Neural Network Based Direct Model Reference Adaptive Control Technique For Improving Tracking Performance in Nonlinear Systems.
| dc.contributor.advisor | Dereje, Shiferaw (PhD) | |
| dc.contributor.author | Alemie, Assefa | |
| dc.date.accessioned | 2019-11-11T06:44:08Z | |
| dc.date.accessioned | 2023-11-28T14:20:31Z | |
| dc.date.available | 2019-11-11T06:44:08Z | |
| dc.date.available | 2023-11-28T14:20:31Z | |
| dc.date.issued | 2019-07 | |
| dc.description.abstract | This thesis investigates the application of a neural network based model reference adaptive intelligent controller for controlling of the nonlinear systems. In this scheme, the intelligent supervisory loop is incorporated into the conventional model reference adaptive controller framework by utilizing an online growing multilayer back propagation neural network structure in parallel with it. The idea is to control the plant by minimizing the tracking error between the desired reference model and the nonlinear system using conventional model reference adaptive controller by estimating the adaptation law using a multilayer back propagation neural network. In the conventional model reference adaptive controller (MRAC) scheme, the controller is designed to realize the plant output converges to reference model output based on the plant, which is linear. This scheme is effective for controlling the linear plant with unknown parameters. However, using MRAC to control the nonlinear system in real time is difficult. The Neural Network is used to compensate the nonlinearity of the plant that is not taken into consideration in the conventional MRAC. The proposed neural network based model reference adaptive controller can significantly improve the system behavior and force the system to follow the reference model and minimize the error between the model and the plant output. Adaptive law using Lyapunov stability criteria for updating the controller parameters online has been formulated. The effectiveness of the proposed control scheme is verified by developing the simulation results for simple pendulum and Vander poll oscillation as a benchmark study in MATLAB/SIMULINK software. It is observed from the simulation results that the proposed neural network based Direct MRAC has 3.13sec rise time, 5.15sec settling time for 0.1rad disturbance and 3.12sec rise time, 5.21sec settling time for 0.2rad disturbance. Whereas, the conventional direct model reference adaptive control has 5.42sec rise time, 15.5sec settling time for 0.1rad disturbance and 5.01sec rise time, 15.52sec settling time for 0.2rad disturbance. It is shown that the proposed neural network based Direct MRAC has small rising time, steadystate error | en_US |
| dc.identifier.uri | http://etd.aau.edu.et/handle/12345678/20074 | |
| dc.language.iso | en_US | en_US |
| dc.publisher | Addis Ababa University | en_US |
| dc.subject | Model reference adaptive control (MRAC) | en_US |
| dc.subject | Artificial Neural Network (ANN) | en_US |
| dc.subject | Multilayer Backpropagation Neural Network | en_US |
| dc.title | Neural Network Based Direct Model Reference Adaptive Control Technique For Improving Tracking Performance in Nonlinear Systems. | en_US |
| dc.type | Thesis | en_US |