Control Systems Notes

1. Introduction to Control Systems

• System: A combination of components that act together to perform a

certain objective.

• Control System: A system that regulates or directs the behavior of other

systems.

• Open-Loop Control System: Output has no effect on the control action.

◦ Simple, less expensive, but less accurate and no compensation for

disturbances. (e.g., toaster, washing machine timer).

• Closed-Loop Control System (Feedback Control System): Output

affects the control action.

◦ Uses feedback to compare the output with the desired input (reference).

◦ More accurate, reduced sensitivity to disturbances, improved transient

response, but more complex and risk of instability. (e.g., thermostat,
cruise control).

• Components of a Closed-Loop System:

◦ Reference Input (R(s) or r(t)): Desired output.

◦ Controller: Generates the control signal.

◦ Actuator: Device that implements the control signal (e.g., motor).

◦ Plant/Process: The system to be controlled.

◦ Output (C(s) or c(t)): Actual output.

◦ Sensor/Transducer: Measures the output.

◦ Feedback Element: Converts output to appropriate form for comparison.

◦ Error Signal (E(s) or e(t)): Difference between reference and feedback

signal.

• Types of Control Systems:

◦ Linear vs. Non-linear: Based on the linearity of governing equations.

◦ Time-Invariant vs. Time-Variant: Parameters constant or changing

with time.
◦ Continuous-Time vs. Discrete-Time: Based on signal nature.

◦ SISO (Single Input Single Output) vs. MIMO (Multiple Input

Multiple Output).

2. System Modeling

• Mathematical Models: Represent physical systems using equations.

◦ Differential Equations: Describe continuous-time systems.

◦ Difference Equations: Describe discrete-time systems.

• Laplace Transform: Used to convert differential equations into algebraic

equations in the s-domain.

◦ Allows easier analysis of LTI systems.

• Transfer Function G(s): Ratio of the Laplace transform of the output to

the Laplace transform of the input, assuming zero initial conditions.

◦ G(s) = Y(s) / X(s).

◦ Characterizes LTI systems.

• Poles and Zeros:

◦ Poles: Roots of the denominator of G(s). Indicate system's natural

frequencies and influence stability.

◦ Zeros: Roots of the numerator of G(s). Influence system response.

• Block Diagram Reduction: Techniques to simplify complex block

diagrams into a single transfer function.

◦ Series, Parallel, Feedback connections.

◦ Forward path gain, Loop gain.

• Signal Flow Graphs (SFG): Alternative to block diagrams for

representing systems.

◦ Nodes (variables), Branches (transfer functions).

◦ Mason's Gain Formula: T = (1/Δ) Σ P_k Δ_k.

P_k: k-th forward path gain.

Δ: Determinant of the graph (1 - Σ(individual loop gains) + Σ(product of

two non-touching loop gains) - ...).
 Δ_k: Δ for the part of the graph not touching the k-th forward path.

• State-Space Representation: A set of first-order differential (or

difference) equations.

◦ x'(t) = Ax(t) + Bu(t) (State Equation)

◦ y(t) = Cx(t) + Du(t) (Output Equation)

◦ x(t): state vector, u(t): input vector, y(t): output vector.

◦ A: system matrix, B: input matrix, C: output matrix, D: direct transmission

matrix.

◦ Provides a complete description of system behavior, including internal

states.

◦ Transfer function from state-space: G(s) = C(sI - A)⁻¹B + D.

3. Time-Domain Analysis

• Test Signals: Step, Ramp, Impulse, Parabolic. Used to evaluate system

response.

• Transient Response: Behavior of the system as it moves from its initial

state to a new steady state.

◦ First-Order Systems: Characterized by a single time constant (τ).

Response: x(t) = x_f + (x_0 - x_f)e^(-t/τ).

◦ Second-Order Systems: Characterized by undamped natural frequency

(ω_n) and damping ratio (ζ).

Underdamped (0 < ζ < 1): Oscillatory, decaying response.

Rise Time (t_r): Time to go from 10% to 90% of final value.

Peak Time (t_p): Time to reach the first peak. t_p = π / (ω_n√(1-ζ²)).

Percent Overshoot (%OS): ((C_peak - C_final) / C_final) * 100%. %OS =

e^(-ζπ / √(1-ζ²)) * 100%.

Settling Time (t_s): Time to settle within a certain percentage (e.g., 2% or

5%) of the final value. t_s ≈ 4/(ζω_n) for 2%.

Critically Damped (ζ = 1): Fastest response without oscillation.

Overdamped (ζ > 1): Slow response without oscillation.

• Steady-State Response: Behavior of the system after the transient has

died out.
 • Steady-State Error (e_ss): Difference between the desired output and
the actual output as time approaches infinity.

◦ For a unity feedback system, E(s) = R(s) / (1 + G(s)H(s)).

◦ e_ss = lim(s→0) sE(s).

◦ Type of System: Number of pure integrators (poles at s=0) in the open-

loop transfer function G(s)H(s).

Type 0: Constant position error for step input, infinite error for
ramp/parabolic.

Type 1: Zero position error, constant velocity error for ramp, infinite error
for parabolic.

Type 2: Zero position/velocity error, constant acceleration error for

parabolic.

◦ Static Error Coefficients: K_p (position), K_v (velocity), K_a

(acceleration).

K_p = lim(s→0) G(s)H(s)

K_v = lim(s→0) sG(s)H(s)

K_a = lim(s→0) s²G(s)H(s)

4. Stability Analysis

• Definition: A system is stable if its output remains bounded for every

bounded input (BIBO stability). For LTI systems, this means all poles of the
closed-loop transfer function must be in the left-half of the s-plane.

• Routh-Hurwitz Criterion: Determines the number of poles in the right-

half s-plane from the coefficients of the characteristic equation.

◦ No sign changes in the first column of the Routh array implies no RHP
poles.

◦ Special cases: Zero in first column, entire row of zeros.

• Root Locus: Graphical method that shows the locations of the closed-
loop poles as a system parameter (typically gain K) is varied from 0 to ∞.

◦ Rules for Construction:

Number of branches = number of poles.

Symmetry about the real axis.

 Starts at open-loop poles, ends at open-loop zeros (or ∞).

Real axis segments: To the left of an odd number of poles+zeros.

Asymptotes: Angle = (2k+1)180° / (P-Z), Centroid = (Σpoles - Σzeros) / (P-

Z).

Breakaway/Break-in points: dK/ds = 0 (where K = -1/G(s)H(s)).

Angle of departure/arrival from complex poles/zeros.

Intersections with jω-axis (Routh-Hurwitz or substitute s=jω).

• Frequency Response Analysis:

◦ Bode Plots: Magnitude (dB) and Phase (degrees) vs. log(ω).

Gain Margin (GM): Amount of gain change at phase crossover frequency

until instability.

Phase Margin (PM): Amount of phase change at gain crossover frequency

until instability.

GM > 0 dB and PM > 0° for stability.

◦ Nyquist Plot: Polar plot of G(jω)H(jω) for ω from 0 to ∞.

Nyquist Stability Criterion: Z = N + P (Z = closed-loop RHP poles, N =

number of encirclements of (-1,0), P = open-loop RHP poles). For stability,
Z = 0.

Focus on (-1,0) point (critical point).

◦ Gain Crossover Frequency (ω_gc): Frequency where |G(jω)H(jω)| = 1

(0 dB).

◦ Phase Crossover Frequency (ω_pc): Frequency where ∠G(jω)H(jω) = -

180°.

5. Control System Design

• Controllers/Compensators: Used to modify system performance

(stability, transient response, steady-state error).

• Proportional (P) Controller: K_p.

◦ Increases speed, reduces steady-state error (but doesn't eliminate for

Type 0).

◦ Increases overshoot, may make system less stable.

 • Integral (I) Controller: K_i / s.

◦ Eliminates steady-state error for step inputs (by adding a pole at origin,
increasing system type).

◦ Slows down response, can increase overshoot and decrease stability.

• Derivative (D) Controller: K_d * s.

◦ Improves transient response, increases damping, reduces overshoot,

speeds up response.

◦ Sensitive to noise, not good for steady-state error.

• PID Controller: Proportional-Integral-Derivative (K_p + K_i/s + K_d s).

◦ Combines benefits of P, I, D. Most widely used controller.

◦ Tunes for optimal performance:

P: Adjust rise time, reduce steady-state error.

I: Eliminate steady-state error.

D: Reduce overshoot, improve settling time.

• Lead Compensator: Adds a zero closer to the origin than a pole.

◦ Increases phase margin, improves stability, speeds up transient response.

◦ Used to reshape Root Locus or improve phase margin in Bode.

• Lag Compensator: Adds a pole closer to the origin than a zero.

◦ Improves steady-state error (increases gain at low frequencies).

◦ Decreases bandwidth, slows down transient response, reduces phase

margin.

• Lag-Lead Compensator: Combines benefits of both.

◦ Improves both steady-state error and transient response/stability.

6. Digital Control Systems (Brief Overview)

• Sampling: Converts continuous signals to discrete.

• Z-Transform: Used for analysis of discrete-time systems.

• Discrete Controllers: Digital versions of P, I, D.

 • Mapping from S-plane to Z-plane: s = (1/T) ln(z).

◦ Left-half s-plane maps to inside unit circle in z-plane (stability).

◦ Imaginary axis in s-plane maps to unit circle in z-plane.