MENG 366: Automatic Control

Term Project

Students, in groups of no more than five, are to model, analyze, and control a physical system using the tools and techniques learned in this class. It is expected that students will analyze their system using both analytical techniques as well as numerical methods.

The goal of this project is to illustrate the modeling, analysis, control and simulation techniques learned in this class on a physical system. A significant component of the project is to model and analyze models of any physical system and then to build the suitable control action. In this way, students can be exposed to the influence of control action on the system behavior.

Each project must allow for the following:

1. The model must admit a closed-form solution (i.e., it must be linear, with constant coefficients).
2. The control action must be considered.
3. You must present both analytical and numerical results.

Each report will be organized as follows:

• Introduction: what are you doing and what information will you obtain;
• Modeling: identify appropriate modeling assumptions and develop the equations of motion which correspond to the different sets of assumptions;
• Linear Analysis: linearize the above models about some operating point and analyze the resulting behavior predicted by each model;
• Simulation: use MATLAB and/or SIMULINK to verify the analytical predictions of both the qualitative and quantitative behavior of the system with and with out control system.
• Conclusions: model extensions, including the validity of the constitutive assumptions, discussion of results, effect of modeling assumptions, effect of choosing a specific control action and varying the control gain etc.

The idea is for you to figure out how your system is going to behave. For example, what happens as the parameters vary, what happens when the control gain is varied, how will the predicted dynamics vary with modeling assumptions, etc. You will employ both analytical techniques (modeling and solving the equations in closed-form) as well as numerical methods (MATLAB and/or SIMULINK) to get the system response.