Asymptotic performances of a constructive algorithm

We present a numerical study of a neural tree learning algorithm, the “triolearning≓ strategy. We study the behaviour of the algorithm as a function of the size of the training set. The results show that a limited number of examples can be used to estimate both the network performance and the network complexity that would result from running the algorithm on a large data set.     

Numerical simulation of the MHD equations by a kinetic-type method

We introduce a flux-splitting formula for the approximation of the ideal MHD equations in conservative form. The Faraday equation is considered as the average of an abstract kinetic equation, giving a flux-splitting formula. For the other part of the equations, we generalize formally the classical half-Maxwellian flux-splitting of the Euler equations. Numerical results on MHD shock tube problems are displayed.     

Sensorless fault tolerant control for induction motors

In this paper, a sensorless fault tolerant controller for induction motors is developed. In the proposed approach, a robust controller based on backstepping strategy is designed in order to compensate for both the load torque disturbance and the rotor resistance variation caused by the broken rotor bars faults. The proposed approach needs neither fault detection and isolation schemes nor controller re-design. Moreover, to avoid the use of speed and flux sensors, a second order sliding mode observer is introduced to estimate the flux and the speed. The observer converges in a finite time and leads to good estimates of the flux and the speed even in the presence of the rotor resistance variation and the load torque disturbance. Since the observer converges in the finite time, the stability of the closed-loop system (controller with observer) is shown in two steps. First, the boundedness of the closed-loop system trajectories before the convergence of the observer is proved. Second, the convergence of the closed-loop system trajectories is proved after the convergence of the observer. To highlight the efficiency and applicability of the proposed control scheme, simulation and experimental results are conducted for a 1.5 kW induction motor.     

Finite Volume Time Domain modelling of microwave breakdown and plasma formation in a metallic aperture

The modelling of plasma formation during microwave breakdown is a difficult task because of the strong non-linear coupling between Maxwell?s equations and plasma equations, and of the large plasma density gradients that form during breakdown. An original Finite Volume Time Domain (FVTD) method has been developed to solve Maxwell?s equations coupled with a simplified fluid plasma model and is described in this paper. This method is illustrated with the study of the shielding of a metallic aperture by the plasma generated by an incident high power electromagnetic wave. Typical results obtained with the FVTD method for this shielding problem are shown.     

三维分子结构检索系统的结构索引与匹配

介绍了三维分子结构检索系统的结构索引与匹配,应用位筛成功地实现了三维结构数据的索引,可以快速得到候选结构,然后使用原子－原子匹配算法获得精确的命中结构。     

Region-Based Memory Management

This paper describes a memory management discipline for programs that perform dynamic memory allocation and de-allocation. At runtime, all values are put intoregions. The store consists of a stack of regions. All points of region allocation and de-allocation are inferred automatically, using a type and effect based program analysis. The scheme does not assume the presence of a garbage collector. The scheme was first presented in 1994 (M. Tofte and J.-P. Talpin,in“Proceedings of the 21st ACM SIGPLAN–SIGACT Symposium on Principles of Programming Languages,” pp. 188–201); subsequently, it has been tested in The ML Kit with Regions, a region-based, garbage-collection free implementation of the Standard ML Core language, which includes recursive datatypes, higher-order functions and updatable references L. Birkedal, M. Tofte, and M. Vejlstrup, (1996),in“Proceedings of the 23 rd ACM SIGPLAN–SIGACT Symposium on Principles of Programming Languages,” pp. 171–183. This paper defines a region-based dynamic semantics for a skeletal programming language extracted from Standard ML. We present the inference system which specifies where regions can be allocated and de-allocated and a detailed proof that the system is sound with respect to a standard semantics. We conclude by giving some advice on how to write programs that run well on a stack of regions, based on practical experience with the ML Kit.     

Optimisation de ľefficacité de couplage entre un guide ďonde confiné et une fibre optique monomode par ľintermédiaire ďune lentille boule

Theoretical coupling efficiency between a channel waveguide and a single-mode optical fibre using a ball lens is optimized by varying the parameters of the optical system such as the relative position of the components, the refractive index and the size of the lens. A specific software has been used for the optimization. Numerical calculations have been done under the Gaussian approximation of the fields distributions. The losses due to waveguides misalignments with respect to their ideal position have been evaluated. All these theoretical results are compared with those obtained from buttcoupling.