Linköping Electronic Conference Proceedings
Not a member yet
1113 research outputs found
Sort by
Sequence Tagging in EFL Email Texts as Feedback for Language Learners
Get PDFWhen predicting scores for different aspects of a learner text, automated scoring algorithms usually cannot provide information about which part of text a score is referring to. We therefore propose a method to automatically segment learner texts as a way towards providing visual feedback. We train a neural sequence tagging model and use it to segment EFL email texts into functional segments. Our algorithm reaches a token-based accuracy of 90% when trained per prompt and between 83 and 87% in a cross-prompt scenario
Speech Technology to Support Phonics Learning for Kindergarten Children at Risk of Dyslexia
Get PDFWe present the AiRO learning environment for kindergarten children at risk of developing dyslexia. The AiRO frontend, easy to use for pupils down to 5 years old, introduces each spelling task with pictural and auditive cues. AiRO responds to spelling attempts with phonetic renderings (synthetic voice). Below, we introduce the didactic and technical principles behind AiRO before presenting our first experiment with 50 kindergarten pupils. Our subjects were pre- and post-tested on reading an spelling. After four weeks of AiRObased training the experimental group significantly out-performed the control group, suggesting that a new CALLbased pedagogical approach to prevent dyslexia for some children may be within reach
TEI and Git in ParlaMint: Collaborative Development of Language Resources
Get PDFThis paper discusses the encoding, validation and development of language resources of the completed ParlaMint I and on-going ParlaMint II CLARIN projects, which centre on the collaborative development of a large set of interoperable corpora of parliamentary proceedings. It focuses on the ParlaMint encoding and the GitHub development platform and the evaluation of their use by project partners. We introduce the use of TEI ODD for the encoding guidelines and validation schemas. We motivate and explain using Git to develop and maintain the encoding schemas, validation and conversion scripts and valid samples of the corpora. Apart from revision control, issues, and publishing documentation, the GitHub platform also supports integrating code execution with pull requests, which significantly automates the data submission process. The paper also presents the results of a survey on the use of TEI and Git in the ParlaMint projects among the project participants. Overall, participants were mostly positive about their experiencewith TEI and Git, although some difficulties were reported. Some partners also expressed doubts about whether the current scheme is flexible enough to support encoding the unique features of sometimes drastically different parliamentary systems. On the other hand, reactions to Git were very positive in terms of information and feedback received via GitHub Issues, effectiveness in the communication process, and plans to use Git in the future. However, there was less agreement on whether the requirements and workflows were adequately explained.The reported difficulties will serve as a basis for further Git workflow optimisation in ParlaMint
“We just did it!” – Female employees in Swedish sigint during the Second World War
Get PDFWhen the Swedish sigint agency, FRA, was formed in 1942, the civilian personnel consisted of 67 percent women. This paper explores the roles and duties performed by women at the FRA during the wartime years and their working conditions
Multirotor drone sizing and trajectory optimization within Modelon Impact
Get PDFThe design of multirotor drones often relies on optimizing its performance in terms of maximum speed requirements and hover time. This is well suited to undefined tasks. In the case of repetitive tasks, the drone trajectory can be added as a third degree of freedom. This paper focuses on the use of Modelon Impact and its dynamic optimization capabilities to reach a multirotor drone design and 1-D trajectory optimization. In comparison to other options investigated by the authors in a separate publication, Modelon Impact based optimization proved to be much simpler, more robust, and faster – for this use case
Development of a Modelica Model for the Texas A&M Smart and Connected Homes Testbed
Get PDFThe Texas A&M Smart and Connected Homes Testbed was developed to enable testing and fair comparison of different smart home technologies and grid-interactive capabilities. Having an accurate building model allows researchers to design and optimize smart home systems before implementing them into the experimental testbed. The Modelica Buildings library gives researchers the flexibility to prototype buildings and energy systems to apply to their research projects. This study develops a Modelica model using the Buildings library for the building envelope of the testbed homes using data from the home manufacturer. The A/C system model is developed from the device’s rated performance data. Each model goes through independent testing before implementation in the complete building model. In the future, real data from the lab homes will be used to tune these models to ensure accurate performance before a final model is made for testing
Object-Oriented Modelling of Flexible Cables based on Absolute Nodal Coordinate Formulation
Get PDFCable-pulley system consists of several segments of cables, winches, and pulleys, which is used in a wide range of engineering applications such as lifting equipment and pulley systems, however its dynamics simulation has been a tough issue in the Modelica community. The absolute nodal coordinate formulation (ANCF) uses global displacements and slopes at nodes to describe the geometry of the deformed body, which allows the derivation of constant mass matrices and zero-valued quadratic velocity dependent centrifugal and Coriolis forces, demonstrating its powerful capacity to model flexible multi-body systems for nearly two decades. This paper presents an object-oriented approach to model cable-pulley system, where flexible cables are discretized using ANCF cable elements. It is compatible with the Modelica Multibody Library by using a unified Frame interface and enables coupled analysis of cables and rigid bodies. The paper provides a rich set of application examples showing the ease and efficiency of the Modelica-based component drag-and-drop modelling way for modelling cable-pulley system
Application of the OpenModelica-Matlab Interface to Integrated Simulation and Successive Linearization Based Model Predictive Control
Get PDFThis paper presents the implementation of successive linearization based model predictive control (SLMPC) efforts through the interfacing of OpenModelica and Matlab using the OMMatlab tool. The dynamic system (here a chemical process) and the model predictive control (MPC) algorithm are implemented in OpenModelica and Matlab, respectively. The model linearization procedure is carried out through OMMatlab, which is highly optimized in terms of run-time by using a single executable file and adapting it at each sample time. Also, necessary theories for a continuous model discretization are discussed for both nonlinear Modelica and linearized continuous models. A procedure for constructing an Extended Kalman Filter (EKF) from a continuous Modelica model is also presented. The usability of the OpenModelica-Matlab interface for SLMPC is demonstrated by control of liquid levels in a tanks-in-series problem
A Library to Simulate Processes in the Factory Hall
Get PDFThe Modelica language is well suited to model systems with discrete and continuous dynamics. If one wants to model the flow of items through manufacturing steps such as preparation, mounting, or transport in the shop floor, this feature is crucial. The library ProcessSimulation can be used to model such processes. By default, it omits the technical details of the process steps, and focuses on the flow of material items through the process steps. In addition to that, a base model to calculate the energy consumption in the different manufacturing steps is provided. It can be enriched with technical details of the components. The library can be used for the calculation of (net) energy consumption, but also for task planning
Heat Exchanger Surrogates for a Vapor Compression System
Get PDFGiven the computationally intensive nature of heat exchanger simulators, utilizing a data-driven surrogate model for efficiently computing the heat exchanger outputs is desirable. This study focuses on developing integrated surrogate models of heat exchangers for a vapor compression system in Modelica. The surrogate models are designed to serve as steady-state equivalents based on an efficient physics-based model which was calibrated using reference data obtained from a more advanced simulation model. Subsequently, the calibrated model was employed to generate the training and testing data for the development of Gaussian Process (GP) and Multi-Layer Perceptron (MLP) surrogates. The obtained findings indicate that GPs exhibit high accuracy when applied to the heat exchanger's outputs with smooth behavior. GPs also demonstrate excellent data efficiency compared to MLPs. In cases where the GP struggles to model specific outputs effectively, MLPs are able to capture the more complex behavior. Moreover, hyperparameter optimization is employed to identify optimal MLP topologies. Finally, the fast and compact surrogate model was integrated into the Modelica/Dymola environment. This adaptation allowed the surrogate models to be directly combined with the physical model of the heat exchanger