This document contains 6 questions regarding control systems and state space representations. The questions cover topics like:
1) Providing two state space representations for the same pulse transfer function and explaining why their controllability and observability may differ.
2) Designing state feedback gain to place closed loop poles at given locations for a discrete time system.
3) Designing a state feedback controller to place closed loop poles at given locations and providing a block diagram.
4) Determining observer feedback matrix to place observer eigen values at given locations.
5) Designing a full order observer for a continuous time system with given system matrices.
6) Designing a full order observer for another continuous time system, with
This document contains 6 questions regarding control systems and state space representations. The questions cover topics like:
1) Providing two state space representations for the same pulse transfer function and explaining why their controllability and observability may differ.
2) Designing state feedback gain to place closed loop poles at given locations for a discrete time system.
3) Designing a state feedback controller to place closed loop poles at given locations and providing a block diagram.
4) Determining observer feedback matrix to place observer eigen values at given locations.
5) Designing a full order observer for a continuous time system with given system matrices.
6) Designing a full order observer for another continuous time system, with
Q1: Consider the following pulse-transfer function:
𝑌(𝑧) 𝑧 −1 (1 + 0.8𝑧 −1 ) = 𝑈(𝑧) 1 + 1.3𝑧 −1 + 0.4𝑧 −2 A state space representation for this system can be given by- 𝑥1 (𝑘 + 1) 0 1 𝑥 (𝑘) 0 [ ]=[ ] [ 1 ] + [ ] 𝑈(𝑘) 𝑥2 (𝑘 + 1) −0.4 −1.3 𝑥2 (𝑘) 2 𝑥 (𝑘) 𝑌(𝑘) = [0.8 1] [ 1 ] 𝑥2 (𝑘) A different state-space representation for the same system can be given by- 𝑥1 (𝑘 + 1) 0 −0.4 𝑥1 (𝑘) 0.8 [ ]=[ ][ ] + [ ] 𝑈(𝑘) 𝑥2 (𝑘 + 1) 1 −1.3 𝑥2 (𝑘) 1 𝑥1 (𝑘) 𝑌(𝑘) = [0 1] [ ] 𝑥2 (𝑘) Check the controllability and observability for both the representation. Explain what causes the apparent difference in the controllability and observability of the same system. Q2: Consider the system- 𝑥(𝑘 + 1) = 𝐺 𝑥(𝑘) + 𝐻 𝑈(𝑘) 0 1 0 Where, 𝐺 = [ ] 𝑎𝑛𝑑 𝐻 = [ ] −0.16 −1 1 The controller is given by 𝑈 = −𝐾𝑥. Determine the state feedback gain matrix K, such that the system will have closed-loop pole at: 𝑧 = 0.5 ± 𝑗0.5 Q3: A discrete time regulator system has the plant : 0 1 0 0 𝑥(𝑘 + 1) = [ 0 0 1 ] 𝑥(𝑘) + [ 0] 𝑈(𝑘) −4 −2 −1 1 1 1 Design a state feedback controller which will place the closed-loop poles at − 2 ± 𝑗 2 and 0. Give a block diagram of the control configuration. Q4: Consider the system defined by 𝑥(𝑘 + 1) = 𝐺 𝑥(𝑘) + 𝐻 𝑈(𝑘) 𝑦(𝑘) = 𝐶 𝑥(𝑘) 0 −0.16 0 Where, 𝐺 = [ ] , 𝐻 = [ ] 𝑎𝑛𝑑 𝐶 = [0 1] 1 −1 1 Determine the observer feedback matrix KC such that the desired eigen values for the observer matrix are : 𝜇 = 0.5 ± 𝑗0.5 Q5: Consider the system- 𝑥̇ = 𝐴𝑥 + 𝐵𝑢 𝑦 = 𝐶𝑥 0 20.6 0 Where, 𝐴 = [ ] , 𝐵 = [ ] 𝑎𝑛𝑑 𝐶 = [0 1] 1 0 1 Use the observed state feedback such that, 𝑈 = −𝑘𝑥̃ Design a full order observer. Assume that the desired eigen values of the observer matrix are -10 and -10. Q6: Consider the system defined as- 𝑥̇ = 𝐴𝑥 + 𝐵𝑢 𝑦 = 𝐶𝑥 0 1 0 0 Where, 𝐴 = [ 0 0 1] , 𝐵 = [0] 𝑎𝑛𝑑 𝐶 = [1 0 0] −5 −6 0 1 Design a full order state observer, assuming that the desired poles for observer are located at s=-10, -10 and -15.