Lead Compensators

Definition

Consider a unity feedback system with reference signal r(t), output signal y(t), plant P(s) and controller C(s).

Unity feedback block diagram

A lead compensator controller C(s) is of the form:

C(s)=Kcs+zs+p

where p>z.

By setting z=αp and K=Kcα where 0<α<1, we can rewrite it in Bode form:

C(s)=Ksαp+1sp+1

Lead compensation approximates a PD controller, is used to stabilize a system and improve the transient response (i.e improve phase margin).

Design

Steps to design a lead compensator under given specifications:

  1. Choose K to meet a steady state error specification
  2. Find phase margin of KP(s) (either analytically or from the Bode plot)
  3. Find how much extra phase is required to meet the phase margin specification. Set ϕmax to be the required phase + an additional 10
  4. Find α=1sin(ϕmax)1+sin(ϕmax)
  5. Set ωmax to be ωcg of KP(s) and calculate p=ωmaxα and z=αωmax
  6. Check if the compensator achieves the specifications. If not, iterate.

Lag Compensators

Definition

Consider a unity feedback system with reference signal r(t), output signal y(t), plant P(s) and controller C(s).

Unity feedback block diagram

A lag compensator controller C(s) is of the form:

C(s)=Kcs+zs+p

where z>p.

By setting z=βp where β>1, we can rewrite it in Bode form:

C(s)=Kcβsβp+1sp+1

Lag compensation approximates a PI controller, is used to boost the DC gain.

Design

Steps to design a lead compensator under given specifications:

  1. Choose Kc to meet the phase margin specification (with 10 buffer) by moving ωcg to the left
  2. Find the low frequency gain of KcP(s), and determine how much extra gain β is needed for tracking specification
  3. Choose z=βp to be one decade below ωcg of KcP(s)
  4. Choose p=zβ
  5. Check if the compensator achieves the specifications. If not, iterate.