الفهرس | Only 14 pages are availabe for public view |
Abstract Helicopter is one of the complicated systems. For studying the dynamics of a complex system like helicopter and test different control theories to control the position of the helicopter (yaw and pitch angle), a simple model which retain the main characteristics and complexity of helicopter is used, and this model is called: Twin Rotor Multi-Input Multi-Output system (TRMS). In this thesis, a comprehensive study for the TRMS model is achieved starting with identifying all the TRMS components and subsystems, using a black-box system identification technique. The identification process starts with identifying the rotors tacho generators constants, then identifying the rotors coupled with propeller input voltage-speed relationship, then rotors coupled with propeller speed-thrust force relationship is estimated using a special test rig. Then, all the identified parts are aggregated by the aid of physical relationships to generate a complete and accurate model for the TRMS. After completing the identification process, different types of controllers are suggested to control the TRMS, starting with classical PID controller tuned by the standard Ziegler Nichols tuning rules, then an algorithm is written for tuning and selecting the optimum gains for the PID controller based on particle swarm optimization (PSO) technique. The output response from the two controllers is compared, then a fuzzy-like PID controller is built based on the controller with the best response. At the end, a fractional-order PID is designed, the output response from all the controllers is compared to deduce the controller which generates the optimum output response. After completing this study, it appears that the fractional order PID controller generates the optimum system response for the twin-rotor and improves the system stability against external disturbance. In pitch direction, the maximum overshoot has been set to 0.01546 radians (1.546% increase compared to classical PID) with rising time equals to 1.319 seconds (98.13% decrease compared to classical PID), and settling time equals to 3.395 seconds (96.91% decrease compared to classical PID). In yaw direction, the maximum overshoot has been decreased to 0.03646 radians (90.45% decrease compared to classical PID) and the rise time to 4.028 seconds (31% increase compared to classical PID), and the settling time to 12.246 seconds (46.3% decrease compared to classical PID). |