Showing 1–2 of 2 results for author: Freed, A

Multi-Axis and Multi-Vector Gradient Estimations: Using Multi-Sampled Complex Unit Vectors to Estimate Gradients of Real Functions

Authors: Ergun Akleman, Alan Freed

Abstract: In this preliminary study, we provide two methods for estimating the gradients of functions of real value. Both methods are built on derivative estimations that are calculated using the standard method or the Squire-Trapp method for any given direction. Gradients are computed as the average of derivatives in uniformly sampled directions. The first method uses a uniformly distributed set of axes th… ▽ More In this preliminary study, we provide two methods for estimating the gradients of functions of real value. Both methods are built on derivative estimations that are calculated using the standard method or the Squire-Trapp method for any given direction. Gradients are computed as the average of derivatives in uniformly sampled directions. The first method uses a uniformly distributed set of axes that consists of orthogonal unit vectors that span the space. The second method only uses a uniformly distributed set of unit vectors. Both methods essentially minimize the error through an average of estimations to cancel error terms. Both methods are essentially a conceptual generalization of the method used to estimate normal fractal surfaces. △ Less

Submitted 30 November, 2023; originally announced December 2023.

Comments: 10 pages

A Technical Note: Two-Step PECE Methods for Approximating Solutions To First- and Second-Order ODEs

Authors: Alan D. Freed

Abstract: Two-step predictor/corrector methods are provided to solve three classes of problems that present themselves as systems of ordinary differential equations (ODEs). In the first class, velocities are given from which displacements are to be solved. In the second class, velocities and accelerations are given from which displacements are to be solved. And in the third class, accelerations are given fr… ▽ More Two-step predictor/corrector methods are provided to solve three classes of problems that present themselves as systems of ordinary differential equations (ODEs). In the first class, velocities are given from which displacements are to be solved. In the second class, velocities and accelerations are given from which displacements are to be solved. And in the third class, accelerations are given from which velocities and displacements are to be solved. Two-step methods are not self starting, so compatible one-step methods are provided to take that first step with. An algorithm is presented for controlling the step size so that the local truncation error does not exceed a specified tolerance. △ Less

Submitted 7 July, 2017; originally announced July 2017.

Comments: 21 pages, 5 figures, 1 table, a supplementary document for a class I teach in numerical methods to mechanical engineers

MSC Class: 65L06