TY - CONF A1 - Weichert, Dorina A1 - Kister, Alexander A1 - Houben, Sebastian A1 - Link, Patrick A1 - Ernis, Gunar ED - Kiyavash, Negar ED - Mooij, Joris M. T1 - Robust Entropy Search for Safe Efficient Bayesian Optimization N2 - The practical use of Bayesian Optimization (BO) in engineering applications imposes special requirements: high sampling efficiency on the one hand and finding a robust solution on the other hand. We address the case of adversarial robustness, where all parameters are controllable during the optimization process, but a subset of them is uncontrollable or even adversely perturbed at the time of application. To this end, we develop an efficient information-based acquisition function that we call Robust Entropy Search (RES). We empirically demonstrate its benefits in experiments on synthetic and real-life data. The results show that RES reliably finds robust optima, outperforming state-of-the-art algorithms. T2 - UAI 2024 CY - Barcelona, Spain DA - 15.07.2024 KW - Bayesian Optimization KW - Gaussian process KW - Active learning PY - 2024 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-613523 UR - https://proceedings.mlr.press/v244/weichert24a.html SN - 2640-3498 VL - 244 SP - 3711 EP - 3729 PB - Proceedings of Machine Learning Research AN - OPUS4-61352 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Metz, Christian A1 - Franz, Philipp A1 - Fischer, C. A1 - Wachtendorf, Volker A1 - Maierhofer, Christiane T1 - Active thermography for quality assurance of 3D-printed polymer structures N2 - Additively manufactured test specimens made of polyamide 12 (PA 12) by Laser Sintering (LS) as well as of acrylnitril-butadien-styrol (ABS) by Fused Layer Modeling (FLM), were tested with active thermography. For this, two different excitation methods (flash and impulse excitation) were used and compared, regarding the suitability for the detection of constructed and imprinted defects. To increase the quality of the thermograms, data processing methods like thermal signal reconstruction (TSR) and Fourier-Transformation were applied. Furthermore, the long-term stability of the probes towards environmental stress, like UV-radiation, heat, water contact and frost is being investigated in the presented project with artificial weathering tests. T2 - 14th Quantitative InfraRed Thermography Conference CY - Berlin, Germany DA - 25.06.2018 KW - Additive manufacturing KW - Active thermography KW - Artificial weathering PY - 2018 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-453919 SP - Tu.3.A.2, 1 EP - 9 AN - OPUS4-45391 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Krankenhagen, Rainer A1 - Ziegler, Mathias A1 - Maierhofer, Christiane ED - Maldague, X. T1 - Systematic errors in the evaluation of uncorrected data from thermographic lock-in measurements N2 - Lock-in thermography (LT) is based on the correct evaluation of phase differences between the temperature oscillations at different surface regions of the object under test during periodic heating. Since the usual heating procedures contain a DC component, the actual heating pattern achieved is not harmonic. This causes systematic deviations when phase differences are determined by means of harmonic analysis, e.g. with FFT analysis. The resulting errors depend clearly on the ratio between DC and AC amplitude, which is demonstrated at simulated and experimentally recorded temperature transients. Further experimental LT data obtained by different oscillating energy inputs showed a variety of possible shapes of transients with different DC components. T2 - 14th QIRT Conference CY - Berlin, Germany DA - 26.06.2018 KW - NDT KW - Lock-in Thermography KW - FFT PY - 2018 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-453768 SP - 539 EP - 547 PB - QIRT Council AN - OPUS4-45376 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Talebi, Elnaz A1 - Korzen, Manfred A1 - Espinós, A. A1 - Hothan, Sascha ED - Albero, Vicente T1 - The effect of damage location on the performance of seismically damaged concrete filled steel tube columns at fire N2 - In this paper, a nonlinear three-dimensional finite element (FE) model was developed and validated to study the effect of seismic damage location on the response of concrete filled tube (CFT) columns at fire after earthquakes. Three analyses were conducted consecutively in the modelling, namely, cyclic, thermal and structural. Results of the cyclic loading analysis comprising residual deformations were applied as the Initial condition to the thermal-stress model, replicating the seismic performance of column. Following, a nonlinear sequentially coupled-thermal stress Analysis was carried out to investigate the fire response of CFT columns after the seismic event. Three damage scenarios were contemplated, considering any possible potential damages that could be generated by the earthquake loading on CFT columns. The accuracy of the proposed FE model was examined by comparing the numerical results with that of available tests on fire and cyclic loading. By means of the validated model, the performance of damaged CFT columns was then investigated under fire after earthquakes. The level of damage was assumed as a high damage level, presuming that the column reached 50% of ist lateral resistance while still maintaining its overall stability after the earthquake. The results were presented broadly, including the axial deformation history as well as the fire resistance time for CFT columns. To have a comprehensive insight on the influence of damage location in columns, the fire response of damaged specimens was compared with that of an intact one. T2 - 12th International Conference on Advances in Steel-Concrete Composite Structures (ASCCS 2018) CY - Valencia, Spain DA - 27.06. 2018 KW - Fire after earthquake KW - Concrete filled tube (CFT) column KW - Finite element model KW - Seismically damaged column KW - Damage location KW - Fire performance PY - 2018 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-457694 SN - 978-84-9048-601-6 DO - https://doi.org/10.4995/ASCCS2018.2018.8534 SP - 835 EP - 842 PB - Editorial Universitat Politècnica de València CY - València AN - OPUS4-45769 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Burgholzer, P. A1 - Berer, T. A1 - Ziegler, Mathias A1 - Thiel, Erik A1 - Ahmadi, Samim A1 - Gruber, J. A1 - Mayr, G. A1 - Hendorfer, G. T1 - Blind structured illumination as excitation for super-resolution photothermal radiometry N2 - Using an infrared camera for radiometric imaging allows the contactless temperature measurement of multiple surface pixels simultaneously. From the measured surface data, a sub-surface structure, embedded inside a sample or tissue, can be reconstructed and imaged when heated by an excitation light pulse. The main drawback in radiometric imaging is the degradation of the spatial resolution with increasing depth, which results in blurred images for deeper lying structures. We circumvent this degradation with blind structured illumination, combined with a non-linear joint sparsity reconstruction algorithm. The ground-breaking concept of super-resolution can be transferred from optics to thermographic imaging. T2 - Conference QIRT 2018 CY - Berlin, Germany DA - 25.06.2018 KW - Thermography KW - Laser thermography KW - Super resolution PY - 2018 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-454506 SN - 978-3-940283-94-8 DO - https://doi.org/10.1080/17686733.2019.1655247 SP - We.3.A.2, 1 EP - 7 AN - OPUS4-45450 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Maierhofer, Christiane A1 - Krankenhagen, Rainer A1 - Röllig, Mathias A1 - Heckel, Thomas A1 - Brackrock, Daniel A1 - Gaal, Mate T1 - Quantification of impact damages in CFRP and GFRP structures with thermography and ultrasonics N2 - The extent of damage caused by impacts in fibre reinforced composites depends on the energy of the impacts, on the velocity and the shape of the impacting body, on the material and structure of the composite and on the geometry of the structure. Here, mainly the thickness of the component is essential. The non-destructive evaluation of these damages can be carried out using both ultrasound and active thermography methods. A comparison of the detection sensitivity of these methods for the different damages is carried out in this paper depending on the fibre composite material used (CFRP and GFRP), the thickness of the material and the impact energy. The NDT methods used after the damage are supplemented by thermographic measurements with high temporal resolution, which were already recorded during the impact. T2 - 14th Quantitative InfraRed Thermography Conference CY - Berlin, Germany DA - 25.06.2018 KW - Active thermography KW - Passive thermography KW - Ultrasonics KW - CFRP KW - GFRP KW - Impact PY - 2018 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-454952 UR - http://www.qirt.org/archives/qirt2018/papers/126.pdf DO - https://doi.org/10.21611/qirt.2018.126 SP - 933 EP - 940 PB - DGZfP e. V. AN - OPUS4-45495 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Paul, Nathalie A1 - Kister, Alexander A1 - Schnellhardt, Thorben A1 - Fetz, Maximilian A1 - Hecker, Dirk A1 - Wirtz, Tim ED - Meo, Rosa ED - Silvestri, Fabrizio T1 - Reinforcement Learning for Segmented Manufacturing N2 - The manufacturing of large components is, in comparison to small components, cost intensive. This is due to the sheer size of the components and the limited scalability in number of produced items. To take advantage of the effects of small component production we segment the large components into smaller parts and schedule the production of these parts on regular-sized machine tools. We propose to apply and adapt recent developments in reinforcement learning in combination with heuristics to efficiently solve the resulting segmentation and assignment problem. In particular, we solve the assignment problem up to a factor of 8 faster and only a few percentages less accurate than a classic solver from operations research. T2 - European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases 2023 CY - Turin, Italy DA - 18.09.2023 KW - Reinforcement Learning KW - Assignment Problem KW - Large component manufacturing PY - 2025 DO - https://doi.org/10.1007/978-3-031-74640-6_38 VL - 1 IS - 1 SP - 470 EP - 485 PB - Springer Cham AN - OPUS4-63031 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Viefhues, Eva A1 - Döhler, M. A1 - Hille, Falk A1 - Mevel, L. T1 - Fault detection for linear parameter varying systems under changes in the process noise covariance N2 - Detecting changes in the eigenstructure of linear systems is a comprehensively investigated subject. In particular, change detection methods based on hypothesis testing using Gaussian residuals have been developed previously. In such residuals, a reference model is confronted to data from the current system. In this paper, linear output-only systems depending on a varying external physical parameter are considered. These systems are driven by process noise, whose covariance may also vary between measurements. To deal with the varying parameter, an interpolation approach is pursued, where a limited number of reference models – each estimated from data measured in a reference state – are interpolated to approximate an adequate reference model for the current parameter. The problem becomes more complex when the different points of interpolation correspond to different noise conditions. Then conflicts may arise between the detection of changes in the eigenstructure due to a fault and the detection of changes due to different noise conditions. For this case, a new change detection approach is developed based on the interpolation of the eigenstructure at the reference points. The resulting approach is capable of change detection when both the external physical parameter and the process noise conditions are varying. This approach is validated on a numerical simulation of a mechanical system. T2 - 21st IFAC World Congress CY - Online meeting DA - 13.07.2020 KW - Linear parameter varying systems KW - Fault detectionchanging KW - Process noise KW - Subspace-based residual KW - Model interpolation PY - 2020 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-524652 DO - https://doi.org/10.1016/j.ifacol.2020.12.868 SN - 2405-8963 VL - 53 IS - 2 SP - 13668 EP - 13673 PB - Elsevier AN - OPUS4-52465 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Küttenbaum, Stefan A1 - Taffe, A. A1 - Braml, T. A1 - Maack, Stefan ED - Alexander, M. G. ED - Beushausen, H. ED - Dehn, F. ED - Moyo, P. T1 - Reliability assessment of existing bridge constructions based on results of non-destructive testing N2 - The non-destructive testing methods available for civil engineering (NDT-CE) enable the measurements of quantitative parameters, which realistically describe the characteristics of existing buildings. In the past, methods for quality evaluation and concepts for validation expanded into NDT-CE to improve the objectivity of measured data. Thereby, a metrological foundation was developed to collect statistically sound and structurally relevant information about the inner construction of structures without destructive interventions. More recently, the demand for recalculations of structural safety was identified. This paper summarizes a basic research study on structural analyses of bridges in combination with NDT. The aim is to use measurement data of nondestructive testing methods as stochastic quantities in static calculations. Therefore, a methodical interface between the guide to the expression of uncertainty in measurement and probabilistic approximation procedures (e.g. FORM) has been proven to be suitable. The motivation is to relate the scientific approach of the structural analysis with real information coming from existing structures and not with those found in the literature. A case study about the probabilistic bending proof of a reinforced concrete bridge with statistically verified data from ultrasonic measurements shows that the measuring results fulfil the requirements concerning precision, trueness, objectivity and reliability. T2 - International Conference on Concrete Repair, Rehabilitation and Retrofitting (ICCRRR 2018) CY - Cape Town, South Africa DA - 19.11.2018 KW - NDT KW - Concrete KW - Probabilistic reassessment KW - Bridge PY - 2018 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-467898 DO - https://doi.org/10.1051/matecconf/201819906001 SN - 2261-236X VL - 199 SP - 06001, 1 EP - 9 PB - MATEC Web of Conferences AN - OPUS4-46789 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Maack, Stefan A1 - Villalobos, S. A1 - Scott, D. T1 - Validation of artificial defects for non-destructive testing measurements on a reference structure N2 - Non-destructive testing was established over the last decades as an important tool for assessing damages, material characterization and quality assurance in civil engineering. For example, Ground Penetrating Radar (GPR) can be used to scan large areas of concrete structures to determine the spatial position of the reinforcement. With the ultrasonic echo method, the thickness of concrete structures can be easily determined even if a high density of reinforcement is given. Various methods and processes have been developed for the validation of NDT procedures aiming at ensuring the quality of measurements in practical use. The Probability of Detection (POD) for example, is an available method to compare different technical devices with each other quantitatively regarding their performance. With this method, the best suited testing device for a specific inspection task under defined boundary conditions can be selected. By using the Guide to the Expression of Uncertainty in Measurement (GUM), it is possible to quantify the measurement uncertainty of an inspection procedure for a specific task. Another important aspect to improve the acceptance of Non-destructive testing methods is the development of reference specimens. Reference specimens serve for the calibration and further development of NDT methods under realistic conditions in different laboratories under the same conditions. A particular challenge here is the most realistic representation of a damage that can occur at building sites. Possible damages include for example horizontal and vertical cracks or honeycombs in concrete. Such a reference structure was built for the development of a new design of power plant constructions. Comparative studies on the manufacturing of realistic honeycombs and delaminations were carried out in advance on a test specimen. The results of this study are presented here. T2 - ICCRRR 2018 - Concrete Repair, Rehabilitation and Retrofitting CY - Cape Town, South Africa DA - 19.11.2018 KW - Zerstörungsfreie Prüfung KW - Nondestructive testing KW - Istzustandserfassung KW - Ultrasonic PY - 2018 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-464497 DO - https://doi.org/10.1051/matecconf/201819906006 SN - 2261-236X VL - 199 SP - 1 EP - 9 PB - EDP Sciences CY - Les Ulis AN - OPUS4-46449 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER -