heat integration - introduction - iv - ntnu av slides/heat-part1... · t. gundersen heat...

T. Gundersen

Heat Integration - Introduction •  Sub-Topics

§  Design of Heat Exchanger Networks §  Selection of Utilities

Ø  Consumption and Production §  “Correct” Integration

Ø  Distillation and Evaporation Ø  Heat Pumps Ø  Turbines (Heat Engines)

§  Heat and Power Considerations •  Project Applications

§  New Design (“grassroot”) §  Reconstruction (“retrofit”)

Process, Energy and System

Heat 1

Heat Integration − Introduction

T. Gundersen

x L

dx

A , U Th,in

mCph

mCpc

Th,out

Tc,in

Tc,out

Pure Countercurrent Heat Exchanger

Heat 2



Qh = ΔHh = mCph ⋅ Th,in −Th,out( )Qc = ΔHc = mCpc ⋅ Tc,out −Tc,in( )

Q = A ⋅U( ) ⋅ ΔTLMQh =Qc =QΔTLM = ??

WS-1: Heat Integration

T. Gundersen

Stream Ts Tt mCp ΔH °C °C kW/°C kW

H1 300 100 1.5 300 H2 200 100 5.0 500 C1 50 250 4.0 800

Steam 280 280 (var) Cooling Water 15 20 (var)

Specification: ΔTmin = 20°C

Find: QH,min , QC,min

Tpinch , Umin Umin,MER

and Network

Notice: 1) H1 and H2 provide as much heat as C1 needs (800 kW) 2) Ts (C1) < Tt (H1,H2) - 20°C & Ts (H1) > Tt (C1) + 20°C

Heat 3



Procesintegration i industrien

Possible Solutions

I II C1

H1 H2

H

C

B

I II C1

H2 H1

H

C D

I II C1

H2 H1

H

C C

T. Gundersen

I II C1

H1 H2

H

C

A 300 200

50

Heat 4



More Solutions

I

II

C1

H1 H2

H

C F

I II C1

H2 H1

H

C

III

E

Question: 1) Minimum Energy ? 2) Which Network ?

T. Gundersen Heat 5



Summary for WS-1

T. Gundersen

Design QH = QC (kW)

A 280 B 280 C 380 D 142.5 E See later . . . . F Around 149 (optimal split)

X Any better ???

Heat 6



We need Best Performance Targets !!

T. Gundersen

•  Data Extraction (assume R/S given) •  Performance Targets

§  Energy, Area, Units / Shells §  Total Annual Cost (gives value for ΔTmin)

•  Process Modifications (change R/S ?) •  Design of Network (minimum energy)

§  Decomposition at the Pinch §  Qualitative and Quantitative Tools

•  Optimization (given the basic structure) §  Heat Load Loops & Paths and Stream Splits

Phases in the Design of Heat Exchanger Networks

Heat 7



Illustrating Example

Reactor

Feed

Product

Distillation Column

Compressor

50°

210°

160°

210°

130°

220°

160°

270°

60° Reboiler

Condenser

T. Gundersen Heat 8



Phase 1: Data Extraction • Time consuming (80%) and Critical Phase • New Design: Material / Energy Balances • Retrofit: Various Sources of Data

§  Measurements (that are often incorrect) §  Simulations (M+E balances è Consistent) §  Design Basis (original, but was it updated ?)

•  Should keep Degrees of Freedom open •  Practical Limitations must be included,

but Cost Effects should be evaluated before such Constraints are accepted

T. Gundersen Heat 9

Heat Integration − Data Extraction


Necessary Data (1) • Process Streams

§  Flowrates m kg/s, tons/h §  Specific Heat Capacity Cp kJ/kg°C §  Supply Temperature Ts °C §  Target Temperature Tt °C §  Heat of Vaporization ΔHvap kJ/kg §  Film Heat Transfer Coeffs. h kW/m2 °C

• Utility Systems §  Temperature(s) and Specific Heat Content §  Price per Energy Unit (Amount)

T. Gundersen Heat 10



Necessary Data (2) •  Cost Data for Heat Exchangers

§  Relation between Area and Purchase Price §  Example: Cost = a + b * (A) C

•  Economical Data §  Number of Operating Hours per year §  Specification on required Payback Time §  Installation Factors §  Maximum Investment (for Retrofit Projects) §  Minimum allowed Approach Temperature

(ΔTmin), possibly based on Pre-optimization




Data Extraction for the Example Stream ID Ts Tt mCp ΔH h

°C °C kW/°C kW kW/m2°C

Reactor outlet H1 270 160 18 1980 0.5 Product H2 220 60 22 3520 0.5 Feed Stream C1 50 210 20 3200 0.5 Recycle C2 160 210 50 2500 0.5

Reboiler C3 220 220 2000 1.0 Condenser H3 130 130 2000 1.0

HP Steam HP 250 250 (≅ 40 bar) (var.) 2.5 MP Steam MP 200 200 (≅ 15 bar) (var.) 2.5 LP Steam LP 150 150 (≅ 5 bar) (var.) 2.5 Cooling Water CW 15 20 (var.) 1.0




Phase 2: Targeting •  Basis:

§  Minimum Approach Temperature: ΔTmin •  Results:

§  Minimum External Heating: QH,min §  Minimum External Cooling: QC,min §  Minimum Number of Units: Umin §  Minimum Heat Transfer Area (total): Amin

•  Pre−Optimization: Near optimal ΔTmin §  QH,min , QC,min , Umin , Amin = f (ΔTmin) §  TAC = f(QH,min , QC,min , Umin , Amin) = f (ΔTmin)


Heat Integration − Targeting


Composite Curves

Stream Ts Tt mCp ΔH H1 270 160 18 1980 H2 220 60 22 3520

H(kW)

270

220

160

60

2000 4000 6000

T(°C)

A

B C

D

270

220

160

60

2000 4000 6000

T(°C)

H(kW)

A

B + C

D




Composite Curves for the Example

2000 4000 6000 0

300

250

200

150

100

50

T (°C)

H (kW)

QH,min

QC,min

Pinch

QRecovery

ΔTmin

Note: The Column Reboiler / Condenser

are not included




Trade-off: Area vs. Energy 300

250

200

150

100

50

T (°C)

H (kW)

2000 4000 6000 0

QH,min

QC,min

+δ

+δ




Pinch Point and Decomposition QH,min + δ

QC,min + δ

δ

300

250

200

150

100

50

H(kW)

2000 4000 6000 0

T (°C)

Surplus

Deficit




Summary of Targeting by Graphical Methods

•  Provides an Overview and Understanding •  Identifies Key Information about the

Heat Recovery Problem, such as: §  Minimum External Heating, QH,min §  Minimum External Cooling, QC,min §  The Process Pinch Point (Bottleneck)

•  More Graphical Diagrams later •  Time consuming and subject to Errors




Heat Cascade

Numerical Methods

Intervals

270ºC - - - - - - - 250ºC

230ºC - - - - - - - 210ºC

220ºC - - - - - - - 200ºC

180ºC - - - - - - - 160ºC

160ºC - - - - - - - 140ºC

70ºC - - - - - - - - 50ºC

H1

H2

CW

C1

C2

ST

720 kW

180 kW

720 kW

880 kW

440 kW

1980 kW

500 kW

200 kW

800 kW

1800 kW

+ 720

- 520

- 1200

2000 kW

400 kW

+ 180

+ 220

+ 400

60ºC - - - - - - - - 40ºC

360 kW

220 kW

ΔTmin = 20°C




Pinch Point changes with ΔTmin




0

50

100

150

200

250

300

0 2000 4000 6000 8000 10000

H (kW)

T (°C)

Pinch Point Candidates




0

50

100

150

200

250

300

0 2000 4000 6000 8000 10000

H (kW)

T (°C) Pinch Candidates ( ) are the Supply Temperatures of Streams or Stream ”Segments”

Total mCp is then increasing

T. Gundersen


Heat 22

Motivating Example – Remember?


Area = 629 m2 Area = 533 m2

The Actual Flowsheet & Data Extraction

T. Gundersen


Heat 23

Focus: Thermal Energy and the Energy Balance First: Checking the Mass Balance:

(1) 1.089 + 1.614 = 2.703 (OK)

(2) 6.931 + 2.703 = 9.634 (OK)

(3) 7.841 + 0.179 + 1.614 = 9.634 (OK)

Symbol: kg/s


1

3 2

T. Gundersen


Heat 24

Simple Data Extraction

Ref.: Ian C. Kemp, ”Pinch Analysis and Process Integration: A User Guide on Process Integration for the Efficient Use of Energy”, 2nd. Ed., IChemE, Elsevier, 2007

40°C

1

2

3

4

Str. Ts Tt ΔH mCp

1 5 200 2233 11.451

2 40 200 413 2.581

3 35 126 992 10.901

4 200 35 2570 15.576

T in ºC, ΔH in kW and mCp in kW/ºC


”Average”

Composite Curves and Energy Targets



0

50

100

150

200

250

0 500 1000 1500 2000 2500 3000 3500 4000

Q (kW)

T (°

C)

PRO_PI1: ΔTmin = 10°C QH,min = 1068 kW and QC,min = 0 kW


Energy Requirements vs. ΔTmin



PRO_PI1: ΔTmin = 22°C QH,min = 1068 kW and QC,min = 0 kW ΔTmin = 23°C QH,min = 1083 kW and QC,min = 15 kW

ΔTthrsh = 22.1°C Threshold Process becomes Pinched

0 200 400 600 800

1000 1200 1400 1600 1800 2000

0 10 20 30 40 50 60 70 80

Global temperature difference (K)

Q (k

W)

QH,exist = 1722 kW

HRATexist = 64°C


Q: What about Phase changes & Cp=Cp(T)?

T. Gundersen


Heat 27

Feed To Column

Reactor

ST

ST CW

Flash

I

H1

C

III

II

H2

40ºC

200ºC

190ºC 153ºC

200ºC

128ºC

5ºC

115.5ºC 114ºC 35ºC

141ºC Recycles

1

2

3

4

I II 1 4

II II IIII 1 2 4

III IIIIII 3 4

H1H1 1

H2 H2H2 1 2

CC 4

III4

"tot"4

??

992 kW

1652 kW

70 kW

654 kW992 39.68 kW/ C

153 1282570 15.576 kW/ C

200 35

Q H HQ H H H

Q H HQ H

Q H HQ H

mCp

mCp

= Δ = Δ =

= Δ +Δ = Δ

= Δ = Δ =

= Δ =

= Δ +Δ =

= Δ =

= = °−

= = °−

Q: Can we solve this Puzzle?

654 kW 1652 kW

70 kW

126ºC


T. Gundersen


Heat 28

0

50

100

150

200

250

0 500 1000 1500 2000 2500 3000 3500 4000 Q (kW)

T (°

C)

0

50

100

150

200

250

0 500 1000 1500 2000 2500 3000 3500 4000 Q (kW)

T (°

C)

Simple

Detai led


Energy Requirements vs. ΔTmin



PRO_PI1: ΔTmin = 10°C QH,min = 1068 kW and QC,min = 0 kW ΔTmin = 11°C QH,min = 1072 kW and QC,min = 4 kW

ΔTthrsh = 10.5°C Threshold Process becomes Pinched

0

500

1000

1500

2000

2500

0 10 20 30 40 50 60 70 80

Global temperature difference (K)

Q (k

W)

QH,exist = 1722 kW

HRATexist = 42.4°C


Lessons learned from the Example



•  Data Extraction is a Critical Activity §  We have to solve a “Puzzle” §  Plant Data are often missing or incorrect §  Collecting and ”Processing” Data is Time

consuming (80% of Conceptual Studies) §  Remember: ”Garbage in means Garbage out”

•  Energy Targeting is much Simpler •  Next: Heat Exchanger Network Synthesis

(Design & Optimization) is manageable


Fewest Number of Units

1000 ST

1980 H1

3520 H2

C1 3200

CW 800

C2 2500

1000 2200 1320

1180

800

N = Number of Streams and Utilities N = nH + nC + nutil = 2 + 2 + 2 = 6 Umin = (N-1)

L = Loops S = Subnetwork U = Units Euler’s Rule: U = N + L - S

Assume: L = 0, S = 1




Umin,MER = (N-1)above + (N-1)below

Maximum Energy Recovery requires Decomposition at Pinch Point(s)

Process Pinch 180°

C2 210° 160°

C1 210° 50° H2 220° 60°

H1 270° 160°

160°

HP 250°

CW 15° 20°




heat integration - introduction - iv - ntnu av slides/heat-part1... · t. gundersen heat...

Documents