Novel Strain Field Imaging Using Electronic Speckle Pattern Interferometry
Heikkinen, Juuso Jalmari (2016)
2016
Teknis-luonnontieteellinen koulutusohjelma
Luonnontieteiden tiedekunta - Faculty of Natural Sciences
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Hyväksymispäivämäärä
2016-08-17
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:tty-201608054396
https://urn.fi/URN:NBN:fi:tty-201608054396
Tiivistelmä
Growing demands for cost and energy efficiency set requirements for developing lighter and stronger products. The strength of material is characterized by maximum tolerable stress. For a given material, stress is directly related to strain, and strain measurement provides a practical way for estimating mechanical strength. Traditional strain gauges are simple and very robust, but their use is limited, since the measurements are taken at only discrete locations; long setup times are required; and maximum strain resolution is limited.
Electronic speckle pattern interferometry (ESPI) is a very sensitive optical method for measuring full field displacements on the surface of a test object. Strain field is obtained by numerically differentiating the displacement field. Full field recording makes ESPI practical for quality control applications, since material faults cause strain concentrations that would be impractical or even impossible to detect using strain gauges. However, ESPI recordings contain noise that is amplified by the differentiation process. In the presence of large strains the noise level further rises due to increasing speckle decorrelation effect. Hence, the strain measurement range is ultimately limited by the ability to successfully filter out measurement noise.
In this thesis, a filtering algorithm was developed to obtain good performance in the presence of large strains. In the algorithm, the measured displacement field is first repetitively mean filtered. The strain field is then calculated from the filtered displacement field. Amplified residual noise is removed by repetitively filtering the calculated strain field. Unlike in existing methods, data is handled in phasor format. This removes the need to unwrap the data, which often is the most critical and time- consuming step. In addition, the repetitive mean filtering is computationally simple and essentially self-adaptive. Compared to existing ESPI filtering methods, the developed algorithm achieves comparable or better performance but is also significantly faster. This enables real time monitoring of the strain field.
As a second objective, a novel measurement setup was developed. Current ESPI systems require high-coherence lasers that tend to be costly and bulky. In this thesis, a compact and affordable laser diode was used in combination with a reflection diffraction grating. The grating relaxes coherence requirements so that even a low- coherence laser diode can be applied. The novel measurement setup shows promising performance but the obtained strain resolution is slightly lower than that of high- coherence ESPI. It is expected that with small hardware improvements the laser diode based ESPI can narrow the gap to the high-coherence ESPI.
Electronic speckle pattern interferometry (ESPI) is a very sensitive optical method for measuring full field displacements on the surface of a test object. Strain field is obtained by numerically differentiating the displacement field. Full field recording makes ESPI practical for quality control applications, since material faults cause strain concentrations that would be impractical or even impossible to detect using strain gauges. However, ESPI recordings contain noise that is amplified by the differentiation process. In the presence of large strains the noise level further rises due to increasing speckle decorrelation effect. Hence, the strain measurement range is ultimately limited by the ability to successfully filter out measurement noise.
In this thesis, a filtering algorithm was developed to obtain good performance in the presence of large strains. In the algorithm, the measured displacement field is first repetitively mean filtered. The strain field is then calculated from the filtered displacement field. Amplified residual noise is removed by repetitively filtering the calculated strain field. Unlike in existing methods, data is handled in phasor format. This removes the need to unwrap the data, which often is the most critical and time- consuming step. In addition, the repetitive mean filtering is computationally simple and essentially self-adaptive. Compared to existing ESPI filtering methods, the developed algorithm achieves comparable or better performance but is also significantly faster. This enables real time monitoring of the strain field.
As a second objective, a novel measurement setup was developed. Current ESPI systems require high-coherence lasers that tend to be costly and bulky. In this thesis, a compact and affordable laser diode was used in combination with a reflection diffraction grating. The grating relaxes coherence requirements so that even a low- coherence laser diode can be applied. The novel measurement setup shows promising performance but the obtained strain resolution is slightly lower than that of high- coherence ESPI. It is expected that with small hardware improvements the laser diode based ESPI can narrow the gap to the high-coherence ESPI.