Chapter 2: State Space Definitions Flashcards
1
Q
Advantage of State Space Design (3)
A
- Reveals all internal poles/zeros
- Linear algebra can be used
- Supports design and analysis of feedback control systems
2
Q
Matrices of State Space Equations (4)
A
- A - System Matrix
- B - Input Matrix
- C - Output Matrix
- D - Feedforward Matrix
3
Q
State Space and Transfer Equation
- C.T.
- D.T.
A
- H(s) = C * inverse(sI-A) * B + D
2. H(z) = C * inverse(zI-A) * B + D
4
Q
Transformation of State Space Matrices where x’ = Tx (4)
A
- A’ = T * A * inverse(T)
- B’ = inverse(T) * B
- C’ = C * inverse(T)
- D’ = D
5
Q
Parallel Canonical Form Properties
A
- Poles of transfer function are eigenvalues of the system
2. H(s) = bn + c1/(s-p1) + … + cn/(s-pn)
6
Q
Cascade Canonical Properties
A
- Diagonal elements are eigenvalues of system matrix and poles of transfer function
- Series of 1st Order D.E. H(s) = H1(s) * H2(s) * … * Hn(s)
7
Q
Phase Variable Canonical Properties
A
- Used when there are no input derivatives.
- x1 = y, x2 = x1’, … , xn = xn-1’
- Simplifies control gain selection
8
Q
Controllable Canonical Properties
A
- Used to design optimal pole placement
- Y(s)/U(s) = P(s)/U(s) * Y(s)/P(s)
- Output of last integrator is x1 = p, xn is output to first integrator
9
Q
Observable Canonical Properties
A
- Estimation of state vector by observing input/output of system
- Observability canonical form is used if observable form doesn’t exist
- Obtain highest order on left side and integrate
10
Q
Static Equilibrium
- C.T. State
- D.T. State
- Output
A
- x’ = f(xe, u0) = 0
- x(k+1) = f(xe,u0) = xe
- ye = g(xe,u0)
11
Q
Linearization of Function
A
- Taylor Series Approximation
2. f(xe + Dx) = f(xe) + df/dx Dx + 1/2! d2f/dx2 Dx^2 + … where x = xe
12
Q
Small Signal Model
f(xe + D*x)
A
f(xe) + df/dx D*x where x = xe