Ensemble Empirical Mode Decomposition based methodology for ultrasonic testing of coarse grain austenitic stainless steels Govind K. Sharma a , Anish Kumar a,⇑ , T. Jayakumar a , B. Purnachandra Rao a , N. Mariyappa b a Metallurgy and Materials Group, Indira Gandhi Centre for Atomic Research (IGCAR), Kalpakkam, India b National Institute of Mental Health and Neuro Sciences (NIMHANS), Bangalore, India article info Article history: Received 7 March 2014 Received in revised form 6 November 2014 Accepted 18 November 2014 Available online 28 November 2014 Keywords: Ultrasonic scattering Austenitic stainless steel Ensemble Empirical Mode Decomposition (EEMD) Intrinsic Mode Functions (IMFs) Signal minimisation abstract A signal processing methodology is proposed in this paper for effective reconstruction of ultrasonic sig- nals in coarse grained high scattering austenitic stainless steel. The proposed methodology is comprised of the Ensemble Empirical Mode Decomposition (EEMD) processing of ultrasonic signals and application of signal minimisation algorithm on selected Intrinsic Mode Functions (IMFs) obtained by EEMD. The methodology is applied to ultrasonic signals obtained from austenitic stainless steel specimens of differ- ent grain size, with and without defects. The influence of probe frequency and data length of a signal on EEMD decomposition is also investigated. For a particular sampling rate and probe frequency, the same range of IMFs can be used to reconstruct the ultrasonic signal, irrespective of the grain size in the range of 30–210 lm investigated in this study. This methodology is successfully employed for detection of defects in a 50 mm thick coarse grain austenitic stainless steel specimens. Signal to noise ratio improvement of better than 15 dB is observed for the ultrasonic signal obtained from a 25 mm deep flat bottom hole in 200 lm grain size specimen. For ultrasonic signals obtained from defects at different depths, a minimum of 7 dB extra enhancement in SNR is achieved as compared to the sum of selected IMF approach. The application of minimisation algorithm with EEMD processed signal in the proposed methodology proves to be effective for adaptive signal reconstruction with improved signal to noise ratio. This methodology was further employed for successful imaging of defects in a B-scan. Ó 2014 Elsevier B.V. All rights reserved. 1. Introduction Defect detection in coarse grain materials is difficult by ultra- sonic time domain amplitude evaluation since the amplitude of defect echoes can be affected by many factors such as grain size and inspection frequency. Often, the defect echoes are comparable to the microstructural ultrasonic scattering noise also known as grain noise. The detection of defects in the presence of grain noise becomes difficult even if the defects are substantially larger than the grains. The grain echo decorrelation technique resulting from shift in the transducer position has been reported for enhancement of signal to noise ratio (SNR) [1,2]. A broadband transducer is used to transmit a set of narrow band signals. Further, rectification and averaging operations on the received echoes result in, defect to grain echo enhancement. In recent years, array of programmable transmitters and receivers have been used to improve the ultrasonic flaw detection in strong scattering media. Efforts have been made to separate single scattering echo from multiple scat- tering background. Various combinations of single scattering filter [3], random matrix approach [4] and total focussing method (TFM) with virtual array approach [5] have been used in conjunction with decomposition of time reversal operator (DORT) to improve the SNR and enhance the probability of detection of defects embedded in strongly scattering media. Another approach known as Split Spectrum Processing (SSP) was proposed by Newhouse et al. [6] for enhancement of SNR in large grain materials. This approach produces frequency diverse quasi-decorrelated signals from the received broad band signal by digital filtering. The echo of a broad band transmitted signal is split into many spectral components which are processed and sub- sequently recombined [6]. The recombination algorithms make use of the fact that the echoes from closely spaced unresolvable scat- terers such as grains are dominated by interference. The ampli- tudes from these scatterers depend on frequency content of the ultrasonic signal. However, the amplitudes from the echoes of large and well separated scatterers are less affected by changes in ultra- sonic frequency. This is known as frequency diversity. The signal http://dx.doi.org/10.1016/j.ultras.2014.11.008 0041-624X/Ó 2014 Elsevier B.V. All rights reserved. ⇑ Corresponding author at: Ultrasonic Measurements Section, Non-Destructive Evaluation Division, Indira Gandhi Centre for Atomic Research (IGCAR), Kalpakkam, India. Tel.: +91 44 27480232. E-mail addresses: anish@igcar.gov.in, anishkumar@live.co.uk (A. Kumar). Ultrasonics 57 (2015) 167–178 Contents lists available at ScienceDirect Ultrasonics journal homepage: www.elsevier.com/locate/ultras