Software Application for Adaptive Identification and Controller Tuning

This paper introduces a new software for adaptive identification and controller tuning, with the use of higher-order neural units and gradient descent based techniques (including backpropagation through time). The software allows the user to load real process data offline and to identify the plant or control loop as a whole. Furthermore the software experimentally investigates potentials for optimisation of the control loop response, via a non-linear adaptive state-feedback controller. The software is aimed as a quick tool for students, scholars and practitioners who wish to check potentials for optimisation of their control loop (utilising available process data and non-linear controller).