Project 5598 - Multi-Mode ultrasonic imaging and Data Fusion for laser ultrasonics
Project Type: Individual
Maximum Number of Students: 1
Project Description
Ultrasound is a simple and powerful technique that allows to detect defects inside optically opaque materials without causing any damage, a process called non-destructive evaluation (NDE). During the past twenty years ultrasonic phased arrays have found their way in NDE as they offer ultrasonic imaging instead of simple defect detection. Conventional arrays comprise a number of elements that can independently generate and detect ultrasonic waveforms. Currently, the “gold standard” in ultrasonic imaging for NDE is a data acquisition method where the array is used to capture signals of all possible combinations between transmission and detection array elements. This method of data acquisition is called the “Full Matrix Capture” (FMC). FMC has given the power to perform. the focusing and steering of ultrasonic waves in post processing using a variety of imaging algorithms. One of those algorithms is the Total Focusing Method (TFM) where the ultrasonic beam is focused at each point in the ultrasonic image. TFM is currently considered the benchmark in NDE as it is almost impossible to perform. this technique during real time and gives high definition ultrasonic images. Even more recently, FMC and TFM have been made possible for laser ultrasonics, which is an ultrasonic technique that uses lasers to generate and detect ultrasound. Laser ultrasound has the unique advantage of being a remote ultrasonic technique, making it suitable for extreme environments, such as the hot environment of a manufacturing process or even space, which is why there is a strong industrial interest for this technique.
The concept of Laser Induced Phased Arrays (LIPA) synthesises an array of ultrasonic generation sources and detectors in post processing by mechanically scanning laser beams. Following data acquisition with LIPA, TFM ultrasonic images can be produced. Composing a TFM image is usually done by choosing the ultrasonic wave mode (longitudinal or shear waves) of the excited ultrasound. All wave modes are excited during laser ultrasonic generation and when waves are reflected off internal boundaries of the investigated structure, mode conversion takes place, which causes the appearance of additional modes. All these modes are detected and contain different kind of information about the structure and should be taken into account, especially when characterisation of a defect is desired. Data fusion of the information from all useful modes can be used in order to synthesise a single TFM image for efficient imaging.
The aim of this project is to develop a TFM algorithm tailored to laser ultrasonics that will be able to image various ultrasonic modes excited in the material and fuse the information in a single ultrasonic image. Part of the project will require capturing experimental FMC data sets from conventional, transducer arrays and LIPAs. Matlab knowledge and an interest in signal processing are essential.
Key Objectives
Understand the principles of phased array ultrasonics and laser ultrasonics.
Understand the principles of FMC and TFM for the purpose of NDE.
Familiarisation with the use of a TFM algorithm for laser ultrasonics.
Development of multi-mode data analysis and TFM ultrasonic images.
Data fusion of various ultrasonic modes into a single ultrasonic image
Comparison of images from multi-mode analysis with conventional single-mode TFM.
Demonstrate the potential of multi-mode and data fusion for the ultrasonic imaging of an industry relevant component.
Supervisor
Stratoudaki, Dr Theodosia (SENIOR LECTURER - EEE)
Class Codes
19496 EEE Individual Project
EE900 MSc Project
EE990 MSc Project
EM401 EME Individual Project
Areas
Signal Processing
Ultrasonics
Topics
Sensor Systems
Optical Technologies
Image & Video Processing
Mechanical Engineering
Data Analysis
Non Destructive Testing
Ultrasonics
Transducer Systems
Transducer Design
Laser Technologies
Activities
Computer-based
Experimental
Measurement
Analysis
Research
Small signal
Programming