Phase aberration correction in ultrasonic phased array non-destructive testing systems

ZHAO Xia, WANG Zhao-ba, LIU Bin

(Science and Technology on Electronic Test & Measurement Laboratory, North University of China, Taiyuan 030051, China)

?

Phase aberration correction in ultrasonic phased array non-destructive testing systems

ZHAO Xia, WANG Zhao-ba, LIU Bin

(ScienceandTechnologyonElectronicTest&MeasurementLaboratory,NorthUniversityofChina,Taiyuan030051,China)

Phase aberration correction for medical ultrasound systems has attracted a great deal of attention. Since phased array techniques are now widely employed for industrial non-destructive testing (NDT) applications in various fields, the problem of phase aberrations in the process of NDT testing is considered. The technique of cross-covariance for phase aberration correction is presented. The performance of the technique for phase aberration correction is tested by means of echo signals obtained in practical non-destructive testing experiment. The results show that the technique has the better accuracy of phase correction.

ultrasonic phased array; non-destructive testing; phase aberration; correction; cross-covariance method

0 Introduction

Nowadays, ultrasonic phased array techniques are widely used in industrial non-destructive testing (NDT) field[1]. In ultrasonic phased array medical systems, a constant acoustic velocity is usually assumed when calculating the steering and focusing vectors of transmission and reception beams. Due to material inhomogeneity, the acoustic velocity is not constant but variable. As a result, both the received and transmitted pulses are improperly focused and the resolution of the systems is degraded. This effect is known as phase aberration of ultrasound systems. Through experiments, it can be found that such phenomenon also exists in ultrasonic phased array NDT systems.

To achieve optimal performance of the system, the time delays must be estimated and corrected. Some different methods for correction of ultrasonic phase aberration had been deduced in Refs.[2-6]. Among them, Flax and O’Donnel introduced a cross-correlation method for estimation of arrival time fluctuations in the received aberrated signal[4-5]. It is a classical method and has been extensively used for ultrasonic phase aberration.

This paper aims to find the causes of phase aberration in non-destructive testing system and effective method to correct the phase for ultrasonic wave aberration.

1 Cause of phase aberration

In ultrasonic phased array testing, each element of the transducer is triggered in different times to generate a beam with a constructive interference, as shown in Fig.1.

Fig.1 Ultrasonic phased array testing principle

The echo goes through the media and is reflected by the flaw. The reflected echo hits each element of the transducer with a computable time shift. Then the reflected echoes are time-shifted and summed together. The summed signal is an A-scan. It emphasizes the information from the desired focal point and attenuates other information from other points in the material.

During transmission, the phased array unit is triggered to generate a set of pulses with different time delays based on focal laws. These pulses determine the generation time of echo emitted by each element. By this way, a beam with a specific angle or focused in a desired depth can be created. This beam goes through the media and is reflected by any flaws.

During receiving, each element receives only one pulse. Focusing is performed by means of delay-and-sum (DAS) operations on the echoes received by each element. That means that the echoes are time-shifted based on the receiving focal law. Then all echoes will be summed together to form a signal and sent to acquisition instruments.

The focusing delays can be calculated by the following traditional formula

where Δtnis the required delay for thenth element (n=…,-2,-1,0,1,2,…),cis the wave speed in the medium,rnis the distance between focal point and the nth element andt0is a constant to keep the delays positive.

From Eq.(1), it can be seen that phased array ultrasound systems assume a constant acoustic velocity in the medium that is used to calculate delays for steering and focusing the ultrasound beam. Due to non-homogeneous material, the acoustic velocity is not constant but variable. This variation in acoustic velocity causes the computed steering and focusing delays of phased array transducer to be inaccurate. The beam is further disrupted. The different components of the ultrasound beam arrive at the focus out of phase. In receiving beam forming process, different time-shifted components of the ultrasound beam have phase error, therefore, the summation of the components results in a reduced resolution and signal-to-noise ratio of the system.

2 Phenomenon of phase aberration in ultrasonic phased array non-destructive testing

Lab testing system is composed of an ultrasonic phased array testing system and a linear array transducer with 64 elements from R/D Tech Company. The experiment was conducted on a steel block which had a particle defect, as shown in Fig.2.

Fig.2 Ultrasonic phased array testing of a steel block

The echoes were recorded by the 64 elements and the 64 signals are displayed from channel 1 to channel 64, as shown in Fig.3.

These original signals contain the needed information related to the defects in testing area below the phased array transducer. It means that if there are any defects or interface, the reflected echoes will be received by each element. Therefore, the original signals can be used to detect defects. The detecting process is shown in Fig.4. As shown in Fig.4(a), to find out the information of each focal point in testing area below the phased array transducer, the required delay for each element is calculated based on Eq.(1) and by means of DAS operations are performed on the original signals received by the array elements. Then by judging the amplitude of summed signals, it can be determined that whether there are any defects.

可以根據改良措施的不同將治理方法分為物理、化學、生物和綜合改良，不同地區由于土壤鹽堿化程度、性質不同，其具體適合的改良措施也不盡相同。

Take the defect point as example, as shown in Fig4(b). According to Eq.(1), the time delay of each elements is calculated. Furthermore, the defect echoes at each array element are shifted in time and summed, as shown in Fig.5.

Fig.3 Echo signals received by 64 elements of phased array transducer

Fig.4 Detecting process of defects

Theoretically, the defect echoes after time-shift should be aligned. However, from Fig.5, the beam forming removes only the time delays due to geometric path length differences between the defect and the received elements, therefore, the echo-signals received from different phased array elements are not aligned after geometrical time delays compensation. The amplitude of the summed pulse is only 8 360，as shown in Fig.5(b). To solve this problem, it need to be corrected depending on the estimation and compensation of the phase aberration between echoes of elements.

Fig.5 Defect echoes at each array elements shifted in time and summed

3 Cross-covariance correction method

Here are considered phase aberrations that are associated with time shifts of echoes received from array elements. Flax and O'Donnell proposed the classical technique for phase correction known as cross-correlation[4-5]. It is based on estimation of the cross-correlation between echoes received by adjacent transducer elements.

Supposexandyare sequences of two echo signals received by two element channels, the cross-correlation sequence of the two signals is

By analyzing, it can be found that the mean value of the echo signals received by each element channel is nonzero. As a result, the cross-correlation method is put forward which removes the influence of the nonzero mean value of the echoes. That is covariance method.

The cross-covariance of two mean-removed signals can be determined by

(3)

Based on cross-covariance theory, the most similar point of two sets of signals is found when the amplitude of cross-correlation of the two signals reaches the maximum value. If the peak position of the cross-covariance of the two signals is atMpoint, where two signals haveNpoints, respectively. The delay between the two signals can be defined as

The delay between the two signals will be a positive or a negative number. It demonstrates that one signal is behind or before another signal.

Based on the theory mentioned above, we choose non-aligned defect echoes to do cross-covariance of mean-removed calculation, as shown in Fig.6.

In Fig.6, two pairs of non-aligned defect echoes are selected: one is the defect echoes received by the 32nd element and 39th element, respectively, and the other is the defect echoes received by the 32nd element and 46th element, respectively. Each defect echo has 131 points. The cross-covariance calculation results are shown in Fig.7.

Fig.6 Aligned defect echoes

In Fig.7, for the first pair, the peak position of the calculation results is achieved at point 129. For the second pair, the peak position of the calculation results is achieved at point 117. Based on Eq.(4), it can be found that the delay between the first pair of signals is 2 points and the delay between the second pair of signals is 14 points. Therefore, this method is effective in delay estimation of two random echoes and can be used to do phase correction.

Fig.7 Cross-covariance calculation of each echoes received by different element: (a) defect echo received by the 39th element; (b) defect echo received by the 32nd element; (c) the cross-covariance of mean-removed calculation of the two echoes received by the 32nd element and 39th element; (d) defect echo received by the 46th element; (e) defect echo received by the 32nd element; (f) the cross-covariance calculation of the two echoes received by the 32nd element and 46th element

4 Correction results and analysis

For the unaligned signals in Fig.5, the cross-covariance method is used to correct it. The echo from the central element is used as the reference signal. The cross-covariance of reference signal and other signals from the group of elements beginning from the first element to the last element is calculated according to Eqs.(3) and (4). Then the relative time shift are measured between two signals from the peak position of these cross-correlation of mean-removed functions. The each echo signal is time shifted, as shown in Fig.8.

From Fig.8, it can be seen that the defect echo signals are aligned after correction and the amplitude of the summed pulse reaches to 8 960. The result shows that the signal-to-noise ratio is enhanced greatly.

Fig.8 Defect echoes at each array elements after correction

5 Conclusion

Through experiments, it can be concluded that the phase aberration which is common in ultrasound medical systems also exits in ultrasonic NDT systems. By analyzing the characteristic of the signals, the cross-covariance method is we proposed for phase aberration correction. This method proves effective by experiments for correction of signals with phase aberration. It can enhance the signal-to-noise ratio and improve the resolution of the system.

In actual industrial applications, the correction technique of cross-covariance can be used only to remove major phase aberrations resulting from sound velocity inhomogeneity. Obviously, additional adaptive compensation techniques must be used to remove beam forming problems, amplitude aberrations, and any remaining phase aberrations.

[1] Drinkwater B W, Wilcox P D. Ultrasonic arrays for non-destructive evaluation: A review. Nondestructive testing and Evaluation, 2006, 39(7): 25-541.

[2] Behar V. Techniques for phase correction in coherent ultrasound imaging systems. Ultrasonics, 2002, 39 (9): 603-610.

[3] Ries L L, Smith S W. Phase aberration correction in two dimensions using a deformable array transducer. Ultrasonics, 1995, 17(3): 227-247.

[4] Ivancevich N M, Pinton G F, Nicoletto H A, et al. Real-time 3-D contrast-enhanced transcranial ultrasound and aberration correction. Ultrasound in Medicine & Biology, 2008, 34(9): 1387-1395.

[5] O’Donnell M. Phase aberration correction system for real-time medical ultrasound imaging. In: Proceedings of the Annual Conference on Engineering in Medicine and Biology, 1990, 1: 8-10.

[6] Flax S, O'Donnell M. Phase aberration correction using signals from point reflectors and diffuse scatterers: basic principles. IEEE Transactions on Ultrasonics, Ferroelectrics and Frequency Control, 1988, 35(6): 58-767.

[7] Long R. Ultrasonic phased array inspection using full matrix capture. Insight Non-Destructive Testing and Condition Monitoring, 2012, 54(7): 380-385.

[8] Frederick C, Porter A, Zimmerman D. High-density polyethylene piping butt-fusion joint examination using ultrasonic phased array. Pressure Vessels and Piping Division of American Society of Mechanical Engineers, 2010, 6: 1295-1299.

超聲相控陣檢測回波信號相位偏差校正技術研究

趙 霞， 王召巴, 劉賓

(中北大學 電子測試技術國家重點實驗室， 山西 太原 030051)

近年來, 超聲相控陣技術廣泛應用于工業無損檢測領域。 在檢測過程中， 檢測材料和檢測系統等方面的擾動因素會造成超聲檢測回波信號明顯的相位偏差， 影響檢測效果。 本文分析研究了造成檢測信號產生相位偏差的原因以及對檢測效果造成的影響， 提出采用信號互協方差分析法實現檢測回波信號相位偏差校正。 實驗結果表明， 該方法可有效校正各通道回波信號的相位偏差， 提高檢測精度。

超聲相控陣； 無損檢測； 相位偏差； 校正； 互協方差分析法

ZHAO Xia, WANG Zhao-ba, LIU Bin. Phase aberration correction in ultrasonic phased array non-destructive testing systems. Journal of Measurement Science and Instrumentation, 2015, 6(1)： 47-52.

10.3969/j.issn.1674-8042.2015.01.009

s： National Natural Science Foundation of China(No.61201412);Ntural Science Foundation of Shanxi Province(No.2012021011-5)

ZHAO Xia (Zhaoxianuc@126.com)

1674-8042(2015)01-0047-06 doi： 10.3969/j.issn.1674-8042.2015.01.009

Received date： 2014-09-12

CLD number： TN911.7 Document code： A