Javascript must be enabled for the correct page display

A predictive model for Nafion-Based IPMC Soft Actuators

Langius, Ruben (2019) A predictive model for Nafion-Based IPMC Soft Actuators. Bachelor's Thesis, Artificial Intelligence.


Download (7MB) | Preview
[img] Text
Restricted to Registered users only

Download (105kB)


This research focuses on the development of a predictive model for a Nafion IPMC soft actuator. Nafion-117 is a synthetic polymer that is often researched for potential implementations in soft-robotics. For this research an actuator was made of Nafion-117, which was modelled with the help of a neural network. Before fabricating the final test samples, key variables that affected the actuator’s performance as well as the optimum build technique were defined by rigorous testing. The neural network was built based on a feed-forward model. To train this neural network, a data set consisting of 80.000 force measurements of 2 test samples was created. The network was trained and optimised on this data set, after which the resulting network was tested on a separate data set that was collected using a separate third test sample. After training, the validation set returned a root mean square error of 0.042. Prediction on the test sample resulted in a root mean square error of 0.034. Therefore, it can be concluded that this model generalises for Nafion IPMC actuators.

Item Type: Thesis (Bachelor's Thesis)
Supervisor name: Carloni, R.
Degree programme: Artificial Intelligence
Thesis type: Bachelor's Thesis
Language: English
Date Deposited: 20 Nov 2019
Last Modified: 11 Jun 2021 12:37

Actions (login required)

View Item View Item