Conceptual Design of Chemical Plants by Multi-Objective Optimization of Economic and Environmental Criteria Relatore: Prof. Davide MANCA Tesi di Piernico SEPIACCI matr. 837275 Anno Accademico 2015-2016 SCUOLA DI INGEGNERIA INDUSTRIALE E DELL’INFORMAZIONE DIPARTIMENTO DI CHIMICA, MATERIALI E INGEGNERIA CHIMICA «GIULIO NATTA» LAUREA MAGISTRALE IN INGEGNERIA CHIMICA
20
Embed
Conceptual Design of Chemical Plants by Multi …pselab.chem.polimi.it/wp-content/uploads/2016/12/Sepiacci-Thesis... · Multi-Objective Optimization of Economic and Environmental
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Conceptual Design of Chemical Plants by
Multi-Objective Optimization of Economic
and Environmental Criteria
Relatore: Prof. Davide MANCATesi di
Piernico SEPIACCI matr. 837275
Anno Accademico 2015-2016
SCUOLA DI INGEGNERIA INDUSTRIALE E DELL’INFORMAZIONE
DIPARTIMENTO DI CHIMICA, MATERIALI E INGEGNERIA CHIMICA «GIULIO NATTA»
LAUREA MAGISTRALE IN INGEGNERIA CHIMICA
Piernico Sepiacci – Milan, 21 December 2016
Definition of sustainability
A sustainable product or process:
• constraints resource consumption and waste generation to an acceptable level;
• makes a positive contribution to the satisfaction of human needs;
• provides enduring economic value to the business enterprise.
2
Economy Society
Environment
Sustainable Development
Economic System
Goods and Services
Materials and Energy
Emissions and Wastes
Piernico Sepiacci – Milan, 21 December 2016
Definition of sustainability
A sustainable product or process:
• constraints resource consumption and waste generation to an acceptable level;
• makes a positive contribution to the satisfaction of human needs;
• provides enduring economic value to the business enterprise.
3
Economy
Environment
Sustainable Development
Economic System
Materials and Energy
Emissions and Wastes
Piernico Sepiacci – Milan, 21 December 2016
Structure of the work
4
Economic SustainabilityUnder Market Uncertainty
Economic Objective Function
Environmental Sustainabilityby Application of WAR Algorithm
Environmental Objective Function
Process Modelingand Simulation
Multi-Objective Optimization
Reference Component
Selection
Time Series Analysis
Time Delay
Correlation
Econometric Modeling
Energy Generation
Chemical Process
( )ep
outI ( )cp
outI
( )cp
inI( )ep
inI
( )cp
weI( )ep
weI
Impact Indicators Selection
Impact Specific to Chemical k
kPro
fita
bili
ty
Pollution
Statistical Analysis of Pareto Fronts on Several Forecast Scenarios
HTPI HTPE TTP ATP
GWP PCOP AP ODP
Piernico Sepiacci – Milan, 21 December 2016
Process modeling and simulation
5
Process Modelingand Simulation
Multi-Objective Optimization
Pro
fita
bili
ty
Pollution
Statistical Analysis of Pareto Fronts on Several Forecast Scenarios
Piernico Sepiacci – Milan, 21 December 2016
The cumene manufacturing process
6
Reactions Kinetics
1) Cumene reaction
2) DIPB reaction
3) Transalkylation
3 6 6 6 9 12C H C H C H
3 6 9 12 12 18C H C H C H
12 18 6 6 9 122C H C H C H
7
1 2.8 10 exp 104,181/ ( ) B Pr RT C C
9
2 2.32 10 exp 146,774 / ( ) C Pr RT C C
8
3,
9 2
3,
2.529 10 exp 100,000 / ( )
3.877 10 exp 127,240 / ( )
f B D
b C
r RT x x
r RT x
Pathak, A. S., Agarwal, S., Gera, V., & Kaistha, N. (2011). Design and control of a vapor-phase conventional process and reactive distillation process for cumene production. Industrial & Engineering Chemistry Research, 50(6), 3312-3326.
Piernico Sepiacci – Milan, 21 December 2016
Economic sustainability
7
Economic SustainabilityUnder Market Uncertainty
Economic Objective Function
Reference Component
Selection
Time Series Analysis
Time Delay
Correlation
Econometric Modeling
Piernico Sepiacci – Milan, 21 December 2016
The feasibility assessment of chemical plants traditionally follows the economic guidelines suggested
in the Conceptual Design of Chemical Processes by Douglas (1988).
Hierarchy of decisions
1. Batch vs. Continuous;
2. Input-Output Structure of the flowsheet;
3. Recycle structure of the flowsheet;
4. General structure of the separation system;
5. Heat exchanger networks.
Conceptual design
8
-1,50E+08
-1,00E+08
-5,00E+07
0,00E+00
5,00E+07
1,00E+08
1,50E+08
2004 2005 2006 2007 2008 2009 2010 2011 2012 2013
Cu
mu
late
d E
P4
[U
SD]
Time [mo]
20
40
60
80
100
120
140
160
180
200
2004 2005 2006 2007 2008 2009 2010 2011 2012 2013
Pri
ce [
USD
/km
ol]
Time [mo]
Cumene Raw materials
2 ( ) ( )EP Product Price Raw Materials Cost
3 2 ( )EP EP Reactor Cost CAPEX OPEX
4 3 ( )EP EP Separation Cost CAPEX OPEX
1
4 4 4nMonths
t
t
Cumulated EP EP nMonths EP
Piernico Sepiacci – Milan, 21 December 2016
Methodology
1. Selection of a suitable reference component, which must be:
• chosen according to the market field of the chemical plant;
• a key component for either the process or the sector where the plant operates.
For the O&G sector and petrochemical industry, a good candidate for the reference component is crude oil.
2. Definition of the sampling time and time horizon of the economic assessment;
3. Identification of an econometric model for the reference component;
4. Identification of an econometric model for the raw materials and (by)products;
5. Identification of an econometric model for the utilities;
6. Use of the identified econometric models to determine the economic impact of the designed
plant in terms of Dynamic Economic Potentials (DEPs).
Predictive Conceptual Design (PCD)
9
Piernico Sepiacci – Milan, 21 December 2016
Time series analysis
10
Use of moving-averaged valuesMoving average allows eliminating most of the high-frequency fluctuations and catching the price trend.
(Auto)correlation analysis(Auto)correlograms report the (auto)correlation index of the time series X and Y based on the time lag between them.
Identification of the candidate econometric model
Evaluation of the adaptive parametersIt requires a linear regression procedure that minimizes the sum of square errors between real and model prices.
1
1
0
1 n
t
i
m Yn
Pri
ce s
eri
es
Time
Real data Moving-averaged data
cov ,corr ,
var var
X YX Y
X Y
cov ,corr ,
var var
t t j
t t j
t t j
Y YY Y
Y Y
0
1
0 1 2 3 4 5 6 7
Time lag
Correlation index
0 1 1 2 2 1 1 2 2... ...t t t q t q t t p t pY a a X a X a X bY b Y b Y