FTT Deutschland GmbH
FTT Deutschland (FTTD) is a small company with its office in Wildau, south of Berlin. We are employing
about 30 engineers developing customized solutions for advanced turbomachinery testing & validation programs. For our sonic-infrared (SIR) inspection development program we are looking for a motivated student with interest in experimental lab work and background in process automation and image processing to work the task described as follows.Turbomachinery and aerospace applications use a wide range of highly loaded and safety-critical components.
To assure a safe operation it is crucial to inspect for integrity of these components. Fluorescent Penetrant Inspection (FPI) as the industry standard is widely used to inspect for defects such as open surface cracks. FPI requires manual work, it employs chemicals and large amounts of water. SIR inspection relies on a digital workflow and offers an environmentally friendly alternative to FPI. FTTD is building a demonstrator of a SIR inspection machine in its facilities in Wildau.
The working principle of this vibrothermographic inspection device is such that an ultrasonic excitation source is exciting the inspected part with frequencies up to 20kHz. The resonances that occur, cause a relative displacement at the crack flanks of a potential defect. Due to the friction at the crack flanks heat is
introduced into the inspection part. A high-precision IR-camera records the surface temperature of
the inspected part. The steering of inspection and post-processing is happening in an inspection
computer, as shown in Fig. 1.
It is one of the key challenges to efficiently post-process the recorded thermal images by algorithms
to identify patterns in the temperature fields that indicate the location of a potential defect. Commonly these algorithms are named defect recognition algorithms (DRA). The simpler DRAs process the temperature change of during excitation time and a) subtract the (unexcited) cold image
Proprietary information of FTT Deutschland GmbH (“FTTD”). Use, disclosure, dissemination,
distribution, or copying to any outside parties without express written permission of FTTD is strictly
forbidden. or b) correlate the thermal profile with (reference) temperature curves of known (typical) defects.
Further developed algorithms use statistical methods such as principal component analysis to
condense the data to meaningful measures to identify defects. A general functional breakdown for
any of the DRAs can be summarized as follows:
- Generate Crack Model Based on SIR Parameters
- Background Subtract SIR Inspection Data
- Apply Filter to Reduce Spatial and Temporal Noise
- Data Reduction
- Apply Correlation Methods
- Process and Display Results to the User
FTTD has a reference Labview program available that assures steering of excitation device and reading
of camera data. As a first step, the capabilities of the existing software are to be understood. The focus
shall be to gather available interfaces and understand raw data files. Reference inspections shall be
performed by execution of inspection routines such that multiple reference datasets are generated.
This thesis is aiming on implementation of one new DRA into an existing Labview environment. FTTD
has documentation of various DRA concepts available that can be used as a guideline. Further it is
seen as a task of the thesis to study methodologies & recent advancements in image / pattern
recognition and judge for their applicability onto the available image data. The thesis shall recommend
for a new DRA methodology and implement it into the available Labview environment (or alternatively
an external routine such as Python would be acceptable).
The enhanced inspection routine shall be tested. It is desired but not a requirement for this thesis to
have this process working in real time (alternatively, a post-processing step is acceptable). Ultimately
the outcome of the thesis shall be documented.
One of our students is currently working in the company and has been developing on the topic. We are now looking for a successor to continue the work.
Tasks:
– Literature study and making familiar with the following methods:
o Basics for vibro-thermographic inspection process
o Labview software environment
o Image recording and pattern recognition
– Review of exiting Labview software package
– Familiarization with DRA concepts
– Review of pattern recognition software methodology
– Down selection of new DRA method and implementation strategy
– Enhancement of Labview based inspection software package by new DRA
– Perform inspections to generate reference datasets by applying enhanced software
– Documentation
If we have aroused your interest, you can send your application documents to the address below or, if you have any queries, please contact Alexander Lange directly at:
FTTD Focal point
Alexander Lange
Tel: +(49) 3375 49720 15
Email: alange@fttgmbh.de
Um sich für diesen Job zu bewerben, sende deine Unterlagen per E-Mail an alange@fttgmbh.de
