Crack Location Identification Research Articles

Application of Machine Learning (ML)-based multi-classifications to identify corrosion fatigue cracking phenomena in Naval steel weldments

Corrosion fatigue (CF) crack detection in welded ship hull structures is quite crucial as well as challenging for ensuring their structural integrity. This is conventionally done by offline ship hull survey and inspection methods, which have large time and cost implications. Data-driven approaches of ship hull crack detection are promising in this regard. Corrosion fatigue crack growth rate (CFCGR) behaviour of shipbuilding steel was studied for its base metal (BM), fatigue-prone heat affected zones (HAZs) of Submerged Arc Welding (SAW), and shielded metal arc welding (SMAW) joints of Naval-grade steel. The results indicated the highest CFCGR for BM, intermediate for SAW, and smallest for SMAW welds. This was attributed to 44% lath martensite observed in the SMAW HAZ microstructure as compared to 33% in SAW HAZ. A novel ML-based method has been proposed for CF crack location identification by a multi-classification approach using CFCGR data for structural health monitoring (SHM) applications. Prediction of crack location (either BM, SMAW HAZ, or SAW HAZ) was investigated using K-nearest neighbour (KNN), linear support vector machine classifier (LSVC), Naïve Bayes (NB), decision tree (DT), random forest (RF), and artificial neural network (ANN) models. DT and RF models were found to perform very well even without scaling and hyper-parameter tuning operations and exhibited 98–99% classification accuracy. Scaling and tuning operations resulted in significant improvements in accuracy of KNN models from 43% to 84% (minmax scaling) and 98% (standard scaling). Significant improvement in the accuracy from 37% to 98% ANN model was achieved by standard scaling. The proposed method shall be useful in data-driven real-time crack location identification in ship hull structures.

Identification of crack location in metallic biomaterial cantilever beam subjected to moving load base on central difference approximation

Abstract If not detected early, the cracks in structural components may ultimately result in the failure of the structure. This issue becomes even more critical when the component under investigation is a prosthesis placed in the human body. This study presents a crack location identification method based on the time domain in a cantilever beam of metallic biomaterials (CBMB). The absolute difference between the central difference approximation of the root mean square (RMS) of displacement of points on the cracked and uncracked beams was applied as a cracked location indicator. Captured time-domain data (displacement) at each node of the cracked and uncracked beams were processed into a central difference approximation of the RMS of displacement. Then, the crack could be detected by a sudden change of the cracked location indicator. The feasibility and effectiveness of the proposed method were validated by numerical simulations. The finite-element simulation of a CBMB with a transverse notch was analyzed in the numerical study. The notch or crack was detected along the beam under a moving load at various locations. A set of simulation experiments and numerical calculations was performed to determine whether the proposed identification method would accurately detect the location of a crack in a cantilever beam under a moving load compared to the location found by an exact solution method. The results showed that the proposed method was not only as able as the analytical method but also robust against noise: it was able to detect a crack precisely under 5% noise.

Modal displacement vs Curvature functions as damage identifier for masonry structures

Identification of crack location and severity is a vital step in assessment of deteriorated structures. The current research addresses the sensitivity of using modal displacement, modal curvature, and modal strain energy functions as Damage Indices (D.I.s) for crack detection. Comparison of these indicators at cracked and undamaged conditions was performed. Firstly, Damage Index (D.I.) method was utilized on synthetic data obtained with the aid of finite element modelling on different flexural elements. The technique was again used to identify cracks for physical scaled-down masonry walls. Three clayey bricks walls with different thicknesses and crack configurations were prepared for this purpose. Each wall was tested in its virgin and damaged states. Formal modal analysis was conducted. It was concluded that all the used D.I.s were accepted as damage indicators. It can also be stated that the relative severity between cracks can be obtained. Numerically, D.I. method was found effective in cases of multiple or inclined cracks. Generally, the identification based on curvature functions, were better than displacement functions for the same number of monitoring points and crack severity. Curvature can’t, however, be easily obtained for sophisticated structures with limited data. Moreover, the detection process became easier as the crack dimensions increased.

Fault detection in rotor system by discrete wavelet neural network algorithm

This study identifies a method for detection of irregularities such as open cracks or grooves on a rotating stepped shaft with multiple discs, based on the wavelet transforms. Cracks are represented as reduction in diameter of shaft (groove) with small width. Single as well as multiple grooves are considered on stepped shaft at locations of stress concentration. Translational or rotational response curves/mode shapes are extracted from finite element analysis of rotors with and without grooves. Discrete and continuous 1D wavelet transforms are applied on resultant response curve or mode shapes. The results show that rotational response curves or mode shapes are more sensitive to shaft cracks and key contributors to identify the location of cracks than translation response curves or mode shapes. Discrete wavelet transforms are accurate enough to locate the groove of smaller size. Effectiveness of detection by wavelets transforms is analysed for single as well as multiple grooves with increase in groove depth. Increase in groove depth can be quantified by increase in wavelet coefficient, and it can be an indicator. White Gaussian noise with low signal-to-noise ratio is added to response curves and analysed for crack location identification. Intelligent techniques such as artificial neural networks are used to quantify the location and depth of crack. Discrete wavelet transforms coefficients are provided as input to the neural network. Feed forward artificial neural networks are trained with Levenberg–Marquardt back propagation algorithm. Trained networks are able to quantify the crack location and depth accurately.

Analysis and simulations for a phase‐field fracture model at finite strains based on modified invariants

AbstractPhase‐field models have already been proven to predict complex fracture patterns for brittle fracture at small strains. In this paper we discuss a model for phase‐field fracture at finite deformations in more detail. Among the identification of crack location and projection of crack growth the numerical stability is one of the main challenges in solid mechanics. Here we present a phase‐field model at finite strains, which takes into account the anisotropy of damage by applying an anisotropic split of the modified invariants of the right Cauchy‐Green strain tensor. We introduce a suitable weak notion of solution that also allows for a spatial and temporal discretization of the model. In this framework we study the existence of solutions and we show that the time‐discrete solutions converge in a weak sense to a solution of the time‐continuous formulation of the model. Numerical examples in two and three space dimensions illustrate the range of validity of the analytical results.

Inverse method for identification of edge crack using correlation model

Crack in a structural member alters local stiffness that affects the dynamic characteristics such as natural frequency and mode shape of the structure. Identification of crack location and size is one key destination of structural health monitoring. In this paper, we propose a correlation-based inverse method for identification of crack location and size in a cantilever beam. First, natural frequencies of healthy and cracked cantilever beams were computed using finite element method and verified with experimental tests. Next, changes in natural frequencies with respect to crack locations and sizes were analyzed and the correlation model between natural frequency and crack parameters was constructed by non-linear curve-fitting. Finally, in the inverse analysis, correlation model was used to identify crack location and size based on the knowledge of the first and second natural frequencies from the measurements. Results indicate that the inverse method with new correlation model can identify crack location and size. The proposed identification method demonstrated for simple cantilever beam can be adapted for crack identification in complex structures. The proposed method is simple and easy to identify crack location and depth; its main advantage is that knowledge of only the first and second natural frequencies are required to predict crack parameters.

Crack Depth and Crack Location Identification using Artificial Neural Network

Crack Depth and Crack Location Identification using Artificial Neural Network

Inverse method to identify crack parameters in multi-span beam using genetic algorithm

The increased flexibility due to the presence of an open crack in the multi-span beam is made bases for identification of crack location and size. The genetic algorithm is coupled with the inverse dynamics of the cracked beam to get effective solution. The modal parameters, obtained through non-destructive vibration testing, are the input to the code and the location and size of crack are the output. The crack is modeled as rotational spring, and support conditions are generalized in the formulation for cracked multi-span beam.

Identification of Crack Location in Beam Structures Using Wavelet Transform and Fractal Dimension

Identification of structural crack location has become an intensely investigated subject due to its practical importance. In this paper, a hybrid method is presented to detect crack locations using wavelet transform and fractal dimension (FD) for beam structures. Wavelet transform is employed to decompose the mode shape of the cracked beam. In many cases, small crack location cannot be identified from approximation signal and detailed signals. And FD estimation method is applied to calculate FD parameters of detailed signals. The crack locations will be detected accurately by FD singularity of the detailed signals. The effectiveness of the proposed method is validated by numerical simulations and experimental investigations for a cantilever beam. The results indicate that the proposed method is feasible and can been extended to more complex structures.

Non-Destructive testing based method for crack detection in beams

Beams are the constituent elements of several machine parts and sophisticated structures. In this paper, efforts are made to develop suitable methods that can serve as the basis to detect crack location and to crack size from measured axial vibration data. This method is used to address the inverse problem of assessing the crack location and crack size in various beam structure. The method is based on measurement of axial natural frequencies, which are global parameter and can be easily measured from any point on the structure. In theoretical analysis, the relationship between the natural frequencies, crack location, and crack size has been developed. For identification of crack location and crack size, it was shown that data on the variation of the first two natural frequencies is sufficient. The experimental analysis is done to verify the practical applicability of the theoretical method developed.

Three-Dimensional Identification of Crack Location in Conducting Slabs Using Wavelets

Different nondestructive crack inspection approaches have been previously proposed for a wide range of industrial applications. Among those approaches, field measurement resulting from impressed current crack perturbation has been suggested. In this paper, semi-orthogonal compactly supported spline wavelets are used to efficiently identify the 3D spatial location of embedded cracks in conducting slabs of finite thicknesses. Within the proposed approach a couple of oppositely arranged field sensor pairs are employed on top and bottom of the conducting slab under consideration. Details of the crack detection methodologies and simulations are given in the paper.

Crack location identification of rotating rotor systems using operating deflection shape data

Crack location identification, as one key destination of structural health monitoring, is still a challenge for operating rotor systems. The operating deflection shape (ODS), which represents a visual description of the structural vibration patterns under operating conditions, has been gaining importance for structure damage detection in recent years. The ODS carries damage information of a structure, however, it is also difficult to detect weak cracks of rotor directly. The approximate waveform capacity dimension (AWCD) method was successfully applied to damage detection of plates and beam-like structures. In this paper, a strategic approach that combines ODS and weighted AWCD is proposed for crack location identification of the rotating rotor. To eliminate the false peaks of AWCD and obtain desirable results, a weight factor and ODS curvature data are introduced to the expression of the weighted AWCD. The effectiveness of the proposed method is validated by numerical simulation and experimental investigation in a cracked rotor system. The results indicate that the proposed approach not only provides good identifying performance for incipient rotor cracks, but also effectively eliminates the fault peaks introduced by the inflexion locations of ODSs. Moreover, the proposed approach proves promising in detecting crack locations of rotating rotor systems.

Crack Position Recognition of Metal Deep Drawing Parts Based on BP Neural Network

Crack position of metal drawing parts molding was analyzed by the BP neural network. First analysis of the drawing parts forming process may crack in different position. The BP neural network location identification was introduced in the basic process. 11 characteristic parameters from the drawing parts may crack position were gathered by acoustic emission signal acquisition system of deep drawing process. Then the BP neural network was designed rational, and carried out appropriate conduct to train and test. Establishing deep drawing parts of the relations between the different positions crack acoustic emission characteristic parameters and the corresponding position. Crack location was identified, in order to achieve the purpose of positioning the work piece forming process. The better method of acoustic emission location issues are resolved, metal deep drawing forming of crack location identification for basis. Provide the basis for metal drawing parts forming crack location identification.

Three-steps-meshing based multiple crack identification for structures and its experimental studies

Multiple crack identification plays an important role in vibration-based crack identification of structures. Traditional crack detection method of single crack is difficult to be used in multiple crack diagnosis. A three-step-meshing method for the multiple cracks identification in structures is presented. Firstly, the changes in natural frequency of a structure with various crack locations and depth are accurately obtained by means of wavelet finite element method, and then the damage coefficient method is used to determine the number and the region of cracks. Secondly, different regions in the cracked structure are divided into meshes with different scales, and then the small unit containing cracks in the damaged area is gradually located by iterative computation. Lastly, by finding the points of intersection of three frequency contour lines in the small unit, the crack location and depth are identified. In order to verify the effectiveness of the presented method, a multiple cracks identification experiment is carried out. The diagnostic tests on a cantilever beam under two working conditions show the accuracy of the proposed method: with a maximum error of crack location identification 2.7% and of depth identification 5.2%. The method is able to detect multiple crack of beam with less subdivision and higher precision, and can be developed as a multiple crack detection approach for complicated structures.

Approximate analytical method for damage detection in free–free beam by measurement of axial vibrations

Nowadays, sophisticated structures and machinery parts are constructed by using metallic beams. Beams are widely used as structural element in civil, mechanical, naval, and aeronautical engineering. In structures and machinery, one undesirable phenomenon is crack initiation in which the impact cannot be seen overnight. Cracks develop gradually through time that lead finally to catastrophic failure. Therefore, crack should be monitored regularly with more care. This will lead to more effective preventive measure and ensure continuous operation of the structure and machine. Damage in structure alters its dynamic characteristics. The change is characterized by change in modal parameters, that is, modal frequencies. Thus, vibration technique can be suitably used as a nondestructive test for crack detection of component to be tested. Mostly modal frequencies are used for monitoring the crack because modal frequencies are properties of the whole structure component. In this paper, efforts are made to develop suitable methods that can serve as the basis to detection of crack location and crack size from measured axial vibration data. This method is used to address the inverse problem of assessing the crack location and crack size in various beam structure. The method is based on measurement of axial natural frequencies, which are global parameter and can be easily measured from any point on the structure and also indeed, the advantage in modeling complexity. In theoretical analysis, the relationship between the natural frequencies, crack location, and crack size has been developed. For identification of crack location and crack size, it was shown that data on the variation of the first two natural frequencies is sufficient. The experimental analysis is done to verify the practical applicability of the theoretical method developed.

Approximate Analytical Method for Damage Detection in Beam

Damage in structure alters its dynamic characteristics. The change is characterized by change in modal parameters i.e., modal frequencies. Mostly modal frequencies are used for monitoring the crack because modal frequencies are properties of the whole component. Thus vibration technique can be suitably used as a nondestructive test for testing crack detection of component. In this article efforts are made to develop a suitable method that can serve as a basis for detection of crack location and crack size from measured axial vibration data. This method is used to address the inverse problem of assessing the crack location and crack size in various beam structure. The method is based on measurement of axial natural frequencies, which are global parameter and can be easily measured from any point on the structure. In theoretical analysis the crack is simulated by an equivalent axial spring, connecting the two segments of the beam. Analysis of this approximate model results in algebraic equation, which relates the natural frequencies of the beam and crack location. Also the relationship between the natural frequencies, crack location, and crack size has also been developed. For identification of crack location and crack size, it was shown that data on the variation of the first two natural frequencies is sufficient. The experimental analysis is done to verify the practical applicability of the theoretical method developed.

Identification of crack location and crack size in a simply supported beam by measurement of natural frequencies

In this paper, theoretical and experimental analysis of a vibrating, simply supported beam with a crack is carried out. This method is used to address the inverse problem of assessing the crack location and crack size in various beam structures. The method is based on measurement of natural frequencies, which are of global parameter and can be easily measured from any point on the structure. In theoretical analysis the crack is simulated by an equivalent spring, connecting the two segments of the beam. Analysis of this approximate model results in algebraic equations, which relate the natural frequencies of the beam, and crack location. Also the relationship between the natural frequencies, crack location and crack size has also been developed. For identification of the crack location and crack size, it was shown that data on the variation of the first two natural frequencies is sufficient. The computation of natural frequencies of an uncracked and cracked beam is facilitated by a finite element method package. This database is to be utilized in an analytical method to address the inverse problem to identify the crack location and crack size. Experiments have also been performed on a simply supported beam. The experimental analysis is done to verify the practical applicability of the theoretical method developed. A good agreement between the predicted and actual crack location and crack size is obtained. The results obtained by theoretical and experimental methods are compared graphically.

Identification of crack location and intensity in a cracked beam by fuzzy reasoning

In this article, a soft-computing-fuzzy-logic approach for crack identification in cantilever beam has been considered. The fuzzy controller consists of six input parameters and two output parameters. The input parameters are first three relative natural frequencies and first three relative mode shape differences in dimensionless forms. The output parameters are relative crack location and relative crack depth. Theoretical analyses have been done including the effects of crack depths and crack locations on the natural frequencies and mode shapes. Several fuzzy rules are outlined for the fuzzy controller. Gaussian membership functions are used for the fuzzy controller. The local stiffnesses at crack location of the beam have been calculated using strain energy release rate. The fuzzy rules are used to identify the location and depth of the crack. Finally, the effectiveness of the developed fuzzy controller has been verified by the results obtained from the developed experimental set-up.

A robust identification of single crack location and size only based on pulsations of the cracked system

The purpose of the present work is to establish a method for predicting the location and depth of a crack in a circular cross section beam by only considering the frequencies of the cracked beam. An accurate knowledge of the material properties is not required. The crack location and size is identified by finding the point of intersection of pulsation ratio contour lines of lower vertical and horizontal modes. This process is presented and numerically validated in the case of a simply supported beam with various crack locations and sizes. If the beam has structural symmetry, the identification of crack location is performed by adding an off-center placed mass to the simply supported beam. In order to avoid worse diagnostic, it was demonstrated that a robust identification of crack size and location is possible if two tests are undertaken by adding the mass at the left and then right end of the simply supported beam. Finally, the pulsation ratio contour lines method is generalized in order to be extended to the case of rectangular cross section beams or more complex structures.

Identification of Crack Damage with Wavelet Finite Element

An improved method to identify the crack location and size is presented which takes advantages of wavelet finite element (WFE). The important property of wavelet analysis is the capability to represent functions in a dynamic multiscale manner, so solution with WFE enables a hierarchical approximation to the exact solution. WFE has good ability in modal analysis for singularity problems like a cracked beam. The crack in a beam is modeled with WFE and represented as a rotational spring. The additional flexibility caused by crack in its vicinity is evaluated according to linear and elastic fracture mechanics theory. The WFE stiffness matrix of the crack is constructed and the algorithm for crack identification through the use of vibration-based inspection (VBI) is developed. With the accurate natural frequencies obtained from the transient signal measured, graphs of crack equivalent stiffness versus crack location are plotted, by providing the first three natural frequencies as an input. The intersection of the three curves gives the crack location and size. Experimental studies of cracked shafts are presented to demonstrate the accuracy of the method. The error in identification of crack location and size are both less than 2%. This study provides the new method for the diagnosis of incipient small crack.