Implementation of adaptive control in power system

• Main Author: Mohamad Hazwan Mohamad Tolib.
• Other Authors: Wen Changyun
• Format: Final Year Project
• Language: English
• Published: 2013
• Subjects: DRNTU::Engineering::Electrical and electronic engineering::Control and instrumentation::Control engineering
• Online Access: http://hdl.handle.net/10356/53404
• Institution: Nanyang Technological University

Controlling a system is considered as one of the most successful ways to ensure that the output desired reaches stability as time progresses. Control engineers are tasked to design controllers based on either the specifications of the plant’s system, or based on the user-defined output. Classical controllers make use of plant’s parameters; in order to design such controllers, the designers need the exact parameters beforehand. Examples of classical controllers are P/PI/PD/PID controllers. The problem with regards to classical controllers is that these controllers cannot be determined if the plant’s parameters are unknown. In addition to that, they (classical controllers) are unable to adapt to changes within the plant. This is where adaptive control comes into place. Adaptive control basically means that the (adaptive) controllers can adapt to the changes in the plant’s parameters. An example of a good use of adaptive control is in the aerospace industry. Flight controllers adapt to the changes in mass of the airplane (decreasing mass of engine oil) while in mid air to keep the airplane in motion at constant altitude. Such changes in plant’s transfer functions are very common in many industries. Changes, as mentioned, include a change in mass, expansion of electrical cables due to heat build-up (which will result in a change in impedances), sudden climate resistance experience in vehicles, etc. Because of such changes that designers are unable to predict, this is why designers design adaptive control so as to adapt to these changes. Classical controllers are unable to adapt to these changes because the controllers are based on fixed parameters. Thus, adaptive control is implemented to adapt, not counter, such changes. There’s two ways to design adaptive control; 1) by knowing the actual parameters and 2) unknown plant parameters. When plant parameters are known, the designers can create the controllers based on the initial parameters. When there are changes in plant parameters, the controllers are able to adapt to such changes and can return output back to its original (before the change) based on the plant’s nominal values. When plant parameters are not given, it is slightly more difficult to determine the exact nominal values. So, the designer’s aim when creating such controllers is to make the adaptation approach to the true nominal value as close as possible. In this report, we will be focusing more on how effective adaptive control is on systems with unknown plant parameters. From there we investigate further onto the individual signals and give reasons why adaptive control is applicable in some situations while it is not for some. All simulations is executed using the Matlab/Simulink software.