DocumentCode :
3588113
Title :
A novel approach to model design and tuning through automatic parameter screening and optimization theory and application to a helicopter flight simulator case-study
Author :
Hessel, Matteo ; Borgatelli, Francesco ; Ortalli, Fabio
Author_Institution :
Engineering of Computing Systems, Politecnico di Milano, Piazza Leonardo Da Vinci 32, Italy
fYear :
2014
Firstpage :
24
Lastpage :
35
Abstract :
The aim of this paper is to describe a novel methodology for model-design and tuning in computer simulations, based on automatic parameter screening and optimization. Simulation requires three steps: mathematical modelling, numerical solution, and tuning of the model´s parameters. We address Tuning because, at the state-of-the-art, the development of life-critical simulations requires months to appropriately tune the model. Our methodology can be split in Screening (identification of the relevant parameters to simulate a system) and Optimization (search of optimal values for those parameters). All techniques are fully general, because they leverage ideas from Machine-Learning and Optimization Theory to achieve their goals without directly analysing the simulator´s mathematical model. Concerning screening, we show how Machine-Learning algorithms, based on Neural Networks and Logistic Regression, can be used for ranking the parameters according to their relevance. Concerning optimization, we describe two algorithms: an adaptive hill-climbing procedure and a novel strategy, specific for model tuning, called sequential masking. Eventually, we show the performances achieved and the impact on the time and effort required for tuning a helicopter flight-simulator, proving that the proposed techniques can significantly speed-up the process.
Keywords :
Computational modeling; Databases; Logistics; Mathematical model; Numerical models; Optimization; Tuning; Adaptive Hill-Climbing; Machine-Learning; Model Tuning; Optimization; Screening; Sequential Masking;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Simulation and Modeling Methodologies, Technologies and Applications (SIMULTECH), 2014 International Conference on
Type :
conf
Filename :
7094996
Link To Document :
بازگشت