H14
Integration of Analysis & Test 2

Back to overview

17:35
conference time (CEST, Berlin)
Approach to Automated Testing by Combining Simulations and Bench Tests Conducted on an Electrohydraulic Actuator
27/10/2021 17:35 conference time (CEST, Berlin)
Room: H
M. Dörr, M. Karam, F. Leitenberger, T. Gwosch, S. Matthiesen (Karlsruher Institut für Technologie (IPEK), DEU)
M. Dörr, M. Karam, F. Leitenberger, T. Gwosch, S. Matthiesen (Karlsruher Institut für Technologie (IPEK), DEU)
Analyzing the system behavior and performance of a system by bench testing and simulation is an essential part of product development. Bench tests are expensive and time-consuming. Simulations can be carried out more quickly and at lower costs than bench tests. However, it is often unclear how well the reality of an actuator on a test bench is represented with a simulation. To increase the simulation confidence, the results of a simulation are usually validated on a test bench. If the system behavior in the simulation deviates from the system behavior on the test bench, iterations must be performed and the simulation must be adjusted if necessary. The test planning, test execution, and test evaluation are often carried out separately for simulation and bench testing, with the results in each iteration being used for further test planning. This results in a high effort for the test planning, test execution, and test evaluation activities by the test engineer. An approach to connect and automate test planning, test execution, and test evaluation is automated testing. It is unclear how automated testing is suitable for both bench tests and simulations and whether it reduces the test engineer’s effort. Therefore, this paper aims to present an approach to automated testing which plans, executes, and evaluates both simulations and bench tests automatically to solve an optimization problem with reduced effort. Based on an evaluation of the bench test and simulation results, new bench tests or simulations are automatically planned and executed. The main purpose of the simulation is to generate knowledge through a broad analysis of the system parameters, while the main purpose of bench tests is to validate the individual simulation results. Furthermore, it is possible for the algorithm to automatically improve the simulation by changing individual previously defined parameters based on the test bench data. To illustrate the application of the approach, a case for optimization of a position controller of an electro-hydraulic actuator using particle swarm optimization is presented. The evaluation of the system behavior for the investigated electrohydraulic actuator is done via the position curve and the integral of time multiplied by square error. A comparison of the number of user interactions and simulations performed with the presented approach to automated testing compared to a classical approach with DOE shows a reduction of the test engineer's effort. The presented approach enables automated testing that plans, executes and evaluates both simulations and bench tests. Thus, simulation confidence can be increased compared to pure simulations by automated validation. In addition, the simulation can be improved based on test bench results.
testing, bench tests, simulation, test strategy, automated testing, test planning, genetic algorithm, simulation confidence, electrohydraulic actuator
17:55
conference time (CEST, Berlin)
Testing/Calculation Dialogue Pyramid for Rotomolded Structures: a Predictive Design Tool
27/10/2021 17:55 conference time (CEST, Berlin)
Room: H
E. Laine (ENSMA, FRA); J-C. Grandidier, J-C. Dupré (Institut Pprime, FRA); E. Maziers, (TOTAL Petrochimicals, BEL); S. Lewis (Vision Plastics Limited, NZL)
E. Laine (ENSMA, FRA); J-C. Grandidier, J-C. Dupré (Institut Pprime, FRA); E. Maziers, (TOTAL Petrochimicals, BEL); S. Lewis (Vision Plastics Limited, NZL)
In the current context of research, competitiveness, and technological innovation, numerical simulation assumes an increased role in the design, qualification, and certification of industrial products. However, the passage from the material sample, for the characterization of the behavior [1], to the design of the final product by numerical simulation is effective only by carrying out an intermediate step. This consists of a correlation between a test on a structure close to the one to be designed or a reference structure and the numerical simulation of this test. To develop reliable numerical tools, manufacturers generally rely on a pyramid protocol describing the tests on semi-structural and structural samples to design the full-scale assemblies. If this approach is common in the transport or building sector, in the field of rotomolded polymer tanks it must be deployed and the effectiveness of such a protocol must be demonstrated. More specifically, by transposing this approach to the world of rotomolded polymer structures, it is possible to build the pyramid of dialogue tests/calculations from the characterization specimen to the dimensioning of rotomolded structures by recalling the stake of the numerical model to predict the behavior at 20 or 50 years. The objective of this presentation is to evaluate the implementation of the test pyramid to finely apprehend the behavior of a rotomolded tank within the framework of the regulations (standard). This approach is deployed in an industrial context, where it is important to correlate the tests/simulations in a reasonable time that must be related to the development time of a structure. The behavioral models must allow fast calculations and give reliable indications. The implementation of the pyramid is presented and a comparison of simulations and optical measurements on a single scale structure allows to quantify the limits of the approach. [1] E. Lainé, C. Bouvy, J.-C. Grandidier, G. Vaes, Methodology of Accelerated Characterization for long-term creep prediction of polymer structures to ensure their service life, Polymer Testing 79 (2019).
Polymer, Rotomolded, characterization, creep test, long term, optical method, structure test, tank, design, numerical simulation,
18:15
conference time (CEST, Berlin)
Spacecraft Model Based System Testing – Correlation of Test & Simulation
27/10/2021 18:15 conference time (CEST, Berlin)
Room: H
P. Tremblay, J. Frachon (Maya HTT, CAN)
P. Tremblay, J. Frachon (Maya HTT, CAN)
In 2019, three Canadian satellites, developed by the Canadian Space Agency, were successfully launched into orbit. Using the example of Radarsat Constellation Mission (RCM) satellite, we will present the benefits of a seamless collaboration of simulation & test for complex space systems. How does the FE model of a RCM spacecraft represent reality? One way to find out is to perform modal testing and to compare, or correlate the Test and Simulation Results. This is particularly important in the space industry, as launch loads are derived from validated models of the launch vehicles and the spacecraft. A poor model means uncertain loads which increases program risk. This paper presents how modal simulation is used to prepare the test. Indeed, how can we apply simulation for more productive and realistic testing? Before the RCM modal test is performed, requirements must be defined, as sensors such that the required modes can be clearly identified. How many sensors & exciters are required, where are they located and in which directions do they point? Also, in order to visualize test mode shapes, the test engineers requires a wireframe mesh that connects the sensor and exciter nodes. Finally, test article configuration and boundary conditions have to be investigated. During the modal test campaign, simulation and test results did not correlate properly. Based on quantitative and qualitative tools (MAC, MODMAC and Orthogonality), test and simulation FRFs and mode shapes have been compared. Once correlation establishes the differences between both representations, the FE model has to be updated to more accurately represent reality. Based on an optimizer, which can handle large amounts of variables (physical and material properties); the finite element model has been updated. It was complicated by the test boundary conditions in which the spacecraft was bolted to a concrete floor. This paper presents attempts to account for the flexibility of the interface plate and for the effect of the floor and of the underlying soil in the updating process. This paper unpacks all the challenge of the modal simulation-test interaction. In addition, it presents a fully integrated workflow to reduce iteration-engineering loops, improve simulation-test interaction recently developed to simulate space system. Multi-discipline analyses have been performed on the updated FE model to validate the design against specific loading conditions (acoustic, vibro-acoustic, dynamics…).
Spacecraft, Model Based Testing, correlation, FE model update
18:35
conference time (CEST, Berlin)
Data Assimilation for Prediction Enhancement and Uncertainty Reduction in Process Simulation
27/10/2021 18:35 conference time (CEST, Berlin)
Room: H
O. Fernlund, A. Forghani, G. Fernlun (Convergent Manufacturing Technologies Inc, CAN)
O. Fernlund, A. Forghani, G. Fernlun (Convergent Manufacturing Technologies Inc, CAN)
Today, process simulation is commonly used for designing and troubleshooting manufacturing processes of composite parts. Physics-based process simulations include a complex set of mathematical models that are solved using numerical methods such as finite elements (FE). These models typically include a large set of parameters and require multiple boundary and initial conditions, many of which are uncertain to some degree. Depending on the amount of uncertainty in the model parameters, boundary and initial conditions, there can be a wide range of possible simulation outcomes. Simulation results can be validated against experimental data, but experimental data carry their own uncertainty (e.g. measurement precision, and calibration error to name a few) and are not always available at the location(s) of interest. In addition, standard approaches to reconciling discrepancies between simulation results and experimental data can be a tedious process of trial and error: making small adjustments to simulation inputs until the outputs are consistent with the observed data. This work demonstrates how a Bayesian framework for data assimilation can be used to systematically update model predictions based on experimental data to enhance prediction accuracy and reduce uncertainty. The updated predictions represent optimal pooling between the simulated and measured data streams as they are combined in proportion to their respective uncertainties. A composite laminate with thermocouples (TC’s) embedded through its center was cured in an autoclave, and the TC data were compared with the results of a probabilistic 1D thermochemical process simulation for the same laminate. Gaussian process regression (GPR) was used to combine model predictions (prior) and TC data (observations) to generate new predictions (posterior) with enhanced accuracy and reduced uncertainty. It is shown that data assimilation can significantly improve simulation results inside the part even in cases where prior model prediction uncertainty is high and TC data are only available at remote points outside the part.
Digital technologies, process simulation, data assimilation, Gaussian process regression, Bayesian statistics
×

[TITLE]

[LISTING

[ABSTRACT]

[DATE]

[ROOM]

[KEYWORDS]