Scribd
Upload
114 views

Process Optimization - 01

This document provides an introduction to process optimization and defines key concepts such as design and operating variables, degrees of freedom, and developing a complete process optimization model. It discusses continuous, discrete, and binary variables and how they relate to optimization problems. The document also provides an example of calculating degrees of freedom and developing the equality and inequality constraints and objective function for a mathematical process model.

Uploaded by

Abdul Majid Ia
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
114 views

Process Optimization - 01

This document provides an introduction to process optimization and defines key concepts such as design and operating variables, degrees of freedom, and developing a complete process optimization model. It discusses continuous, discrete, and binary variables and how they relate to optimization problems. The document also provides an example of calculating degrees of freedom and developing the equality and inequality constraints and objective function for a mathematical process model.

Uploaded by

Abdul Majid Ia
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 16

CCB 4333 PROCESS OPTIMIZATION

DEFINITION AND PROBLEM FORMULATION

ZULFAN ADI PUTRA, PDENG

OFFICE: 04-03-31
TELP: 05 368 7562
EMAIL: [email protected]

1
Contents

1. Definition
2. Problem formulation
3. Degrees of Freedom (DOF)

2
Design and operating variable
Design & operating variable is a trade-off Optimum value

3
Types of variables

Continuous - can take any value in a region, e.g. temperature,


pressure, flowrate,

Discrete (e.g. integer) - can only take discrete values, e.g. no.of
reactors, no.of columns in a sequence, no. of trays in a column, ...
Binary variables - can only take 0 or 1, e.g. such equipment
exists or does not exist, ...

4
Objective function
Other resources:
- Fuel (e.g. for energy)
- Water (e.g. for wash)
- Solvents

Raw materials Product

Investment
(Capital Expenditure or CAPEX) By products
Waste
Wastewater
Objective function = f(product, raw materials, energy, water, solvents, equipment design,
waste, etc.)

Optimization is to find what are the design and operating variables


that yield the optimal value of an objective function
5
Process model and constraints
A = mol/hr
B = mol/hr
C = mol/hr

A = 1 mol/hr
Reactor Distillation
B = 1 mol/hr

Catalytic fixed bed reactor


A + B C (Reaction 1)
B + C D (Reaction 2) D = mol/hr
A is converted 50%
Selectivity of B to Reaction 1 is 80%
Relative volatilities A > B > C > D

What are the constraints in the figure?

Constraints + objective function = process model


(equalities and inequalities)

6
Example
Given:
T s = 25 oC
w Cps, Cpw = Specific heat capacities
Fw = ? C = unit cost of water
o
Cpw = 4.2 kJ/kg C U = overall heat transfer coefficient, 1000
Ts = 170OC Tt = 80 oC W/(m2.oC)
A R = annualisation factor
s A= ? Q =? Exchanger capital cost ($) = a A0.84
F = 500 kg/hr
Cps =1kJ/kgOC
t Subscript: w = tube; s = shell side
Tw = ?
Superscript : s = supply temperature;
t = target temperature

Objective: Design a minimum cost heat exchanger


Assume:
Counter current with T min = 10 oC

Please write down:


1. Variables
2. Equality constraints
3. Inequality constraints
4. Objective function

7
A COMPLETE PROCESS MODEL
In general, a mathematical programming model or process model
can be expressed in the following format:
minimise f(x) - objective function
Or maximise

Subject to:

h(x) = 0 - equality constraints

g(x) 0 - inequlity constraints


g(x) 0

Where x is assumed as a set of continuous variables.

8
A COMPLETE PROCESS MODEL
Minimize f = R(aA0.84) + CFw Objective function
Subject to:
H1: Q FsCps(Ts-Tt) = 0
H2: Q FwCpw(Ttw-Tsw) = 0 Equality constraints
H3: Q UA(Tt-Tsw) (Ts-Ttw) / ln{(Tt-Tsw)/ (Ts-Ttw)}

G1: Ts Ttw 10 Inequality constraints


G2: Fw, Ttw, A 0

9
NOTES

if f(x), g(x) and h(x) are all linear, then this is a linear programming problem (LP).

If any of them is non-linear, it becomes a non-linear programming problem (NLP).

If some variables are integer, then it is either mixed integer linear programming
problem (MILP) or mixed integer non-linear programming problem (MINLP).

10
6 Steps to Solve Optimization Problem

1. Analyze the process to define variables


2. Determine the objective function
3. Develop mathematical models for the variables and their
constraints
4. If the problem is too large, then:
Break into manageable parts, and/or
Simplify the objective function
5. Apply suitable optimization technique to solve the problem
6. Check and examine the sensitivity of the results

11
Degrees of Freedom (DOF)

Example
Minimise f = x2 x1 + 2x32 - 4x4x5
Subject to
x3 - 2x1 + x2 = 2

x4 + x1 - 3x2 = 2

x5 + x1 + x2 = 4

Number of equations (NE) is 3


Number of variables (NV) is 5.

12
Rearrange the constraints:
x3 = 2 + 2x1 - x2

x4 = 2 + 3x2 - x1

x 5 = 4 - x1 - x2

If we assume any value for x1 and x2, other three


variables (x3, x4 and x5) are fixed and can be calculated
x1 and x2 independent variables
x3, x4, x5 dependent variables
This implies two degrees of freedom (choosing x1 and x2)

13
For some values of x1 and x2, f can achieve a minimum, determined by optimisation

Therefore, a problem is an optimisation problem if DOF 1

In general

DOF = NV - NE

for this example DOF = 5 - 3 = 2

14
Questions
1. What is DOF if one more equation is added without adding new
variables to our example?

2. How about adding two more equations without introducing new variables?
Is it still an optimisation problem?

3. What happens if more than two equations are added to the example,
without adding new variables?

15
Quiz
Heavy naphtha Catalytic Gasoline high octane
(A) Reforming (E)

You are process engineers. New comers!


Your boss has to optimize the profit for the day.
Gasoil
Hydrotreating
Treated Gasoil Your task is to advise him on what to do.
(B) (F)
Good luck!!!

Residue Gasoline
Hydrocracking
(C) (G)

Availability, max Reactant Processing Selling price


Raw materials Cost ($/kg) Products
(kg/day) requirement cost ($/kg) ($/kg)
A 40000 1.5 E 2/3A + 1/3B 1.5 4
B 30000 2 F 2/3A + 1/3B 0.5 3.3
C 25000 2.5 G 1/2A + 1/6B + 1/3C 1 3.8

On your working table, there is a note:


Develop the mathematical optimization model!

16

You might also like