NON-DESTRUCTIVE IDENTIFICATION OF

MECHANICALLY STRONGER COMPOSITE PLATES

 

Aim

This assignment is designed to give students an introduction to digital signal and image processing (DSIP) through their applications in ultrasonic non-destructive evaluation of engineering structures. You are required to use the Matlab software package to design and implement a DSIP system to identify a mechanically stronger plate manufactured using composite carbon fibre materials.

 

Learning Outcomes

This assignment will enable you to:

Use an appropriate software package (Matlab) to investigate appropriate DSIP algorithms for ultrasonic non-destructive evaluation.

Perform signal analysis in the time and frequency domain.

Design and implement digital filters for signal denoising.

Observe and explain the effects of digital filters on noisy signals.

Generate images from signal processing.

Apply basic image statistics for decision making.

Present results graphically in different ways for performance analysis.

Identify and explain the processing stages in the system that you design and support the explanation with practical results.

Summarise your understanding of these points in a written report.

 

Marking Scheme / Report

The marks for the assignment will constitute 25% of your overall mark for this module, and the marks are allocated according to the assignment requirements discussed in the following.

 

A full report including the program to perform non-destructive identification should be submitted through Turnitin by the deadline below. The report should be structured according to the requirements described in the following, and a check list with detail marking scheme is provided as a further guide. The maximum report length is 15 pages.

 

Relevant files for the assignment are provided as email attachment together with a check list and marking scheme for assignment report writing.

 

Background and Problem:

In a carbon fibre composite part, porosity is a defect that appears as small interlaminar voids. As the level of porosity increases, the mechanical strength of the composite part decreases. To evaluate the porosity level non-destructively, ultrasonic measurement is commonly employed. The process involves transmission of high-intensity ultrasonic waves through the component part under test and assessment of the relative amplitude values of the echoes reflected respectively from the front wall and the back wall of the composite part. A stronger ultrasonic attenuation indicates a less dense composite part with higher porosity.

 

Two ultrasonic data files obtained from two laminated composite material plates with 3 mm thickness are provided as "CompositeA.mat" and “CompositeB.mat”. The data was acquired with a sampling frequency at 100 MHz over an area of 40 mm x 20 mm with 1 mm resolution. You are asked to produce a Matlab program to process the ultrasonic data and to identify which composite plate has the stronger mechanical strength overall.

 

 

Requirements

 

(1) Signal analysis

Ultrasonic signals are noisy in nature due to the back scattering phenomenon produced by the inherent microstructure of the material. The first step is to extract a typical ultrasonic signal from each composite part under test, and to carry out detail signal analysis in the time and frequency domains to identify key signal features such as locations, magnitudes and frequencies for ultrasonic echoes and noise.

 

(2) Noise reduction

With the ultrasonic measurement operating at a particular frequency band, the second step is to implement a suitable digital filtering system to reduce the impact of out-of-band noise on ultrasonic echoes, and to demonstrate the effectiveness of the filter by comparing the output with respect to input in the time and frequency domains. A good filter should yield an output signal that is as similar to the original signal as possible in the two ultrasonic echo intervals (with minimum amplitude distortion) and as near to zero as possible outside the two ultrasonic echo intervals.

 

(3) Attenuation estimation

The fourth step is to estimate the attenuation of the back wall echo with respect to the front wall echo. This involves application of the filter developed in the second step to the whole ultrasonic record of each composite plate, extraction of the two peak values from the front wall and back wall echoes in each ultrasonic signal, and calculation of the echo attenuation at each ultrasonic measurement point. The results should lead to two echo attenuation images for comparative visualisation of two composite plates.

 

(4) Part sentencing

The final step is to show the two distributions of the echo attenuation values obtained from the third step, to compute basic statistics for the two distributions, and to use the results to determine which composite plate has the stronger mechanical strength overall.