A fuzzy logic based dynamic wave model inversion algorithm for canal regulation

A fuzzy logic based centralized control algorithm for irrigation canals is presented. Purpose of the algorithm is to control downstream discharge and water level of pools in the canal, by adjusting discharge release from the upstream end and gates settings. The algorithm is based on the dynamic wave model (Saint-Venant equations) inversion in space, wherein the momentum equation is replaced by a fuzzy rule based model, while retaining the continuity equation in its complete form. The fuzzy rule based model is developed on fuzzification of a new mathematical model for wave velocity, the derivational details of which are given. The advantages of the fuzzy control algorithm, over other conventional control algorithms, are described. It is transparent and intuitive, and no linearizations of the governing equations are involved. Timing of the algorithm and method of computation are explained. It is shown that the tuning is easy and the computations are straightforward. The algorithm provides stable, realistic and robust outputs. The disadvantage of the algorithm is reduced precision in its outputs due to the approximation inherent in the fuzzy logic. Feed back control logic is adopted to eliminate error caused by the system disturbances as well as error caused by the reduced precision in the outputs. The algorithm is tested by applying it to water level control problem in a fictitious canal with a single pool and also in a real canal with a series of pools. It is found that results obtained from the algorithm are comparable to those obtained from conventional control algorithms.

An FETI-preconditioned conjuerate gradient method for large-scale stochastic finite element problems

In the spectral stochastic finite element method for analyzing an uncertain system. the uncertainty is represented by a set of random variables, and a quantity of Interest such as the system response is considered as a function of these random variables Consequently, the underlying Galerkin projection yields a block system of deterministic equations where the blocks are sparse but coupled. The solution of this algebraic system of equations becomes rapidly challenging when the size of the physical system and/or the level of uncertainty is increased This paper addresses this challenge by presenting a preconditioned conjugate gradient method for such block systems where the preconditioning step is based on the dual-primal finite element tearing and interconnecting method equipped with a Krylov subspace reusage technique for accelerating the iterative solution of systems with multiple and repeated right-hand sides. Preliminary performance results on a Linux Cluster suggest that the proposed Solution method is numerically scalable and demonstrate its potential for making the uncertainty quantification Of realistic systems tractable.

Magnetically tunable nonlinear current-voltage characteristics in oxygen deficient manganite perovskites

(La0.667Ca0.333Mn1-xMO3-delta)-O-x (M = Mg, Li or Re) exhibit insulating behaviour and nonlinear current-voltage (J-E) relationship with voltage-limiting characteristics at temperatures below the ferromagnetic transition (T-c). The high current region is set in at field strengths <60 V/cm. Nonlinearity exponent, alpha in the relation J = kE(alpha) increases inversely with temperature. In presence of an external magnetic field, the J-E curves show higher current density at lower field strengths. Microstructural studies indicate that there is no segregation of secondary phases in the grain boundary regions. There is remarkable changes in p(T) as well as J-E curves with the grain size. Annealing studies in lower p(O2) atmospheres indicate that there is significant out-diffusion of oxygen ions through the grain boundary layer (GBL) regions creating oxygen vacancies in the GBL regions. The concentration of Mn4+ ions is lowered at the GBL due to oxygen vacancies, reducing the probability of hopping and resulting in insulating behaviour. Therefore an insulating barrier is introduced between two conducting grains and the carrier motion between the grains is inhibited. Thus below T-c, where sufficient increase in resistivity is observed the conduction may be arising as a result of spin dependent tunneling across the barrier. External electric field lowers the barrier height and establishes carrier transport across the barrier. Above certain field strength, barrier height diminishes significantly and thereby allowing large number of carriers for conduction, giving rise to highly nonlinear conductivity. (C) 2002 Elsevier Science B.V. All rights reserved.