Process Dynamics & Control

Process Dynamics & Control (CH 61016)

Class Teacher: 1. Prof Amar Nath Samanta
2. Prof Parag Deshpande

Class TA
Mr. Shubham Raj
Mr. Vikas Dhamu

Class Timing
Wednesday : 10 AM to 11 AM
Thursday : 9 AM to 10 AM
Friday : 11 AM to 1 PM (PD)

Attendance
• Biometric device
Must Complete Biometric Registration as early as possible.
Process Control Syllabus
• Brief Review of process model and transfer function
• Review of feedback control
• Control of time delay and inverse response processes
• Traditional Advanced Control
• Feedforward Control
• Cascade Control
• Ratio Control
• Override Control
• Model based Control
• Direct Synthesis
• Internal Model Control
• Model Predictive control
• Unconstrained DMC
• Quadratic DMC
• Controller design using State-Space model
• State Feedback Controller
• Linear Quadratic Regulator
• State Observer
• Nonlinear Control
• Feedback linearizing Control
• Backstepping Control
• Sliding mode control
 Process Control

Books:
 Process Dynamics & Control -- Seborg, Edgar, Mellichamp &
Doyle
 Process Control -- W Bequette
 Modern Control Engineering -- K Ogata
Benchmark Systems to be used for the course

• Tank Level System

• 3 tank in series
• Quadruple Tank
• Continuous Stirred Tank Heater
• Continuous Stirred Tank Reactor system
• Isothermal Reactor (Van de Vusse reaction)
• Non-isothermal Reactor
• Biochemical reactor
• Distillation Column
Mathematical Model of tank level system

Control objective:
Control the 3rd tank level by
manipulating pump voltage V
 Mathematical Model of tank level system

dh1 F0 a1 Control objective:

= − 2 gh1
dt A1 A1 Control the 3rd tank level by
manipulating F0
dh2 a1 a2
= 2 gh1 − 2 gh2 Add q2 input flow on the tanks 2 and
dt A2 A2 set q2 as manipulated variable.
Which one is difficult to control?
dh3 a2 a3
= 2 gh2 − 2 gh3
dt A3 A3

Data for Simulation in Matlab/Simulink

A1, A3 = 28 cm2 A2 = 32 cm2 Vs = 3.15 V
a1, a3 = 0.071 cm2 a2 = 0.057 cm2 K=3.14
Quadruple Tank problem
Quadruple Tank problem

dh1 a1 a3 γ 1k1 Control Objective

=
− 2 gh1 + 2 gh3 + v1
dt A1 A1 A1 Both the levels of tank1 and
tank2 should be controlled by
dh2 a2 a4 γ 2 k2 manipulating voltages to the
=
− 2 gh2 + 2 gh4 + v2
dt A2 A2 A2 pumps.
Data for simulation:
dh3 a3 (1 − γ 2 )k2
=
− 2 gh3 + v2 A1, A3 = 28 cm2
dt A3 A3
A2, A4 = 32 cm2
dh2 a4 (1 − γ 1 )k1 a1, a3 = 0.071 cm2
=
− 2 gh4 + v1
dt A4 A4 a2, a4 = 0.057 cm2
k1, k2 = 3.33, 3.35
v1, v2 = 3.0, 3.0
γ1, γ2 = 0.7, 0.6
Jacketed heated stirred tank
F, Ti

Fj, Tji
h
Fj, Tj

F, T

Assumptions:
 Constant hold-up in tank and jacket
 Constant heat capacities and densities
 Incompressible flow
Jacketed heated stirred tank

dT F UA
= (Ti − T ) + (T j − T )
dt V ρ CPV
dT j Fj UA
= (T ji − T j ) − (T j − T )
dt V j ρ j CPjV j
Parameter and steady state values
F = 30 l/m; Fjs = 50 l/m; Ti = 150C;
Tji = 930C; V = 300 l; Vj = 30 l;
rhocp = 1Kcal/K lit; rhocpj= 1.384;
UA = 100 Kcal/min K;
T= 600C; Tj = 73.50C;
Continuous Stirred Tank Reactor

Fin

Fout
A 
→ B → C
k1 k2

A 
→ 2D
k3 Assumption:
1. Exit Condition = Reactor Condition
2. Isothermal Reaction
3. Constant volume, i.e, Fin = Fout=F
4. Only reactant A in feed is consumed, i.e,
other reactant is in large excess. No
product in the feed
Continuous Stirred Tank Reactor
Constant volume : dV/dt = 0
Dynamic model
𝑑𝑑𝐶𝐶𝐴𝐴 𝐹𝐹
• = 𝐶𝐶𝐴𝐴𝐴𝐴 − 𝐶𝐶𝐴𝐴 − 𝑘𝑘1 𝐶𝐶𝐴𝐴 − 𝑘𝑘3 𝐶𝐶𝐴𝐴2 Data for the CSTR
𝑑𝑑𝑡𝑡 𝑉𝑉
𝑑𝑑𝐶𝐶𝐵𝐵 𝐹𝐹 k1 = 5/6; k2 = 5/3
• = − 𝐶𝐶𝐵𝐵 + 𝑘𝑘1 𝐶𝐶𝐴𝐴 − 𝑘𝑘2 𝐶𝐶𝐵𝐵 K3 = 1/6; Caf = 10
𝑑𝑑𝑑𝑑 𝑉𝑉
𝑑𝑑𝐶𝐶𝐶𝐶 𝐹𝐹
• = − 𝐶𝐶𝐶𝐶 + 𝑘𝑘2 𝐶𝐶𝐵𝐵
𝑑𝑑𝑑𝑑 𝑉𝑉
𝑑𝑑𝐶𝐶𝐷𝐷 𝐹𝐹 1
• = − 𝐶𝐶𝐷𝐷 + 𝑘𝑘3 𝐶𝐶𝐴𝐴2
𝑑𝑑𝑑𝑑 𝑉𝑉 2

Since our objective is to control CB, the reduced order model should be
𝑑𝑑𝐶𝐶𝐴𝐴 𝐹𝐹
• = 𝐶𝐶𝐴𝐴𝐴𝐴 − 𝐶𝐶𝐴𝐴 − 𝑘𝑘1 𝐶𝐶𝐴𝐴 − 𝑘𝑘3 𝐶𝐶𝐴𝐴2 = 𝑓𝑓1 (𝐶𝐶𝐴𝐴 , 𝐶𝐶𝐵𝐵 , 𝐹𝐹⁄𝑉𝑉)
𝑑𝑑𝑑𝑑 𝑉𝑉
𝑑𝑑𝐶𝐶𝐵𝐵 𝐹𝐹
• = − 𝐶𝐶𝐵𝐵 + 𝑘𝑘1 𝐶𝐶𝐴𝐴 − 𝑘𝑘2 𝐶𝐶𝐵𝐵 = 𝑓𝑓2 (𝐶𝐶𝐴𝐴 , 𝐶𝐶𝐵𝐵 , 𝐹𝐹⁄𝑉𝑉)
𝑑𝑑𝑑𝑑 𝑉𝑉