Michel Hogge

Space solar arrays and concentrators

Abstract This paper presents some research activities conducted at the Centre Spatial de Liege (CSL) in the field of space solar arrays and concentration. With the new generation of high efficiency solar cells, solar concentration brings new insights for future high power spacecrafts. A trade-off study is presented in this paper. Two different trough concentrators, and a linear Fresnel lens concentrator are compared to rigid arrays. Thermal and optical behaviors are included in the analysis. Several technical aspects are discussed: • Off-pointing with concentrators induces collection loss and illumination non uniformity, reducing the PV efficiency. • Concentrator deployment increases the mission risk. • Reflective trough concentrators are attractive and already proven. Coating is made of VDA (Aluminum). A comprehensive analysis of PV conversion increase with protected silver is presented. • Solar concentration increases the heat load on solar cells, while the conversion efficiency is significantly decreasing at warm temperatures. To conclude, this paper will point out the new trends and the key factors to be addressed for the next generation of solar generators.

Nonrigid registration and multimodality fusion for 3D image-guided neurosurgical planning and navigation

We present a status report on our work in nonrigid registration and multimodality fusion for neurosurgery. The new features are the ability to perform registration using heterogeneous brain models and to perform fusion in 3D. We describe the various elements of the system, the experiments performed, the results obtained, and the lessons learned.

Parameter Identification for Inverse Problems in Metal Forming Simulations

To improve the quality of numerical simulations results, accurate material behavior models are required. Moreover, more and more complex constitutive laws are being proposed to describe peculiar material properties. Parameter identification is an inverse problem taking place in material model development. It consists in evaluating the material parameters which exist in the chosen model, leading to the most accurate model, minimizing the difference between experimental results and Finite Element Method (FEM) simulations. Hence, the parameter identification problem can be formulated as an optimization problem. We propose, in this paper, to solve this optimization problem with eight optimization methods in order to compare their efficiency and robustness. The eight implemented methods come either from literature, such as conjugate gradient method, BFGS, Levenberg-Marquardt, etc., or from original developments, such as a modified GCMMA method and an optimization method combination technique. At last, an optimization method dedicated to parameter identification problem will be proposed.