Scribd
Upload
5 views3 pages

Control

Uploaded by

Krishna kumar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
5 views3 pages

Control

Uploaded by

Krishna kumar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

1 Theoretical Framework

1.1 Introduction to Control Systems and Disturbances


A control system is a set of devices or algorithms designed to regulate the behavior of
other devices or systems to achieve a desired output. Control systems maintain the system
output to track a reference input despite the presence of disturbances or uncertainties.
Control systems are broadly classified into:

• Open-loop systems: Operate without feedback, do not correct deviations.

• Closed-loop (feedback) systems: Continuously monitor output and adjust in-


put to minimize error.

Disturbances are unwanted inputs affecting system performance. They can be:

• External: Environmental effects, load changes, supply variations.

• Internal: Sensor noise, component nonlinearities, parameter variations.

• Constant vs. Time-varying: Constant disturbances can be addressed by integral


control; time-varying require advanced methods.

Effective disturbance rejection improves accuracy, stability, efficiency, and system


longevity.

1.2 Classical PID Control


The Proportional-Integral-Derivative (PID) controller is widely used for its simplicity and
effectiveness. It computes the control signal as:
Z t
de(t)
u(t) = Kp e(t) + Ki e(τ )dτ + Kd (1)
0 dt
where e(t) = r(t) − y(t) is the error between reference and output.
In the Laplace domain, the PID transfer function is:
Ki
CPID (s) = Kp + + Kd s (2)
s
Limitations:

• Integral action removes steady-state error for constant disturbances but reacts
slowly to time-varying disturbances.

• Derivative action amplifies high-frequency noise.

• Tuning is challenging to balance tracking, stability, and disturbance rejection.

• PID lacks internal model or predictive disturbance compensation.

1
1.3 Second-Order System Model
Many physical systems are approximated by a second-order transfer function:

ωn2
Pn (s) = (3)
s2 + 2ζωn s + ωn2
where:

• ωn : Natural undamped frequency (rad/s).

• ζ: Damping ratio (dimensionless).

Physical interpretations:

• ωn determines the system’s speed of response.

• ζ determines the oscillatory nature:

– ζ = 0: Undamped oscillations.
– 0 < ζ < 1: Underdamped, oscillatory decay.
– ζ = 1: Critically damped.
– ζ > 1: Overdamped, slow response.

Time domain response characteristics such as overshoot, settling time, and peak
time depend on these parameters.

1.4 Disturbance Observer (DOB) Concept


DOB estimates unknown disturbances acting on the plant and compensates them proac-
tively.
Consider the actual system:

Y (s) = P (s)U (s) + D(s) (4)


where D(s) is the unknown disturbance.
Using a nominal model Pn (s), the predicted output is:

Yn (s) = Pn (s)U (s) (5)


The difference signal is:

E(s) = Y (s) − Yn (s) = [P (s) − Pn (s)]U (s) + D(s) (6)


Assuming small model mismatch P (s) ≈ Pn (s):

E(s) ≈ D(s) (7)


The estimated disturbance is filtered by a Q-filter Q(s) and the inverse nominal plant:

D̂(s) = Q(s)Pn−1 (s)E(s) = Q(s)Pn−1 (s)[Y (s) − Pn (s)U (s)] (8)

2
1.5 Q-Filter Design
Q(s) is a stable, strictly proper low-pass filter designed to:

• Filter high-frequency noise.

• Ensure causality.

• Maintain unity gain at DC: limω→0 Q(jω) = 1.

A common second-order Q-filter is:

ωc2
Q(s) = √ (9)
s2 + 2ωc s + ωc2
where ωc is the cutoff frequency controlling disturbance rejection bandwidth and noise
sensitivity.

1.6 Control Law with DOB


The control input is adjusted by subtracting the estimated disturbance:

U (s) = UPID (s) − D̂(s) (10)


This allows the PID controller to focus on reference tracking, while DOB handles
disturbance compensation.

1.7 Stability and Robustness


The DOB internal loop stability depends on:

L(s) = Q(s)Pn−1 (s)P (s) (11)


For stability:

• L(s) must be proper and stable.

• Modeling error ∆(s) = P (s) − Pn (s) must be limited.

• The Q-filter bandwidth ωc must balance robustness and performance.

1.8 Summary of Key Equations

Component Equation
2
ωn
Nominal Plant Model Pn (s) = s2 +2ζωn s+ω
2
n
ωc2
Q-Filter Q(s) = √
s2 + 2ωc s+ωc2
Disturbance Estimate D̂(s) = Q(s)Pn−1 (s)[Y (s) − Pn (s)U (s)]
Control Law U (s) = UPID (s) − D̂(s)

Table 1: Key equations used in DOB-based control system

You might also like