Structural system identification is traditionally related to the estimation of modal parameters (natural frequencies, modal damping ratios, and mode shapes). Various well known modal methods often fail to extract these parameters for heavily damped structures with high modal densities due to the high coupling between densely packed adjacent modes. The recent development of the scanning laser Doppler vibrometer (SLDV) technology that provides efficient and massive dynamic data acquisition with high spatial density makes the new non-modal system identification techniques feasible. The proposed non-modal system identification method is based on the singular value decomposition (SVD) of the spatial mobility matrices that are acquired by the SLDV technique. Data reduction, filtering, periodization, and remapping techniques are applied to the measured data in the spatial domain. Linear and polynomial singular vector interpolation and subspace rotation techniques are applied in the frequency domain for the prediction of the spatial mobility over the frequency range of interest. This non-modal method uses measured frequency response data directly and involves neither curve fitting nor modal parameter extraction. The proposed non-modal technique was applied to a commercial business jet airplane fuselage. The measured mobility data of the fuselage were reduced to a much smaller and very efficient data set that could be easily managed, stored, and retrieved for the reconstruction and/or prediction the dynamic responses of the fuselage in both frequency and spatial domains / Ph. D.

Stateflowmodels are used for describing logic and implementing state machines in modern safety-critical software. However, the complete Stateflowmodelling language is hard to formally define, therefore a subset relevant for industrial models has been developed in previous works. Proving that the execution of Stateflow models satisfies certain safety properties is intractable in general. However, bounded model checking (BMC) can be used to either prove that safety properties are satisfied up to a bounded execution depth, commonly referred to as the reachability diameter, or find a concrete counterexample. One particular safety property of interest is an invariant property. This thesis project contributes with the following. A bounded model checking tool based on symbolic execution has been developed and is called Stateflow Model Verification Tool (SMVT). This tool has been tested on synthetic models and industrial models. The performance of Stateflow Model Verification Tool (SMVT) has been measured, but not compared against the Simulink DesignVerifier (SLDV) due to licensing issues. The study has shown that many industrial models share a similar model structure. Furthermore, it has been shown that SMVT can perform well for several models. / Stateflow-modeller används för att beskriva logik and implementation av tillståndsmaskiner i modern säkerhetskritisk mjukvara. Det kompletta Stateflowspråket är väldigt komplext, och därför har forskare tidigare definierat en begränsad version av språket relevant för industriella modeller. Bevisning att exekvering av Stateflow-modeller måste uppfylla säkerhetsegenskaper, är svårlösligt rent generellt. Begränsad modellprovning kan användas antingen för att bevisa att säkerhetsegenskaper uppfylls till ett begränsat exekveringsdjup, eller för att hitta ett motexempel. En väldigt viktig säkerhetsegenskap kallas för invariant. Detta examensarbete bidrar med följande. En begränsad modellprövare baserad på symbolisk exekvering har utvecklats och kallas för SMVT. Detta verktyg har blivit testat på syntetiska modeller samt industriella modeller. Prestandan har blivit mätt, men på grund av Simulink Design Verifier (SLDV) licens har ingen jämförelse kunnat göras. Studien har visat att många industriella modeller delar samma modellstruktur. Vidare har det utvecklade verktyget SMVT visats prestera väl för flertalet modeller.

This research presents improvements to the velocity field reconstruction process achieved through computer vision. The first improvement of the velocity reconstruction process is the automation of the scanning laser Doppler vibrometer (SLDV) pose procedure. This automated process results in superior estimates of the position and orientation of the SLDV. The second improvement is the refinement of the formulation for reconstruction of the velocity field. The refined formulation permits faster computation, evaluation, and interpretation of the reconstructed structural velocity field. Taken together, these new procedures significantly improve the overall velocity reconstruction process which results in better, unbiased out-of-plane velocity estimates in the presence of noise. The automation of the SLDV pose procedure is achieved through a computer vision model of the SLDV. The SLDV is modeled as a projective camera, i.e. an imager which preserves projectivities. This projective camera model permits the precise association of object features with image features. Specifically, circular features in the object space are seen by the SLDV as ellipses in the image space. In order to extract object points, the bitangents among the circular features are constructed and the bitangent points selected. The accuracy and precision of the object points are improved through the use of a calibrated object whose circular features are measured with a coordinate measuring machine. The corresponding image points are determined by constructing the bitangents among the ellipses and selecting the tangent points. Taken together, these object/image bitangent point sets are a significantly improved data set for previously developed SLDV pose algorithms. Experimental verification of this automated pose procedure includes demonstrated repeatability, independent validation of the estimated pose parameters, and comparison of the estimated poses with previous methods. The refinement of the velocity reconstruction formulation is a direct result of the computer vision viewpoint adapted for this research. By viewing the velocity data as images of the harmonically excited structure's velocity field, analytical techniques developed for holographic interferometry are extended and applied to SLDV velocity images. Specifically, the "absolute" and "relative" fringe-order methods are used to reconstruct the velocity field with the "best" set of bases. Full and partial least squares solutions with experimental velocity data are calculated. Statistical confidence bounds of the regressed velocity coefficients are analyzed and interpreted to reveal accurate out-of-plane, but poor in-plane velocity estimates. Additionally, the reconstruction process is extended to recover the velocity field of a family of surfaces in the neighborhood of the "real" surface. This refinement relaxes the need for the exact experimental geometry. Finally, the velocity reconstruction procedure is reformulated so that independent least squares solutions are obtained for the two in-plane directions and the out-of plane direction. This formulation divides the original least squares problem into three smaller problems which can be analyzed and interpreted separately. These refinements to the velocity reconstruction process significantly improve the out-of-plane velocity solution and interpretation of the regressed velocity parameters. / Ph. D.

Structural health monitoring (SHM) of aerospace components is a rapidly emerging field due in part to commercial and military transport vehicles remaining in operation beyond their designed life cycles. Damage detection strategies are sought that provide real-time information of the structure's integrity. One approach that has shown promise to accurately identify and quantify structural defects is based on guided ultrasonic wave (GUW) inspections, where low amplitude attenuation properties allow for long range and large specimen evaluation. One drawback to GUWs is that they exhibit a complex multi-modal response, such that each frequency corresponds to at least two excited modes, and thus intelligent signal processing is required for even the simplest of structures. In addition, GUWs are dispersive, whereby the wave velocity is a function of frequency, and the shape of the wave packet changes over the spatial domain, requiring sophisticated detection algorithms. Moreover, existing damage quantification measures are typically formulated as a comparison of the damaged to undamaged response, which has proven to be highly sensitive to changes in environment, and therefore often unreliable. As a response to these challenges inherent to GUW inspections, this research develops techniques to locate and estimate the severity of the damage. Specifically, a phase gradient based localization algorithm is introduced to identify the defect position independent of excitation frequency and damage size. Mode separation through the filtering technique is central in isolating and extracting single mode components, such as reflected, converted, and transmitted modes that may arise from the incident wave impacting a damage. Spatially-integrated single and multiple component mode coefficients are also formulated with the intent to better characterize wave reflections and conversions and to increase the signal to noise ratios. The techniques are applied to damaged isotropic finite element plate models and experimental data obtained from Scanning Laser Doppler Vibrometry tests. Numerical and experimental parametric studies are conducted, and the current strengths and weaknesses of the proposed approaches are discussed. In particular, limitations to the damage profiling characterization are shown for low ultrasonic frequency regimes, whereas the multiple component mode conversion coefficients provide excellent noise mitigation. Multiple component estimation relies on an experimental technique developed for the estimation of Lamb wave polarization using a 1D Laser Vibrometer. Lastly, suggestions are made to apply the techniques to more structurally complex geometries.

Automated Test Generation (ATG) has been successfully applied in many domains. For the modeling and simulation language Simulink, there has been research on developing tools for ATG with promising results. However, most tools developed as part of academic research and are not publicly available, or severely limited in their ability to be integrated into an industrial workflow. There are commercial ATG tools for Simulink, with Simulink Design Verifier (SLDV) as the de-facto standard tool. For this thesis, we perform an empirical comparison of manual tests to those generated by SLDV. For the comparison, we used 180 components from the propulsion control software developed by our industry partner. All except two components are compatible for test generation to some extent. The majority of components are partially compatible, requiring block replacement or stubbing. Approximation of floating-point numbers is the primary reason for block replacement, which can be performed automatically by SLDV. Two components were incompatible, and 14 required full stubbing of blocks. Using a pre-processing step, the generated tests achieve similar coverage as the manual tests. We performed a Mann–Whitney U test with the hypothesis that the generated tests achieve higher coverage than the manual tests. There are no statistically significant differences for either decision coverage (0.0719), or condition coverage (0.8357). However, for Modified Condition/Decision Coverage, the generated tests achieve higher coverage, and the difference is significant (0.0027). The limitations of ATG were explored by looking at the cases where the generated tests achieved lower coverage than the manual test. We found that the use of floating-point arithmetic and temporal logic increases the time required for test generation, and causes the analysis to hit the time limit. The test generation does not support all custom S-functions and perform stubbing of these blocks. This made the tool unable to reason about persistent storage. Configuration constants have limited support, which was the reason for the coverage difference in three cases. We have concluded that while much effort is required for custom tooling and initial setup, ATG can prove useful for early fault detection in an industrial workflow. ATG would prove especially useful in an automated continuous integration workflow for integration-level conformance testing.