Michel van Tooren

Learn More
Radial basis function neural networks (RBFNNs) are widely used in nonlinear function approximation. One of the challenges in RBFNN modeling is determining how to effectively optimize width parameters to improve approximation accuracy. To solve this problem, a width optimization method, concurrent subspace width optimization (CSWO), is proposed based on a(More)
The design of radial basis function widths of Radial Basis Function Neural Network (RBFNN) is thoroughly studied in this paper. Firstly, the influence of the widths on performance of RBFNN is illustrated with three simple function approximation experiments. Based on the conclusions drawn from the experiments, we find that two key factors including the(More)
With progress of space technology and increase of space mission demand, requirements for robustness and reliability of space systems are ever-increasing. For the whole space mission life cycle, the most important decisions are made in the conceptual design phase, so it is very crucial to take uncertainties into consideration in this initial phase to assure(More)
The ADCP (Advanced Design Concepts and Practice) series workshop is an important part of an International collaboration Innovation team project sponsored by Tsinghua University and another two grants including National with the research group of the German Informatics Society on "Graphics in design and Engineering-GI GRIB". The ADCP project is now three(More)
This paper proposes algorithms for mechanism synthesis and analysis, Index. These algorithms are used to transfer behaviors into structures, and to transfer structures into kinematic and dynamic behaviors of the structures. The algorithms are valid for the synthesis and analysis of different mechnisms. These algorithms can be used iteratively to produce(More)
In the current design process of complex products, such as aircraft, road vehicles, etc. (as well as their subsystems and related manufacturing hardware), there is a strong imbalance between the time absorbed by lengthy and repetitive activities and the limited time left for the investigation of multiple and innovative configurations, and what-if scenarios.(More)
Nowadays designers need a technology that enables them to virtually access their ideas, model the multidisciplinary aspect of a product, manipulate geometry and related knowledge, and investigate multiple design options and configurations. Time efficiency and confidence in the quality of results are two factors which must both be taken into account.(More)
  • 1