.
S9 the site control computer processes partial data and inputs the processed data to the PC terminal computer of the control room by long distance transmission communication mode to realize visual display and statistic function.
In the step S1, in the initial preparation stage, steel structure construction drawing information, steel structure line information, and steel structure material information need to be acquired from a steel structure operator, and meanwhile, a corresponding dynamic characteristic test needs to be organized to obtain the current modal parameters of the steel structure.
In the step S2, in order to improve the accuracy and convergence speed of the intelligent algorithm for judging the damage position of the steel structure, the whole steel structure is dispersed into different sub-components according to the component type, and the sub-components can be subdivided into beam components, rod components, supporting components and the like according to the stress condition of the steel structure.
In step S3, when modeling with the structural analysis program, the steel structure models under the non-damage condition are first established according to the grouping condition of each component of the design information, and the models are grouped according to the type of the sub-component.
In the step S4, a damage is introduced into the original non-damaged model, the location where the damage is introduced is determined according to the sub-component grouping condition, it is ensured that each sub-component introduces at least one damage, the damage introduction mode can adopt reducing the modulus of the damaged location, setting a unit gap, even removing a part of units, and using a structural analysis program to output the modal parameters of different locations of the steel structure in a single damage state.
The step of constructing the digital damage information refers to representing each damage information in a matrix form as follows: beam member damage is reported as 100, rod member damage is reported as 010 support member damage as 001, and so on. The structural sub-members include, but are not limited to, beam members, rod members, and support members, which should be classified according to actual steel structural properties.
In the step S5, the data sample has universality for current detection and subsequent detection of the steel structure to be detected, and the data is still applicable no matter how the performance of the steel structure is degraded in the subsequent service period, and belongs to a permanent structure data model.
And S6, combining, learning and iteratively establishing a damage positioning model by training samples formed by the damage conditions of different components of the steel structure and corresponding modal parameters, calculating the damage position of the steel structure according to the actual measured modal information of the steel structure after the model is established, judging the damage positioning accuracy of the intelligent algorithm, completing the establishment of the intelligent damage positioning model if the accuracy requirement is met, and otherwise, iteratively optimizing the intelligent algorithm again until the accuracy requirement is met. The method is characterized in that the error between the position of the damage positioned by the intelligent algorithm and the actual position of the damage is smaller than the maximum scanning range of the ultrasonic phased array technology, and the accuracy requirement is met.
In the step S7, the design of the multi-channel instrument can be divided into two parts of array emission and array reception, a terminal computer initiates a command, a data input stage is started, the scanning of the sub-component region to be detected is realized by controlling the position of the phased array probe, the data input is stopped after the scanning of all the sub-component detection regions is finished, and a data transfer stage is started. And after the data of the control computer is analyzed and stored by the PC terminal, the next sub-component is checked until all the sub-components are detected.
The phased array probe is composed of focused acoustic beams, the focused acoustic beams are composed of a certain number of array elements, and the PC terminal computer controls the phased array probe to scan the area to be detected by moving the array elements.
The number of the array elements of the focused acoustic beam and the phased array probe is determined by the actual detection range, for example, a focused acoustic beam consisting of 16 array elements is adopted, and the effective detection range is 16 mm.
In the step S9, the array receiving circuit converts the reflected ultrasonic signal into a voltage that can be measured, and finally obtains the data stored in the memory of the field control computer by data processing, and converts the bus into an RS232 bus into an RS485 bus, so as to realize long-distance transmission and then converts the data into an RS 232. In the operation of computer software, the RS232 bus is only required to be programmed to realize bidirectional data communication.
The embodiments described in this specification are merely illustrative of implementations of the inventive concept and the scope of the present invention should not be considered limited to the specific forms set forth in the embodiments but includes equivalent technical means as would be recognized by those skilled in the art based on the inventive concept.
Claims (10)
1. A steel structure detection method based on an intelligent algorithm and an ultrasonic phased array technology is characterized by comprising the following steps:
acquiring initial information of a steel structure;
grouping the steel structural members according to the member types;
establishing a steel structure analysis model by using the steel structure initial information and the construction grouping condition as a modeling basis and by means of a structure analysis program;
simulating the damage of the components at different parts of the steel structure by a structural analysis program, and further obtaining the damaged modal information;
analyzing, inducing and sorting modal data after different parts are damaged to form a training sample;
adopting an intelligent algorithm to take the damage conditions of different components of the steel structure as input parameters and corresponding modal parameters as output parameters, and combining the different input parameters and the output parameters to form a training sample; training and learning the model constructed by the intelligent algorithm by using the combined training sample, continuously performing iterative computation if the learning result does not meet the precision requirement, and establishing a damage positioning model based on the intelligent algorithm until the model learning and training result meets the specified precision requirement;
according to the positioning information given by the intelligent algorithm damage positioning system, a PC terminal computer belonging to the ultrasonic phased array system realizes the movement of an ultrasonic focusing point for area detection by changing the setting of a multi-channel instrument;
the method comprises the following steps that a tester holds an array probe to transmit a plurality of groups of signals to a region to be tested, and a receiving end receives echo signals of the region to be tested and intercepts a plurality of groups of scattered echo signals; inverting the time sequence of each group of received scattering signals, selecting a time point with the maximum amplitude in each group, and positioning the position of a defect;
the field control computer processes the acquired data and inputs the processed data into a PC terminal computer of a control room in a long-distance transmission communication mode, thereby realizing the functions of visual display and statistics.
2. The steel structure detection method based on the intelligent algorithm and the ultrasonic phased array technology as claimed in claim 1, characterized in that: the steel structure initial information comprises construction drawing information of the bridge, steel structure line information, steel structure material information and current state steel structure modal information.
3. The steel structure detection method based on the intelligent algorithm and the ultrasonic phased array technology as claimed in claim 1, characterized in that: in order to improve the accuracy and the convergence speed of the intelligent algorithm for judging the damage position of the steel structure, the steel structure is integrally dispersed into different sub-components according to the component type, and the sub-components can be subdivided into beam components, rod components and supporting components according to the stress condition of the steel structure.
4. The steel structure detection method based on the intelligent algorithm and the ultrasonic phased array technology as claimed in claim 1, characterized in that: when the construction analysis software is used for modeling, a steel structure model under a non-damage condition is established according to the grouping condition of each component of the design information, the models are grouped according to the type of the subcomponents, and the damage condition of each subcomponent is digitized and used for later-stage statistical information to form a training sample.
5. The steel structure detection method based on the intelligent algorithm and the ultrasonic phased array technology as claimed in claim 1, characterized in that: the method comprises the steps of introducing damage into an original non-damage model, determining the part where the damage is introduced according to the grouping condition of the subcomponents, ensuring that each subcomponent is at least subjected to one-time damage introduction, reducing the modulus of the damaged part, setting a unit notch, even removing part of units by introducing the damage, and outputting modal parameters of different parts of a steel structure in a single-damage state by using structural analysis software.
6. The steel structure detection method based on the intelligent algorithm and the ultrasonic phased array technology according to claim 5, characterized in that: the single damage state refers to that only a single component damage working condition needs to be simulated in the structural analysis model each time, and the intelligent algorithm randomly combines each single component damage working condition and the corresponding modal sample to obtain multi-damage working condition corresponding modal information of simultaneous damage of multiple components.
7. The steel structure detection method based on the intelligent algorithm and the ultrasonic phased array technology as claimed in claim 1, characterized in that: after the structure analysis model is established, the steel structure damage position needs to be back calculated according to the actually measured steel structure modal information, the damage positioning precision of the intelligent algorithm is judged, if the precision requirement is met, the establishment of the intelligent damage positioning model is completed, otherwise, the intelligent algorithm needs to be iteratively optimized again until the precision requirement is met; the intelligent algorithm adopts the normalized damage signal index NDSI as a network input vector to carry out specific damage positioning, the normalized damage signal index NDSI is used as the damage identification of the network input vector, and the error between the positioning damage position and the actual damage position of the intelligent algorithm is smaller than the maximum scanning range of the ultrasonic phased array technology, namely the accuracy requirement is met.
8. The steel structure detection method based on the intelligent algorithm and the ultrasonic phased array technology as claimed in claim 1, characterized in that: the multi-channel instrument is divided into an array transmitting part and an array receiving part, is controlled by a PC terminal computer, and realizes scanning of the area to be detected by controlling the position of the phased array probe.
9. The steel structure detection method based on the intelligent algorithm and the ultrasonic phased array technology according to claim 8, characterized in that: the phased array probe is composed of focused acoustic beams, the focused acoustic beams are composed of a certain number of array elements, and the PC terminal computer controls the phased array probe to scan the area to be detected by moving the array elements.
10. The steel structure detection method based on the intelligent algorithm and the ultrasonic phased array technology as claimed in claim 1, characterized in that: the field control computer processes the acquired data as follows: the array receiving circuit converts the reflected ultrasonic signals into voltage which can be measured, data stored in a memory of the field control computer are finally obtained through data processing, and the bus is converted into an RS232 bus and an RS485 bus, so that long-distance transmission is realized and then the data are converted into RS 232.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111546254.XA CN114324584B (en) | 2021-12-16 | 2021-12-16 | Steel structure detection method based on intelligent algorithm and ultrasonic phased array technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111546254.XA CN114324584B (en) | 2021-12-16 | 2021-12-16 | Steel structure detection method based on intelligent algorithm and ultrasonic phased array technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114324584A true CN114324584A (en) | 2022-04-12 |
| CN114324584B CN114324584B (en) | 2023-09-05 |
Family
ID=81052008
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111546254.XA Active CN114324584B (en) | 2021-12-16 | 2021-12-16 | Steel structure detection method based on intelligent algorithm and ultrasonic phased array technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114324584B (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101566601A (en) * | 2009-05-04 | 2009-10-28 | 北京航空航天大学 | System for recognizing tensile damage state of 16Mn steel force-bearing part by adopting neural network and coupling iteration |
| CN102279223A (en) * | 2011-05-16 | 2011-12-14 | 北京航空航天大学 | Identifying and assessing system for deformation and damage of 16Mn steel welded structure on basis of SNF (Strongest Neighbor Filter) and DSD (Deformation Sensitive Decimation) policies |
| CN204612669U (en) * | 2015-05-20 | 2015-09-02 | 中国海洋石油总公司 | A kind of crude oil pipeline wax deposition thickness off-line detection device |
| CN111272886A (en) * | 2020-03-24 | 2020-06-12 | 华东理工大学 | Pressure vessel damage positioning method based on convolutional neural network |
| CN111561901A (en) * | 2020-04-15 | 2020-08-21 | 交通运输部水运科学研究所 | Steel structure strain detection fault identification method based on deep learning |
| CN111767897A (en) * | 2020-07-14 | 2020-10-13 | 上海应用技术大学 | Support vector machine-based steel rail crack defect identification method |
| CN112131781A (en) * | 2020-08-26 | 2020-12-25 | 浙江工业大学 | Steel structure damage detection method based on full-connection neural network and transfer rate function |
| KR102256181B1 (en) * | 2019-12-30 | 2021-05-27 | 한국과학기술원 | Method of inspecting and evaluating coating state of steel structure and system for the same |
| CN114813973A (en) * | 2022-03-09 | 2022-07-29 | 浙江大学 | Bolt loosening detection method combining time reversal technology and BP neural network |
| CN115754022A (en) * | 2022-11-29 | 2023-03-07 | 东南大学 | Steel truss welding node damage degree identification method based on acoustic emission |
-
2021
- 2021-12-16 CN CN202111546254.XA patent/CN114324584B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101566601A (en) * | 2009-05-04 | 2009-10-28 | 北京航空航天大学 | System for recognizing tensile damage state of 16Mn steel force-bearing part by adopting neural network and coupling iteration |
| CN102279223A (en) * | 2011-05-16 | 2011-12-14 | 北京航空航天大学 | Identifying and assessing system for deformation and damage of 16Mn steel welded structure on basis of SNF (Strongest Neighbor Filter) and DSD (Deformation Sensitive Decimation) policies |
| CN204612669U (en) * | 2015-05-20 | 2015-09-02 | 中国海洋石油总公司 | A kind of crude oil pipeline wax deposition thickness off-line detection device |
| KR102256181B1 (en) * | 2019-12-30 | 2021-05-27 | 한국과학기술원 | Method of inspecting and evaluating coating state of steel structure and system for the same |
| CN111272886A (en) * | 2020-03-24 | 2020-06-12 | 华东理工大学 | Pressure vessel damage positioning method based on convolutional neural network |
| CN111561901A (en) * | 2020-04-15 | 2020-08-21 | 交通运输部水运科学研究所 | Steel structure strain detection fault identification method based on deep learning |
| CN111767897A (en) * | 2020-07-14 | 2020-10-13 | 上海应用技术大学 | Support vector machine-based steel rail crack defect identification method |
| CN112131781A (en) * | 2020-08-26 | 2020-12-25 | 浙江工业大学 | Steel structure damage detection method based on full-connection neural network and transfer rate function |
| CN114813973A (en) * | 2022-03-09 | 2022-07-29 | 浙江大学 | Bolt loosening detection method combining time reversal technology and BP neural network |
| CN115754022A (en) * | 2022-11-29 | 2023-03-07 | 东南大学 | Steel truss welding node damage degree identification method based on acoustic emission |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114324584B (en) | 2023-09-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11255825B2 (en) | Wrinkle characterization and performance prediction for composite structures | |
| EA021646B1 (en) | Non-destructive testing, in particular for tubes during manufacturing and in the finished state | |
| CN111754463B (en) | Method for detecting CA mortar layer defects of ballastless track based on convolutional neural network | |
| Pyle et al. | Uncertainty quantification for deep learning in ultrasonic crack characterization | |
| CN112731377A (en) | Dielectric constant inversion method, roadbed detection method and detection system | |
| JP5156707B2 (en) | Ultrasonic inspection method and apparatus | |
| CN108918679A (en) | Based on elastic wave and the prefabricated post sleeve of machine learning grouting lossless detection method | |
| CN115371651A (en) | House building quality evaluation method and system | |
| US20240119199A1 (en) | Method and system for generating time-efficient synthetic non-destructive testing data | |
| CN111678991B (en) | Method for nondestructive testing damage identification of concrete structure | |
| CN111579647B (en) | Concrete member corrosion degree detection method and system based on analytic hierarchy process | |
| CN114324584B (en) | Steel structure detection method based on intelligent algorithm and ultrasonic phased array technology | |
| CN112816557B (en) | Defect detection method, device, equipment and storage medium | |
| US11494873B2 (en) | Automated TFM grid resolution setup tools | |
| CN111189906B (en) | On-line intelligent judging and classifying identification method for alternating current magnetic field defects | |
| CN111241725B (en) | Structure response reconstruction method for generating countermeasure network based on conditions | |
| CN108152373B (en) | System and method for measuring mechanical property of composite material containing pores based on pore size identification | |
| CN113092588A (en) | Ultrasonic phased array detection device and detection method | |
| Wall et al. | Worth of modelling for assessing the intrinsic capability of NDT | |
| CN111210409A (en) | Condition-based generation confrontation network structure damage identification method | |
| Le Berre et al. | Simulation and processing tools for the design and performance evaluation of FMC-TFM techniques | |
| CN117949542B (en) | Industrial pipe gallery column-holding scaffold firmness detection method, medium and system | |
| CN116482114A (en) | Method for identifying damage of steel structural beam after fire disaster | |
| CN117686588B (en) | Nondestructive monitoring system and method for welding quality of gas pipeline | |
| Dekhterev et al. | Diagnostics of defects of monolithic reinforced concrete structures using neural network technologies |
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 |