CN116124902B - Diagnostic method for ultrasonic guided wave damage positioning accuracy - Google Patents

Diagnostic method for ultrasonic guided wave damage positioning accuracy Download PDF

Info

Publication number
CN116124902B
CN116124902B CN202310054298.3A CN202310054298A CN116124902B CN 116124902 B CN116124902 B CN 116124902B CN 202310054298 A CN202310054298 A CN 202310054298A CN 116124902 B CN116124902 B CN 116124902B
Authority
CN
China
Prior art keywords
damage
signal
sigma
guided wave
sparse
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
Application number
CN202310054298.3A
Other languages
Chinese (zh)
Other versions
CN116124902A (en
Inventor
黄永
薛事成
周文松
李惠
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Harbin Institute of Technology
Original Assignee
Harbin Institute of Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Harbin Institute of Technology filed Critical Harbin Institute of Technology
Priority to CN202310054298.3A priority Critical patent/CN116124902B/en
Publication of CN116124902A publication Critical patent/CN116124902A/en
Application granted granted Critical
Publication of CN116124902B publication Critical patent/CN116124902B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • 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
    • G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
    • 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
    • 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
    • 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
    • G01N29/4409—Processing the detected response signal, e.g. electronic circuits specially adapted therefor by comparison
    • G01N29/4418—Processing the detected response signal, e.g. electronic circuits specially adapted therefor by comparison with a model, e.g. best-fit, regression analysis
    • 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
    • G01N29/449—Statistical methods not provided for in G01N29/4409, e.g. averaging, smoothing and interpolation
    • 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
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E30/00—Energy generation of nuclear origin
    • Y02E30/30—Nuclear fission reactors

Landscapes

  • Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Immunology (AREA)
  • Pathology (AREA)
  • Biochemistry (AREA)
  • General Health & Medical Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Data Mining & Analysis (AREA)
  • Signal Processing (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Probability & Statistics with Applications (AREA)
  • Mathematical Analysis (AREA)
  • Computational Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Evolutionary Biology (AREA)
  • General Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Databases & Information Systems (AREA)
  • Algebra (AREA)
  • Operations Research (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Acoustics & Sound (AREA)
  • Investigating Or Analyzing Materials By The Use Of Ultrasonic Waves (AREA)
  • Measurement Of Velocity Or Position Using Acoustic Or Ultrasonic Waves (AREA)

Abstract

The application provides a diagnosis method for ultrasonic guided wave damage positioning accuracy based on Bayes uncertainty quantification. The method comprises the steps of preprocessing a measurement signal, constructing and solving a guided wave damage positioning sparse problem, judging algorithm damage positioning accuracy and the like. According to the method, the damage positioning problem based on ultrasonic guided waves is solved by adopting a multi-task complex-level sparse Bayesian model, and whether the damage positioning result is accurate enough is diagnosed by utilizing the change trend of uncertainty indexes given by the multi-task complex-level sparse Bayesian model under different sensor data numbers. In addition, the multi-task complex-level sparse Bayesian model can estimate the super-parameters in the model according to the data, so that uncertainty quantization indexes given in the Bayesian model in the method are more objective, and whether a damage positioning structure is accurate enough or not can be diagnosed more reliably.

Description

Diagnostic method for ultrasonic guided wave damage positioning accuracy
Technical Field
The application belongs to the technical field of machine learning and civil engineering, and particularly relates to a diagnosis method for ultrasonic guided wave damage positioning accuracy based on Bayes uncertainty quantification.
Background
In the current age when civil engineering construction is becoming saturated, detection and maintenance of built civil structures are receiving more and more attention. Ultrasonic nondestructive testing has found wide application in the detection of structural damage in civil engineering. By means of ultrasonic nondestructive testing technology, whether defects or damage exist in the structure or not can be effectively detected, defect positioning is carried out, damage degree is estimated, and accordingly safety conditions of the built structure can be evaluated and corresponding maintenance measures can be timely implemented. Compared with other ultrasonic detection methods, the nondestructive detection technology based on the ultrasonic guided wave method has the advantages of large detection range and high efficiency, and can greatly reduce the labor cost and human errors. Therefore, development and utilization of a suitable guided wave detection technology to solve the detection problem in actual civil engineering are currently urgent research works to be carried out.
Because the guided wave has the characteristics of dispersion, multi-mode, mode conversion and the like, an ultrasonic guided wave signal received by a sensor generally becomes very complex, so that effective interpretation of the signal becomes very difficult, and the effectiveness of the guided wave detection technology in civil engineering structure detection is severely limited. The method based on signal sparse representation can carry out sparse decomposition on the guided wave signal under a certain overcomplete dictionary matrix, so that effective information in the guided wave signal is obtained. In recent years, the damage positioning of a structure based on a signal sparse representation method has become a research hot spot in the aspect of ultrasonic guided wave signal processing. However, due to the complexity of the guided wave signal, the signal sparse representation method sometimes gives an incorrect damage positioning result, and how to judge the accuracy of the damage result is a problem worthy of research.
The Bayesian probability theory method is used as an important means for processing uncertainty, fully utilizes data measurement information and prior probability information, deduces posterior probability distribution of unknown parameters, and therefore quantifies uncertainty of damage recognition results, namely confidence (reliability) of damage recognition deduced results. In the application, a multi-task complex-level sparse Bayesian learning method is used for carrying out sparse representation on the damage signals, and damage position information is obtained according to the construction of a dictionary matrix. Meanwhile, the uncertainty index of the damage position is calculated through the parameters of posterior probability distribution, so that the confidence (reliability) of damage positioning is obtained. And diagnosing the accuracy of the damage positioning result according to the uncertainty index under the condition of different sensor data volumes.
Disclosure of Invention
The application aims to solve the problems in the prior art, and provides a diagnosis method for ultrasonic guided wave damage positioning accuracy based on Bayes uncertainty quantification. The method is suitable for diagnosing whether the damage positioning result obtained by calculation according to the existing detection data is accurate.
The application is realized through the following technical scheme, and provides a diagnosis method for ultrasonic guided wave damage positioning accuracy based on Bayes uncertainty quantification, which comprises the following steps:
preprocessing a measurement signal, decomposing the signal by adopting an orthogonal matching pursuit method considering the guided wave dispersion effect, normalizing the amplitude of the signal, obtaining the envelope of the signal through Hilbert transform, and obtaining the signal after preprocessing
Step two, meshing the area needing damage detection, and constructing a damage positioning sparse problem under the condition of simultaneously considering a plurality of frequency information based on the physical characteristics of guided wave propagationSolving the sparse problem by adopting a multi-task complex layered sparse Bayesian model, so as to obtain damage point positions and posterior distribution parameter values thereof;
step three, judging the distance between the damage points obtained by the algorithm identification; if the distance between the identified damage points is smaller than the diagonal length of the grid, deleting the node with small amplitude of the non-0 item in the corresponding weight vector v from the grid, and returning to the second step for re-calculation until the grid is not changed;
step four, starting from the sensor data number of 2, repeating the calculation of the step two and the step three, and then adding new measured sensor data one by one; a posterior covariance matrix sigma calculated according to the step two k Calculating uncertainty index sigma avg,k Finally, sigma under different sensor data quantity is obtained avg,k The method comprises the steps of carrying out a first treatment on the surface of the If sigma avg,k The method has the advantages that when a certain data quantity is suddenly reduced and then slowly changes along with the increase of the data quantity, the result of diagnosing damage positioning is accurate and reliable; otherwise, the positioning accuracy of the diagnosis damage is to be improved.
Further, the first step specifically comprises:
step 1.1, the signal measured from the structure to be measured is differenced with the reference signal of the structure to be measured to obtain the damage signal y of the sensor i i (t)；
Step 1.2, decomposing the damage signal by adopting orthogonal matching pursuit to obtain a decomposed signalNormalize the decomposed signal amplitude and record as +.>
Step 1.3, by calculation
Obtaining a signalEnvelope signal +.> wherein />Is a hilbert variation;
step 1.4, by calculation
Obtaining a frequency domain signal after pretreatment
Further, the second step specifically comprises:
step 2.1, constructing a damage positioning sparse problem under the condition of simultaneously considering a plurality of frequency information according to the physical rule of propagation of guided waves in the structure;
step 2.2, according to the sparse representation problem constructed in the step 2.1, performing random embedding to convert into a multi-task complex layered sparse Bayesian learning problem;
step 2.3, solving the sparse problem in the step 2.2 by adopting multi-task complex hierarchical sparse Bayesian learning, and firstly, carrying out the following transformation on the expression of the sparse problem:
wherein
The following formula will be given
Performing cyclic calculation; mu is calculated first k Sum sigma k Re-using the calculated mu k Sum sigma k Calculation ofAnd->Calculated +.>And->For calculating mu k Sum sigma k The iteration is repeated in a circulating way until the numerical value of each variable between the previous iteration and the subsequent iteration is not changed obviously; finally, solving to obtain w k Average value μ of posterior distribution k Sum covariance matrix sigma k 。
Further, the representation of the damage localization sparsity problem is specifically:
wherein ,corresponding to frequency f k Dictionary matrix of lower guided wave mode gamma with elements of row i and column j>Is that wherein k′(γ) (f k ) For preprocessing the signal at the mode gamma frequency f k Wavenumber, d i,j The distance between the ith sensor and the jth position point to be measured is the distance between the ith sensor and the jth position point to be measured; />Is a dictionary matrix under all modes, wherein N is as follows (γ) The total number of guided wave modes in the structure; weight vector +.>Each element corresponds to one position of the region to be measured, if some element is 0, the corresponding position is not damaged, and if some element is not 0, the corresponding position is damaged.
Further, in step 2.2, there is an uncertain prediction measurement signalRepresented as
wherein ,for the uncertain weight vector, v (f k ) As a true weight vector, the unknown prediction error e k From modeling error p k And measuring error n k Composition of->
Further, in the third step, for the result obtained by the lesion localization, the distance between any two lesion points i, j and the distance matrix L between the lesion points may be calculated t The method comprises the following steps:
wherein ,li,j The distance from the ith damage point to the jth damage point is the distance; find all distance values l greater than the diagonal length e of the grid i,j Comparing the values of the points i and j corresponding to the magnitudes of the elements in the upsilon, deleting the points with small magnitudes from the divided grid nodes, returning to the step for recalculation until the grid division is not changed.
Further, the fourth step specifically comprises:
step 4.1, starting from using two sensor data, repeating the second and third steps to calculate, and then adding new sensor data one by one; calculating according to a formula to obtain a frequency f k An uncertainty index for each sensor number;
step 4.2, when the number of the sensor data is increased, each frequency f k The change condition of the uncertainty index is used for diagnosing whether the damage positioning accuracy is enough; if each frequency f k The uncertainty index sigma below avg,k The value of (2) is suddenly reduced when a certain sensor number is generated, and after that, the value slowly changes along with the increase of the sensor data, so that the result of diagnosing damage positioning is accurate and reliable, otherwise, the result of diagnosing damage positioning is inaccurate.
Further, the calculation formula in step 4.1 is:
wherein ,σi,k Is a posterior covariance matrix sigma k The i-th element on the main diagonal, m is the order of the covariance matrix.
The application provides electronic equipment, which comprises a memory and a processor, wherein the memory stores a computer program, and the processor realizes the steps of the ultrasonic guided wave damage positioning accuracy diagnosis method based on Bayesian uncertainty quantification when executing the computer program.
The application provides a computer readable storage medium for storing computer instructions which when executed by a processor realize the steps of the diagnostic method for ultrasonic guided wave damage positioning accuracy based on Bayesian uncertainty quantization.
The application has the beneficial effects that:
1. the application provides a multi-task model for positioning guided wave damage, which can comprehensively utilize information under a plurality of frequencies to improve the accuracy of damage positioning.
2. The application utilizes a multi-task complex layering sparse Bayesian algorithm to effectively quantify posterior uncertainty of the damage positioning result.
3. The application provides a diagnosis method for accurately positioning damage results, and solves the problem that the traditional method does not know whether the damage positioning results are correct.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings that are required to be used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only embodiments of the present application, and that other drawings can be obtained according to the provided drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a method for diagnosing ultrasonic guided wave damage positioning accuracy based on Bayes uncertainty quantization.
FIG. 2 is a schematic diagram of the connection of the ultrasonic guided wave nondestructive testing system according to the present application.
Fig. 3 is a schematic diagram of the arrangement of the sensors and the actual damage location.
Fig. 4 is a schematic diagram of a variation trend of uncertainty index with increasing sensor information in an embodiment of the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
The application aims to provide a diagnosis method for ultrasonic guided wave damage positioning accuracy based on Bayesian uncertainty quantification in order to judge whether a damage positioning result obtained according to existing detection data is accurate. First, the measurement data is preprocessed. Then, an uncertainty index when two sensor data are solved by using a multi-task complex-level sparse Bayesian algorithm. Then, the sensor data volume is increased one by one, and an uncertainty index under each sensor data volume is obtained. Finally, diagnosing whether the damage positioning is accurate enough according to the change trend of the uncertainty index along with the data quantity of the sensor.
Referring to fig. 1-4, the application provides a diagnosis method for ultrasonic guided wave damage positioning accuracy based on bayesian uncertainty quantification, which comprises the following steps:
preprocessing a measurement signal, decomposing the signal by adopting an orthogonal matching pursuit method considering the guided wave dispersion effect, normalizing the amplitude of the signal, obtaining the envelope of the signal through Hilbert transform, and obtaining the signal after preprocessing
Step two, meshing the area needing damage detection, and constructing a damage positioning sparse problem under the condition of simultaneously considering a plurality of frequency information based on the physical characteristics of guided wave propagationSolving the sparse problem by adopting a multi-task complex layered sparse Bayesian model, so as to obtain damage point positions and posterior distribution parameter values thereof;
step three, in order to prevent the same damage from being identified as adjacent grid nodes by the algorithm due to too small grid spacing, the number of the damage is incorrect, and the spacing between the damage points identified by the algorithm is judged; if the distance between the identified damage points is smaller than the diagonal length of the grid, deleting the node with small amplitude of the non-0 item in the corresponding weight vector v from the grid, and returning to the second step for re-calculation until the grid is not changed;
step four, starting from the sensor data number of 2, repeating the calculation of the step two and the step three, and then adding new measured sensor data one by one; a posterior covariance matrix sigma calculated according to the step two k Calculating uncertainty index sigma avg,k Finally, sigma under different sensor data quantity is obtained avg,k The method comprises the steps of carrying out a first treatment on the surface of the If sigma avg,k A steep drop occurs at a certain data amount and then slowly changes with the increase of the data amount, so that the damage is diagnosedThe positioning result is accurate and reliable; otherwise, the positioning accuracy of the diagnosis damage is to be improved.
The first step is specifically as follows:
step 1.1, the signal measured from the structure to be measured is differenced with the reference signal of the structure to be measured to obtain the damage signal y of the sensor i i (t)；
Step 1.2, decomposing the damage signal by adopting orthogonal matching pursuit to obtain a decomposed signalNormalize the decomposed signal amplitude and record as +.>
Step 1.3, by calculation
Obtaining a signalEnvelope signal +.> wherein />Is a hilbert variation;
step 1.4, by calculation
Obtaining a frequency domain signal after pretreatment
The second step is specifically as follows:
step 2.1, constructing a damage positioning sparse problem under the condition of simultaneously considering a plurality of frequency information according to the physical rule of propagation of guided waves in the structure;
step 2.2, performing random embedding (stochastic embedding) according to the sparse representation problem constructed in the step 2.1 to convert the sparse representation problem into a multi-task complex hierarchical sparse Bayesian learning problem;
step 2.3, solving the sparse problem in the step 2.2 by adopting multi-task complex hierarchical sparse Bayesian learning, and firstly, carrying out the following transformation on the expression of the sparse problem:
wherein
The following formula will be given
Performing cyclic calculation; mu is calculated first k Sum sigma k Re-using the calculated mu k Sum sigma k Calculation ofAnd->Calculated +.>And->For calculating mu k Sum sigma k The iteration is repeated in a circulating way until the numerical value of each variable between the previous iteration and the subsequent iteration is not changed obviously; finally, solving to obtain w k Average value μ of posterior distribution k Sum covariance matrix sigma k 。
The damage positioning sparseness problem is specifically expressed as follows:
wherein ,corresponding to frequency f k Dictionary matrix of lower guided wave mode gamma with elements of row i and column j>Is that wherein k′(γ) (f k ) For preprocessing the signal at the mode gamma frequency f k Wavenumber, d i,j The distance between the ith sensor and the jth position point to be measured is the distance between the ith sensor and the jth position point to be measured; />Is a dictionary matrix under all modes, wherein N is as follows (γ) The total number of guided wave modes in the structure; weight vector +.>Each element corresponds to one position of the region to be measured, if some element is 0, the corresponding position is not damaged, and if some element is not 0, the corresponding position is damaged.
In step 2.2, there is an uncertain prediction measurement signalRepresented as
wherein ,for the uncertain weight vector, v (f k ) As a true weight vector, the unknown prediction error e k From modeling error p k And measuring error n k Composition of->
In the third step, for the result obtained by the damage positioning, the distance between any two damage points i, j and the distance matrix L between the damage points can be calculated t The method comprises the following steps:
wherein ,li,j The distance from the ith damage point to the jth damage point is the distance; find all distance values l greater than the diagonal length e of the grid i,j Comparing the values of the points i and j corresponding to the magnitudes of the elements in the upsilon, deleting the points with small magnitudes from the divided grid nodes, returning to the step for recalculation until the grid division is not changed.
The fourth step is specifically as follows:
step 4.1, starting from using two sensor data, repeating the second and third steps to calculate, and then adding new sensor data one by one; calculating according to a formula to obtain a frequency f k An uncertainty index for each sensor number;
step 4.2, when the number of the sensor data is increased, each frequency f k The change condition of the uncertainty index is used for diagnosing whether the damage positioning accuracy is enough; if each frequency f k The uncertainty index sigma below avg,k The value of (2) is suddenly reduced when a certain sensor number is generated, and after that, the value slowly changes along with the increase of the sensor data, so that the result of diagnosing damage positioning is accurate and reliable, otherwise, the result of diagnosing damage positioning is inaccurate.
The calculation formula in step 4.1 is:
wherein ,σi,k Is a posterior covariance matrix sigma k The i-th element on the main diagonal, m is the order of the covariance matrix.
Examples
This embodiment enables the application of the present application to the defect detection of an aluminum plate member, fig. 1 shows a flowchart of the method of the present application, and fig. 2 shows a schematic diagram of the connection of the apparatus, ultrasonic transducer and test piece used for the implementation of this example. Fig. 3 shows a schematic diagram of the arrangement of the sensors and the actual damage location.
The excitation signal is a sine signal of five peaks of a narrow band of a Hanning window. And respectively exciting the plate structure under the nondestructive and damage states to obtain guided wave signals under the nondestructive and damage conditions, and performing subtraction on the guided wave signals to obtain residual signals related to damage. And respectively calculating the two working conditions of the test and the simulation, wherein in the working condition of the test, the center frequency of an excitation signal is 60kHz, a piezoelectric sheet symmetrical excitation plate structure is used, the excitation mode is an S0 mode, the plate material is aluminum, and the thickness is 1mm.
The first step is specifically that: in the embodiment, 6 piezoelectric sensors are arranged on a plate structure, and each sensor is preprocessed one by one to obtain a preprocessed signal
The second step is specifically as follows: meshing the region to be detected, and dividing signalsAt f k Injury localization sparsity problem =complex amplitude substitution construct at 2khz,3khz,4khz,5khz,6khz +.> And (5) performing calculation. The first two sets of data are taken for calculation. Calculating a transformed measurement vector g k . According to the grid division, calculating a complex dictionary matrix +.>And calculate the transformed matrix phi k . Respectively iterative calculation of mu k ,∑ k And->Covariance matrix sigma of posterior distribution finally obtained through iteration k Calculating uncertainty index sigma avg,k 。
The third step is specifically as follows: obtaining positions of damage points through non-0 item positions in mu obtained in the second step, and calculating a distance matrix L between the damage points t . Find all the damage point distances l greater than the diagonal length of the grid i,j Comparing the values of the points i and j corresponding to the magnitudes of the elements in the upsilon, deleting the point with small magnitude from the divided grid nodes, repeating the operation in the second step until the grid division is not changed any more, and then performing sigma avg,k As a final result.
The fourth step isThe method comprises the following steps: new sensor data are added continuously in sequence, and sigma under different sensor numbers is calculated avg,k . As the number of sensors increases one by one, if σ avg,k The steep drop and the slow change after that indicate that the result of the lesion localization is accurate. Otherwise, the damage positioning accuracy is still to be improved. In an embodiment, σ avg,k As shown in fig. 4, the number of sensors is sufficient, and the uncertainty index σ at each frequency increases with the increase of the number of sensors avg,k And after abrupt drop, the damage positioning result is stable and reliable.
The application provides electronic equipment, which comprises a memory and a processor, wherein the memory stores a computer program, and the processor realizes the steps of the ultrasonic guided wave damage positioning accuracy diagnosis method based on Bayesian uncertainty quantification when executing the computer program.
The application provides a computer readable storage medium for storing computer instructions which when executed by a processor realize the steps of the diagnostic method for ultrasonic guided wave damage positioning accuracy based on Bayesian uncertainty quantization.
The memory in embodiments of the present application may be either volatile memory or nonvolatile memory, or may include both volatile and nonvolatile memory. The nonvolatile memory may be a Read Only Memory (ROM), a Programmable ROM (PROM), an Erasable PROM (EPROM), an electrically Erasable EPROM (EEPROM), or a flash memory. The volatile memory may be random access memory (random access memory, RAM) which acts as an external cache. By way of example, and not limitation, many forms of RAM are available, such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous DRAM (SLDRAM), and direct memory bus RAM (DR RAM). It should be noted that the memory of the methods described herein is intended to comprise, without being limited to, these and any other suitable types of memory.
In the above embodiments, it may be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer instructions are loaded and executed on a computer, the processes or functions described in accordance with embodiments of the present application are produced in whole or in part. The computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by a wired (e.g., coaxial cable, fiber optic, digital subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that contains an integration of one or more available media. The usable medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a magnetic tape), an optical medium (e.g., a high-density digital video disc (digital video disc, DVD)), or a semiconductor medium (e.g., a Solid State Disk (SSD)), or the like.
In implementation, the steps of the above method may be performed by integrated logic circuits of hardware in a processor or by instructions in the form of software. The steps of a method disclosed in connection with the embodiments of the present application may be embodied directly in a hardware processor for execution, or in a combination of hardware and software modules in the processor for execution. The software modules may be located in a random access memory, flash memory, read only memory, programmable read only memory, or electrically erasable programmable memory, registers, etc. as well known in the art. The storage medium is located in a memory, and the processor reads the information in the memory and, in combination with its hardware, performs the steps of the above method. To avoid repetition, a detailed description is not provided herein.
It should be noted that the processor in the embodiments of the present application may be an integrated circuit chip with signal processing capability. In implementation, the steps of the above method embodiments may be implemented by integrated logic circuits of hardware in a processor or instructions in software form. The processor may be a general purpose processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic, or discrete hardware components. The disclosed methods, steps, and logic blocks in the embodiments of the present application may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of the method disclosed in connection with the embodiments of the present application may be embodied directly in the execution of a hardware decoding processor, or in the execution of a combination of hardware and software modules in a decoding processor. The software modules may be located in a random access memory, flash memory, read only memory, programmable read only memory, or electrically erasable programmable memory, registers, etc. as well known in the art. The storage medium is located in a memory, and the processor reads the information in the memory and, in combination with its hardware, performs the steps of the above method.
The application provides a diagnosis method for ultrasonic guided wave damage positioning accuracy based on Bayesian uncertainty quantification, which is described in detail above, and specific examples are applied to illustrate the principle and implementation of the application, and the description of the above examples is only used for helping to understand the method and core ideas of the application; meanwhile, as those skilled in the art will have variations in the specific embodiments and application scope in accordance with the ideas of the present application, the present description should not be construed as limiting the present application in view of the above.

Claims (10)

1. A diagnosis method for ultrasonic guided wave damage positioning accuracy based on Bayes uncertainty quantification is characterized by comprising the following steps:
preprocessing a measurement signal, decomposing the signal by adopting an orthogonal matching pursuit method considering the guided wave dispersion effect, normalizing the amplitude of the signal, obtaining the envelope of the signal through Hilbert transform, and obtaining the signal after preprocessing
Step two, meshing the area needing damage detection, and constructing a damage positioning sparse problem under the condition of simultaneously considering a plurality of frequency information based on the physical characteristics of guided wave propagationSolving the sparse problem by adopting a multi-task complex layered sparse Bayesian model, so as to obtain damage point positions and posterior distribution parameter values thereof;
step three, judging the distance between the damage points obtained by the algorithm identification; if the distance between the identified damage points is smaller than the diagonal length of the grid, deleting the node with small amplitude of the non-0 item in the corresponding weight vector v from the grid, and returning to the second step for re-calculation until the grid is not changed;
step four, starting from the sensor data number of 2, repeating the calculation of the step two and the step three, and then adding new measured sensor data one by one; a posterior covariance matrix sigma calculated according to the step two k Calculating uncertainty index sigma avg,k Finally, sigma under different sensor data quantity is obtained avg,k The method comprises the steps of carrying out a first treatment on the surface of the If sigma avg,k The method has the advantages that when a certain data quantity is suddenly reduced and then slowly changes along with the increase of the data quantity, the result of diagnosing damage positioning is accurate and reliable; otherwise the first set of parameters is selected,the positioning accuracy of the diagnostic damage is to be improved.
2. The method according to claim 1, wherein the first step is specifically:
step 1.1, the signal measured from the structure to be measured is differenced with the reference signal of the structure to be measured to obtain the damage signal y of the sensor i i (t)；
Step 1.2, decomposing the damage signal by adopting orthogonal matching pursuit to obtain a decomposed signalNormalize the decomposed signal amplitude and record as +.>
Step 1.3, by calculation
Obtaining a signalEnvelope signal +.> wherein />Is a hilbert variation;
step 1.4, by calculation
Obtaining a frequency domain signal after pretreatment
3. The method according to claim 2, wherein the second step is specifically:
step 2.1, constructing a damage positioning sparse problem under the condition of simultaneously considering a plurality of frequency information according to the physical rule of propagation of guided waves in the structure;
step 2.2, according to the sparse representation problem constructed in the step 2.1, performing random embedding to convert into a multi-task complex layered sparse Bayesian learning problem;
step 2.3, solving the sparse problem in the step 2.2 by adopting multi-task complex hierarchical sparse Bayesian learning, and firstly, carrying out the following transformation on the expression of the sparse problem:
wherein
The following formula will be given
Performing cyclic calculation; mu is calculated first k Sum sigma k Re-using the calculated mu k Sum sigma k Calculation ofAnd->Calculated +.>And->For calculating mu k Sum sigma k The iteration is repeated in a circulating way until the numerical value of each variable between the previous iteration and the subsequent iteration is not changed obviously; finally, solving to obtain w k Average value μ of posterior distribution k Sum covariance matrix sigma k 。
4. A method according to claim 3, characterized in that the representation of the impairment localization sparsity problem is in particular:
wherein ,corresponding to frequency f k Dictionary matrix of lower guided wave mode gamma with elements of row i and column j>Is that wherein k′(γ) (f k ) For preprocessing the signal at the mode gamma frequency f k Wavenumber, d i,j The distance between the ith sensor and the jth position point to be measured is the distance between the ith sensor and the jth position point to be measured; />Is a dictionary matrix under all modes, wherein N is as follows (γ) The total number of guided wave modes in the structure; weight vector +.>Each element corresponds to one position of the region to be measured, if some element is 0, the corresponding position is not damaged, and if some element is not 0, the corresponding position is damaged.
5. The method according to claim 4, characterized in that in step 2.2 there is an uncertain predicted measurement signalRepresented as
wherein ,for the uncertain weight vector, v (f k ) As a true weight vector, the unknown prediction error e k From modeling error p k And measuring error n k Composition of->
6. The method according to claim 5, wherein in the third step, for the result of the lesion localization, a distance between any two lesion points i, j and a distance matrix L between the lesion points can be calculated t The method comprises the following steps:
wherein ,li,j The distance from the ith damage point to the jth damage point is the distance; find all distance values l greater than the diagonal length e of the grid i,j Comparing the values of the points i and j corresponding to the magnitudes of the elements in the upsilon, deleting the points with small magnitudes from the divided grid nodes, returning to the step for recalculation until the grid division is not changed.
7. The method according to claim 6, wherein the fourth step is specifically:
step 4.1, starting from using two sensor data, repeating the second and third steps to calculate, and then adding new sensor data one by one; calculating according to a formula to obtain a frequency f k An uncertainty index for each sensor number;
step 4.2, when the number of the sensor data is increased, each frequency f k The change condition of the uncertainty index is used for diagnosing whether the damage positioning accuracy is enough; if each frequency f k The uncertainty index sigma below avg,k The numerical value of the sensor is suddenly reduced when the number of the sensors is increased, and then the numerical value is slowly changed along with the increase of the number of the sensor data, so that the result of diagnosing damage positioning is accurate and reliable, otherwise, the accuracy of diagnosing damage positioning is to be improved.
8. The method of claim 7, wherein the calculation formula in step 4.1 is:
wherein ,σi,k Is a posterior covariance matrix sigma k The i-th element on the main diagonal, m is the order of the covariance matrix.
9. An electronic device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1-8 when the computer program is executed.
10. A computer readable storage medium storing computer instructions which, when executed by a processor, implement the steps of the method of any one of claims 1-8.
CN202310054298.3A 2023-02-03 2023-02-03 Diagnostic method for ultrasonic guided wave damage positioning accuracy Active CN116124902B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310054298.3A CN116124902B (en) 2023-02-03 2023-02-03 Diagnostic method for ultrasonic guided wave damage positioning accuracy

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310054298.3A CN116124902B (en) 2023-02-03 2023-02-03 Diagnostic method for ultrasonic guided wave damage positioning accuracy

Publications (2)

Publication Number Publication Date
CN116124902A CN116124902A (en) 2023-05-16
CN116124902B true CN116124902B (en) 2023-08-18

Family

ID=86309703

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310054298.3A Active CN116124902B (en) 2023-02-03 2023-02-03 Diagnostic method for ultrasonic guided wave damage positioning accuracy

Country Status (1)

Country Link
CN (1) CN116124902B (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109781864A (en) * 2019-01-29 2019-05-21 哈尔滨工业大学 Signal de-noising and defect inspection method and flaw indication reconstructing method based on guided wave signals sparse decomposition method
CN109871824A (en) * 2019-03-11 2019-06-11 西安交通大学 The multi-modal separation method of supersonic guide-wave and its system based on management loading
CN110542723A (en) * 2019-09-17 2019-12-06 哈尔滨工业大学 guided wave signal sparse decomposition and damage positioning-based two-stage damage position identification method
CN110907540A (en) * 2019-12-04 2020-03-24 哈尔滨工业大学 Ultrasonic guided wave multi-damage identification method based on Bayesian updating and Gibbs sampling

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8924315B2 (en) * 2011-12-13 2014-12-30 Xerox Corporation Multi-task learning using bayesian model with enforced sparsity and leveraging of task correlations
WO2013172876A1 (en) * 2012-05-16 2013-11-21 Hidden Solutions, Llc Method and system for multi-path active defect detection, localization, and characterization with ultrasonic guided waves
FR3084748B1 (en) * 2018-08-01 2024-01-05 Commissariat Energie Atomique RAIL HEALTH CONTROL

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109781864A (en) * 2019-01-29 2019-05-21 哈尔滨工业大学 Signal de-noising and defect inspection method and flaw indication reconstructing method based on guided wave signals sparse decomposition method
CN109871824A (en) * 2019-03-11 2019-06-11 西安交通大学 The multi-modal separation method of supersonic guide-wave and its system based on management loading
CN110542723A (en) * 2019-09-17 2019-12-06 哈尔滨工业大学 guided wave signal sparse decomposition and damage positioning-based two-stage damage position identification method
CN110907540A (en) * 2019-12-04 2020-03-24 哈尔滨工业大学 Ultrasonic guided wave multi-damage identification method based on Bayesian updating and Gibbs sampling

Also Published As

Publication number Publication date
CN116124902A (en) 2023-05-16

Similar Documents

Publication Publication Date Title
Robbins et al. Mean shift testing in correlated data
Chen et al. An innovative deep neural network–based approach for internal cavity detection of timber columns using percussion sound
JP2019070965A (en) Learning device, learning method, and program
CN107037136B (en) Array type ultrasonic detection method for heat exchange dirt of pipeline
Yu et al. Application of Kernel Density Estimation in Lamb Wave‐Based Damage Detection
Wang et al. Towards high-precision data modeling of SHM measurements using an improved sparse Bayesian learning scheme with strong generalization ability
Tian et al. Joint learning model for underwater acoustic target recognition
Li et al. An Orthogonal Matching Pursuit based signal compression and reconstruction approach for electromechanical admittance based structural health monitoring
US20200393347A1 (en) Imaging Method of Internal Defects in Longitudinal Sections of Trees
CN111291481B (en) Bayesian model-based structure early warning analysis method
Entezami Structural health monitoring by time series analysis and statistical distance measures
Song et al. An improved reconstruction method based on auto-adjustable step size sparsity adaptive matching pursuit and adaptive modular dictionary update for acoustic emission signals of rails
Cao et al. Seismic damage identification of high arch dams based on an unsupervised deep learning approach
Ma et al. Classification of damage types in liquid-filled buried pipes based on deep learning
CN116124902B (en) Diagnostic method for ultrasonic guided wave damage positioning accuracy
CN116307269B (en) Photovoltaic power generation power prediction method and device based on artificial intelligence
Li et al. All-phase fast Fourier transform and multiple cross-correlation analysis based on Geiger iteration for acoustic emission sources localization in complex metallic structures
Moon et al. Impact parameter prediction of a simulated metallic loose part using convolutional neural network
Yuan et al. Quantitative damage evaluation of curved plates based on phased array guided wave and deep learning algorithm
CN114841892A (en) Sparse guided wave data recovery method based on full-connection network
Chencho et al. Structural damage quantification using ensemble‐based extremely randomised trees and impulse response functions
Zhang et al. Physics-informed deep learning for structural vibration identification and its application on a benchmark structure
Xi et al. Classification of multiple power quality disturbances based on continuous wavelet transform and lightweight convolutional neural network
Luo et al. Ultrasonic lamb wave damage detection of cfrp composites using the bayesian neural network
CN117494573B (en) Wind speed prediction method and system and electronic equipment

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