Content Details | NUML Online Research Repository

Content Details

Back to Department Listing

Title An Innovative Neuroevolutionary Approach for Heart Beat Analysis with ANN

Author(s) Sabir Ali

Abstract The prominence of artificial neural networks (ANNs) is rising in a variety of applications. Most mathematical models have a form of differential equations (DEs), and recent research work has demonstrated that neural networks (NN) can be used to solve differential equations (DEs). In this thesis, we present a new neuroevolutionary approach called hybrid fractional particle swarm optimization (FO-PSO) to solve a differential equation (DE). Here we find an approximate solution to a 2nd order non-linear ordinary differential equation (ODE), known as the Van der Pol (VdP) heartbeat model (HBM), utilizing artificial neural networks (ANNs) with feed-forward (FF), and also examine the effectiveness of our technique and approach. Fractional order particle swarm optimization (FO-PSO) is a hybrid technique for the fractional order velocity of the particle swarm optimization algorithm. For this thesis, we considered two problems: One of the problems contains a forcing term, while the other does not. Each problem has two scenarios, and each scenario has four cases. These cases arise because of some variations in parameters. We make a comparison of our proposed hybrid FO-PSO–ASA technique’s results with the hybrid genetic algorithm with the interior point technique’s results. 100 independent runs have been performed. In terms of mean absolute deviation, root-mean-square error, and Nash–Sutcliffe efficiency, statistical analyses demonstrate its application, efficacy, and dependability.

Type Thesis/Dissertation

Faculty Engineering and Computer Science

Department Mathematics

Language English

Publication Date 2023-09-04

Subject Mathematics

Publisher

Contributor(s)

Format

Identifier

Source

Relation

Coverage

Rights