A Comparative Study of Predictive Models for Nafion-based IPMC Soft Actuators

Burawudi, Kenny, K (2020) A Comparative Study of Predictive Models for Nafion-based IPMC Soft Actuators. Bachelor's Thesis, Artificial Intelligence.

Preview	Text AI_BA_2020_Kenny_Burawudi.pdf Download (2MB) | Preview
	Text toestemming.pdf Restricted to Registered users only Download (105kB)

Abstract

: Ionic polymer-metal composites are electro-active polymers that, when stimulated by an electric field, convert electrical energy into mechanical energy. The focus of this research is an ionic polymer-metal composite soft actuator that has been realised by Nafion-117 metallised on both sides with platinum. Three models are developed for this study and their predictive ability is compared. The methods used to realise these models are the Multi-layer Perceptron, the curve fitting and a Long-Short Term Memory neural network. The models aim to predict the force with respect to time at different voltages (low to high applied electric fields) and displacements (how much the actuator bends). Their ability to generalise to unseen samples is also evaluated. The Multi-Layer Perceptron produces the best overall results with a root mean squared error of 0.241 mN on data from the unseen sample and computational time for prediction of 1.3 µs.

Item Type:	Thesis (Bachelor's Thesis)
Supervisor name:	Carloni, R.
Degree programme:	Artificial Intelligence
Thesis type:	Bachelor's Thesis
Language:	English
Date Deposited:	31 Aug 2020 07:24
Last Modified:	11 Jun 2021 12:00
URI:	https://fse.studenttheses.ub.rug.nl/id/eprint/23305

Actions (login required)

View Item