 Original Article
 Open Access
 Published:
Identification of elastic parameters of an inclusion by a particle filter using ultrasonic waves
International Journal of Mechanical and Materials Engineering volume 10, Article number: 23 (2015)
Abstract
Background
Ultrasonic waves have been used as a nondestructive testing tool to evaluate defects in steel mill products. Although an ultrasonic wave has the capability to detect defects, the type classification of defects and material characterization of components continue to be challenging tasks. The aim of this paperis to develop an identification method to determine the elastic properties of an inclusion.
Methods
A particle filter (PF) is applied to identify the wave velocity and density of an inclusion. The PF is a data assimilation technique that is based on a Bayesian approach, and requires a reliable simulation tool to describe the system and measurements.
Results
The estimated value by PF showed good convergence to the true value of the wave velocity anddensity of the inclusion.
Conclusions
By using a finite integration technique as the simulation tool, the PF can estimate the wavevelocity and density of the inclusion in steel accurately and successfully.
Background
Detecting inclusions such as acid slags in steel mill products without causing damage is important for guaranteeing their quality and performance. Ultrasonic waves are widely used to detect inclusions, and the size and location of the inclusions can be evaluated from the flight time and amplitude of the scattered wave (Sachse 1975; Bifulco & Sachse 1975). In general, an ultrasonic signal is affected by changes in size and shape as well as the elastic properties of the inclusions. Over the past few decades, many studies on material characterization using ultrasonic waves have been reported. A technique to determine the grain size and homogeneity of materials using backscattered ultrasonic waves was proposed by Willems and Goebbels (Willems & Goebbels 1981). Castagnède et al. showed the decision method to determine the elastic constants in solid using frequency shift information and the velocity of an ultrasonic wave (Castagnède et al. 1990). Luo and Bungey used surface waves to determine the Young modulus and Poisson ratio of concrete materials (Luo & Bungey 1996), and Rogers utilized the phase velocity of Lamb waves for the estimation of the elastic constants in plates (Rogers 1995).
A statespace approach (Gandossi & Simola 2006; Khan & Ramuhalli 2008; Kaipio & Somersalo 2005) based on Bayesian theory has been applied for the identification of defect properties. The Kalman filter (Kalman 1960) is a Bayesian filtering method that can estimate the state variables in a linear dynamic system. The Kalman filter was also extended for solving various linear problems (Fahrmeir 1992; Julier & Uhlmann 1997). Meanwhile, filtering algorithms such as the Monte Carlo filter (Kitagawa 1996) and the Bayesian bootstrap filter (Gordon et al. 1993) were proposed for problems with a nonlinear system. The Monte Carlo and Bootstrap filter methods express the state variables using random data sampling and associated weights. These methods are often referred to as particle filters (Kaipio & Somersalo 2005).
In this paper, a particle filter (PF) is applied for the identification of the elastic properties, wave velocity, and density of an inclusion in a steel mill component. The PF can numerically solve these identification problems to modify the state variable using available observation data. The PF is used in conjunction with a reliable numerical simulation model that describes ultrasonic wave transmission and reception in ultrasonic testing (UT). Here, ultrasonic signals are calculated with a finite integration technique (FIT) (Fellinger et al. 1995; Schubert 2004). We verify the effectiveness of the PF for UT through the demonstrations of the elastic parameters identification. In this paper, shear horizontal (SH) ultrasonic waves are used for the identification of wave velocity and density of an inclusion.
Method
Particle filter (PF)
The PF aims to estimate the state variables x _{ t } for time t while incorporating measured variable data y _{ t }. Generally, the state variables x _{ t } are hidden in the measured data. The relationship between the state variable vector x _{ t } and measured data y _{ t } is represented by the measurement model as
where w _{ t } denotes the measurement noise which is distributed with a mean of κ and a variance of σ ^{2}. The system model which describes the next state vector x _{ tt1 } is written by using the previous state vector x _{ t1 } as
where v _{ t } represents the system noise. The functions h and f in Eqs. (1) and (2) are known. In the PF, a total of N samples (particles) are used to show the probabilistic density of x _{ t }, and the particles are labeled with superscripts as x _{ t } ^{(1)}, x _{ t } ^{(2)}, . . ., x _{ t } ^{(N)}. At every step, the importance likelihood λ _{ t } of each particle is calculated as
where m indicates the number of dimensions of the vector x _{ t }, and R represents the variancecovariance matrix. The flow of calculation on the PF process in this paper is as follows:

A.
x _{ o } ^{(i)} : Initial particles are randomly generated.

B.
For t = 1, . . ., T, steps (a), (b), and (c) described below are carried out.

(a)
For each particle i, perform the following sequence:

1.
Generate the system noise randomly : v _{ t } ^{(i)}

2.
Predict the next state variable : x _{ tt1 } ^{(i)} = f _{ t } (x _{ t1 } ^{(i)}) + v _{ t } ^{(i)}

3.
Calculate the likelihood of the particle : λ _{ t } ^{(i)}

1.

(b)
Compute the summation of the likelihood as
$$ S={\displaystyle \sum_{j=1}^N{\lambda_t}^{(j)}} $$(4) 
(c)
Sample N times with replacement from the set of particles x _{ t } ^{(i)} according to the following normalized importance likelihood

(a)
The state vectors are modified each time and updated. As shown in Fig. 1, the particles with low likelihoods are eliminated. On the other hand, those with high likelihood remain and then split for the generation of new particles. This procedure is referred to as sequential importance resampling (SIR). In the SIR, the expected sampling value of each particle is expressed as Nβ ^{(i)}. Therefore, the ratio of the eliminated particle to the predicted particles \( {\left\{{{\boldsymbol{x}}_{t\Bigt1}}^{(i)}\right\}}_{i=1}^N \) can be expressed as \( \sqrt{\beta^{(i)}\left(1{\beta}^{(i)}\right)/N} \). It should be noted that a small sample number N may cause a large variance of the particle.
The selection of the number N of particles is a key factor for the efficient and accurate identification in the PF. The computational load and the convergence of the filter depend on this number. Most applications select a fixed number of particles in advance, using ad hoc criteria or statistical methods (Boers 1999).
Finite integration technique (FIT)
A numerical tool to calculate the ultrasonic signal is introduced in the PF. Here, we utilize a FIT for modeling SH wave transmission and reception (Nakahata & Kimoto 2012). In this section, we briefly summarize the formulation of the FIT. Cartesian coordinates (x _{1}, x _{2}, x _{3}) are considered, and the antiplane direction is set to the x _{3}axis. The particle velocity is indicated as u _{3}, and the shear stresses are τ _{31} and τ _{32}. The equation of motion and the constitutive law in the integral form are expressed as
where ρ denotes the mass density, μ represents the shear modulus, and n denotes the outward normal. The shear wave velocity in a solid is expressed using the mass density and shear modulus as
Equations (6) and (7) are discretized with integration cells of small squares as shown in Fig. 2. The stress components τ _{31} and τ _{32} are allocated at halftime steps, while the velocity v _{3} is allocated at fulltime steps. The discretization in the time domain is based on a leapfrog timemarching scheme. Let V_{3}, T_{31}, and T_{32} be computational arrays to store the solutions of v _{3}, τ _{31} and τ _{32}, respectively. The updating process in Eqs. (6) and (7) can be expressed as
where δ and ε are defined as
Therefore, Eqs. (9) and (10) are executed by incrementing the time step Δt in sequence. Various boundary conditions were explained by Schubert (Schubert 2004).
Since the FIT uses a unified grid size, an imagebased model can be applied. In imagebased modeling, a numerical model can be constructed from digital images and then pixel data can be directly fed into the FIT (Nakahata et al. 2014). In the PF, a number of simulations have to be performed to model the functional h in Eq. (1). Since a parallel calculation with graphics processing units (GPUs) showed good speed efficiency in our previous paper (Nakahata & Kimoto 2012), we use the same calculation method in this study and accelerate the FIT simulation.
Simulation of identification of elastic parameters
Consider twodimensional (2D) problems of the identification of the SH wave velocity c _{ T } and the density ρ of an inclusion in a steel material. The state variables for this problem are expressed as
In this paper, these two variables are independent of the time. Therefore, the system model can be expressed using the prior state and system noise as
As shown in Fig. 3, we consider an inclusion with a diameter of 10 mm, which is embedded in steel (c _{ T } = 3100 m/s, ρ = 7850 kg/m^{3}). An ultrasonic SH wave with a center frequency of 1.0 MHz is transmitted into the steel material from a transducer with a diameter of 10 mm and located on the top surface. The frequency spectrum of the reflected wave from the inclusion is used in the PF identification.
Results and discussion
In Fig. 4, a numerical example of SH wave propagation calculated with the FIT is shown. This figure illustrates the magnitude of displacement u _{3}(=∫v _{3} dt) at certain time steps. First, the incident wave is emitted from the transducer (Fig. 4a) and then scatters at the upper interface between the steel and the inclusion. The scattered wave is recorded at the same transducer as the first reflection echo (Fig. 4c). Meanwhile, a part of the incident wave propagates through the inside of the inclusion and then scatters at the lower interface between the inclusion and the steel. The scattered wave from the lower interface is recorded as the second reflected wave (Fig. 4e).
In this study, the Fourier spectrum of the reflected echo is used for the evaluation of the likelihood in Eq. (3). The PF uses a large number (normally more than 1000) of particles. In this research, however, ten particles (N = 10) are used due to the simplicity of the problem. As shown in Fig. 3, we assumed that the true values of c _{ T } and ρ of the inclusion were 2400 m/s and 7000 kg/m^{3}, respectively. Although the measured signals were supposed to be used in the identification process, we substituted the artificial signal obtained via numerical calculations for the measured one. Figure 5 shows the transition of the ten particles whose initial positions were located at regular intervals. Initially, c _{ T } and ρ were distributed between 1000 and 2800 m/s, and 1000 and 10000 kg/m^{3}, respectively. From Fig. 5, it can be seen that most particles gathered around the true value after several time steps. The convergent point can indicate the true values of c _{ T } and ρ of the inclusion.
Figure 6 shows another result of the PF that commenced with a different initial particle distribution. In this case, we aligned initial particles at a regular interval with respect to only c _{ T }. At first, c _{ T } and ρ were distributed between 1000 and 3000 m/s, and 4000 and 8500 kg/m^{3}, respectively. The calculation of the PF was terminated after thirty calculation steps. Although good convergence at early steps was not seen, the particles converged on the true value eventually.
Conclusions
In this paper, we proposed an identification method for determining the elastic parameters of an inclusion in a steel material as an UT tool. A PF was applied to identify the wave velocity and density in an inclusion. The PF is a data assimilation technique based on a Bayesian approach. In the PF, a FIT was used to assist the description of the measurement model in UT. In the simulation, the estimated value of the wave velocity and density showed good convergence to the true value.
In the future, we aim to validate our approach using experimentally measured signals and apply it for the identification of multivariable parameters with a large number of particles. Furthermore, we will apply our method to 3D problems using a GPUaccelerated 3D FIT simulation.
References
Bifulco, F, & Sachse, W. (1975). Ultrasonic pulse spectroscopy of a solid inclusion in an elastic solid. Ultrasonics, 13, 113–116.
Boers, Y. (1999). On the number of samples to be drawn in particle filtering. IEE Colloquium on Target Tracking: Algorithms and Applications 5/1 –5/6.
Castagnède, B, Jenkins, JT, Sachse, W, & Baste, S. (1990). Optimal determination of the elastic constants of composite materials from ultrasonic wave speed measurements. Journal of Applied Physics, 67, 2753–2761.
Fahrmeir, L. (1992). Posterior mode estimation by extended kalman filtering for multivariate dynamic generalized linear models. Journal of the American Statistical Association, 87, 501–509.
Fellinger, P, Marklein, R, Langenberg, KJ, & Klaholz, S. (1995). Numerical modeling of elastic wave propagation and scattering with EFIT—elastodynamic finite integration technique. Wave Motion, 21, 47–66.
Gandossi, L, & Simola, K. (2006). Bayesian analysis of flaw sizing data of the NESC III exercise. International Journal of Pressure Vessels and Piping, 83, 654–662.
Gordon, NJ, Salmond, DJ, & Smith, AFM. (1993). Novel approach to nonlinear/nongaussian bayesian state estimation. IEEE Radar and Signal Processing, 140, 107–113.
Julier, SJ, & Uhlmann, JK. (1997). A new extension of the kalman filter to nonlinear systems. Proceedings of SPIE, 3068, 182–193.
Kaipio, J, & Somersalo, E. (2005). Statistical and computational inverse problems. New York: Springer.
Kalman, RE. (1960). A new approach to linear filtering and prediction problems. Journal of Basic Engineering, 82, 35–45.
Khan, T, & Ramuhalli, P. (2008). A recursion bayesian estimation method for solving electromagnetic nondestructive evaluation inverse problems. IEEE Transactions on Magnetics, 44, 1845–1855.
Kitagawa, G. (1996). Monte carlo filter and smoother for nongaussian nonlinear state space models. Journal of Computational and Graphical Statistics, 5, 1–25.
Luo, Q, & Bungey, JH. (1996). Using compression wave ultrasonic transducers to measure the velocity of surface waves and hence determine dynamic modulus of elasticity for concrete. Const Build Mater, 10, 237–242.
Nakahata K, & Kimoto K. (2012). Realtime imagebased FIT simulation using GPU computing and its application to nondestructive testing. Proceedings of 6th European Congress on Computational Methods in Applied Sciences and Engineering, 42854297.
Nakahata, K, Chang, J, Takahashi, M, Ohira, K, & Ogura, Y. (2014). Finite integration technique for coupled acoustic and elastic wave simulation and its application to noncontact ultrasonic testing. Acous Sci Tech, 35, 260–268.
Rogers, WP. (1995). Elastic property measurement using rayleighlamb waves. Research in Nondestructive Evaluation, 6, 185–208.
Sachse, W. (1975). Scattering of ultrasonic pulses from cylindrical inclusions in elastic solids. Iowa State University Digital Repository. http://lib.dr.iastate.edu/cnde_yellowjackets_1975/16. Accessed 8 March 2013.
Schubert, F. (2004). Numerical timedomain modeling of linear and nonlinear ultrasonic wave propagation using finite integration techniques—theory and applications. Ultrasonics, 42, 221–229.
Willems, H, & Goebbels, K. (1981). Characterization of microstructure by backscattered ultrasonic waves. Met Sci, 15, 549–553.
Acknowledgements
This work was partly supported by a GrantinAid for challenging Exploratory Research No.15 K14821 from the Ministry of Education, Culture, Science and Technology (MEXT), Japan.
AI received her B. Eng. (Mechanical) in 2008 from UTM, Malaysia. Upon graduation, she worked as an Asst. Engineer in a biodiesel plant. In 2009, she joined UTeM, Malaysia as a Tutor. Later in 2011, she completed her M. Eng (Mechanical) from UKM, Malaysia. She was appointed as a Lecturer in Faculty of Mechanical Engineering, UTeM since 2011. Currently, she is pursuing her study in Ehime University, Japan focusing on crack detection using ultrasonic nondestructive testing.
AZ received her B. Eng. (Civil and Environmental) from Ehime University, Japan in 2015. Currently she is a master student in the Department of Civil and Environmental Engineering, Ehime University, Japan. Her current research is a study on application of particle filters in ultrasonic testing.
KN received his Ph.D in Civil Engineering from Tohoku University, Japan in 2003. He has started work as an Associate Professor at Ehime University, Japan since 2004. In 2008, he was a research fellow at IZFPD, Fraunhofer Institute, Germany. His interesting fields are numerical modeling of ultrasonic and electromagnetic wave for nondestructive testing.
Author information
Additional information
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
AI drafted the manuscript. AZ carried out the simulation work. KN guided the entire research work and made vital discussions. All authors read and approved the final manuscript.
Rights and permissions
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.
About this article
Received
Accepted
Published
DOI
Keywords
 Ultrasonic testing
 Particle filter
 Material characterization
 Finite integration technique