This paper presents a novel approach to the analysis of the overtraining phenomenon in pattern classiiers. A nonlinearity measure N is introduced which relates the shape of the classiication function to the generalization capability of a classiier. Experiments using the k-nearest neighbour rule, a neural classiier and the quadratic classiier show that the introduced measure N can be used to stu...

Applying the recent developments of scale transition theory, we demonstrate a systematic approach to the problem of scaling up local scale interactions to regional scale dynamics with field data. Dynamics on larger spatial scales differ from the predictions of local dynamics alone because of an interaction between nonlinearity in population dynamics at the local scale and spatial variation in d...

Abstract. The dynamics of a membrane is a coupled system of a moving elastic surface and an incompressible membrane fluid. The difficulties in analyzing such a system include the nonlinearity of the curved space (geometric nonlinearity), the nonlinearity of the fluid dynamics (fluid nonlinearity), and the coupling to the surface incompressibility. In the two-dimensional case, the fluid vanishes...

This paper proposes a robust design scheme for the control of surge phenomenon appearing in a centrifugal compressor. It is known the surge instability in compressor might lead to reduce the operating efficiency and even damage the machinery while both of unstable compressor characteristics and compressor load torque are hard to exactly measure in the practical application. Based on the assumpt...

In this paper, an adaptive control strategy is proposed for the inverters of renewable energy source (RES) to simultaneously control the load voltage, grid current and the amount of instantaneous injected power to the grid in the presence of grid voltage distortions and nonlinearity of load current. In the proposed control strategy, the power quality of the local load can be settled based on th...