Search

Exascale computers are expected to have highly hierarchical architectures with nodes composed by multiple core processors (CPU; central processing unit) and accelerators (GPU; graphics processing unit). The different programming levels generate new difficult algorithm issues. In particular when solving extremely large linear systems, new programming paradigms of Krylov methods should be defined and evaluated with respect to modern state of the art of scientific methods. Iterative Krylov methods involve linear algebra operations such as dot product, norm, addition of vectors and sparse matrix–vector multiplication. These operations are computationally expensive for large size matrices. In this paper, we aim to focus on the best way to perform effectively these operations, in double precision, on GPU in order to make iterative Krylov methods more robust and therefore reduce the computing time. The performance of our algorithms is evaluated on several matrices arising from engineering problems. Numerical experiments illustrate the robustness and accuracy of our implementation compared to the existing libraries. We deal with different preconditioned Krylov methods: Conjugate Gradient for symmetric positive-definite matrices, and Generalized Conjugate Residual, Bi-Conjugate Gradient Conjugate Residual, transpose-free Quasi Minimal Residual, Stabilized BiConjugate Gradient and Stabilized BiConjugate Gradient (L) for the solution of sparse linear systems with non symmetric matrices. We consider and compare several sparse compressed formats, and propose a way to implement effectively Krylov methods on GPU and on multicore CPU. Finally, we give strategies to faster algorithms by auto-tuning the threading design, upon the problem characteristics and the hardware changes. As a conclusion, we propose and analyse hybrid sub-structuring methods that should pave the way to exascale hybrid methods. [ABSTRACT FROM PUBLISHER]

Abstract: Considerable compliance of the lengthy oil-well drill string combined with low tool inertia and stick–slip friction between the tool and the rock bed causes notable undesired torsional vibrations of the drill-string electrical drive. In order to attenuate the torsional vibrations and, thus, improve both the quality and the productivity of the drilling process, an automatically tuned active damping control strategy based on drill-string torque estimation is proposed in the paper. The core of the strategy includes a proportional-integral (PI) controller extended with estimator-based drill-string torque feedback loop. Furthermore, an appropriate algorithm that prevents drill-string back-spinning caused by the limited braking power of the power converter is presented. Finally, an auto-tuning algorithm is proposed, which is built around an adaptive Kalman filter-based estimator of drill-string drive natural frequencies. The drill-string control strategy is verified experimentally using a drill string hardware-in-the-loop setup under the laboratory conditions, as well as on an actual oil-drilling rig. [Copyright &y& Elsevier]

Abstract: In this paper a new auto-tuning algorithm for PI and PID controllers based on relay experiments is proposed to minimize the load disturbance integral error (IE) by maximizing the integral gain, subject to a desired phase margin, and a minimum required gain margin constraint. The main advantage of the proposed auto-tuning algorithm with respect to previous works is that it leads, for most of the processes, to PID tuning with close loop performance similar to PID designed using off-line numerical optimization. Moreover the algorithm is applicable to any linear model structure, including dead time and non-minimum phase systems. [Copyright &y& Elsevier]

The integrated radial basis function (IREF) network has been reported is an efficient algorithm to study the performance of brushless dc motors. However, such an algorithm cannot be implemented readily since it is difficult to lute-tune or even to find the undetermined coefficients in the integrated RBF network. In this paper, a novel auto-tuning algorithm that can effectively guarantee the automatic implementation of the integrated RBF network of a brushless dc motor is reported. [ABSTRACT FROM AUTHOR]