spray dryer exhaust heat recovery

Spray Dryer Exhaust Heat Recovery

A Techno-economic Assessment Model

Tim Walmsley, M Walmsley, M Atkins, J Neale

PRES 2014

Outline

Research Motive/Context

Overarching Goal

Review of Progress

Exhaust Heat Recovery Modelling

Conclusions

New Zealand University of Waikato

Dairy

Cows

NZ Dairy Performance

0%

10%

20%

30%

0

5

10

15

20

1990 2000 2010 2020

Pe

rce

nta

ge o

f to

tal e

xpo

rts

Val

ue

of

exp

ort

s ($

bill

ion

s)

Year

FutureGrowth?

DairyProducts $$

Milk Powders $$

ExportShare %

Source: Statistics New Zealand, 2013

42.2

19.9

17.115.0

6.2

2.9 2.6

32.1

0

5

10

15

20

25

30

35

40

45

Food &beverage

Petroleum& chemical

Pulp &paper

Woodproduct

Non-metallicmineralsproduct

Metalsproduct

Othermanufacturing

Ener

gy u

se b

y N

Z m

anu

fact

ori

ng

in 2

01

2 [

PJ]

Estimated contributionfrom Dairy processing

NZ Manufacturing Process Heat Use 2012

Source: Survey of New Zealand Energy Use: Industrial and trade sector 2012

Coal & N.G.Fuel

Supply

MP Process Demand

ConversionLosses

Milk Powder Production Utility Demands

H

Condenser

Eva

po

rato

r

Eva

po

rato

r

Eva

po

rato

r

TVR

Spray Dryer

Fluidised Bed (1)

Fluidised Bed (2)

Bag

House

Milk Concentrate

Std Milk

H

H

H

Exhaust Air

CO

W (1

)

HCIP

Steam

Water

Vap. (3)

Inlet Air

CO

W (2

)

CO

W (3

)

Evaporators Spray Dryer

Clean-In-Place (Hot Water)Fluidised Bed (3)

H

Dry

Powder

MVR MVR

CIP

Dry

Powder

Milk Powder Plant

Source

Sink

To cooling

tower

C C C C

H

H

C

Vapour

Liquid

Gas

Important!

Aim: To Investigate How to Maximise Economic Heat Recovery in Milk Powder Production?

Progress to Achieving the Research Goal

4) Heat Exchanger

Design

2) Heat Recovery

Loop

1) Direct Heat Integration

3) Milk Powder Fouling

(1a) PDM Heat Integration Schemes (PRES’12)

Walmsley et al. (2013), App Therm Eng

Milk

(1a) PDM Heat Integration Schemes (PRES’12)

MER A: Split milk & direct condenser integration

13

8

53.954

Pinch

Std milk

COW

Cond. vap.

CIP

64.3

H

H

H

H

H

H

H

59.3

48.5

63.4 58.1

Exhaust air

Milk conc.

FB 1

FB 3

FB 2

Dryer inlet air

15

200

32

45

49

65 54

5575

55 1548.5

Eva

po

rato

r

Zo

ne

Dry

er

Zo

ne

CIP

Zo

ne

42.8

62.5

27.562.5

27.5

15

MER B: Cyclic milk matching & direct condenser integration

13

8

53.954

Pinch

Std milk

COW

Cond. vap.

CIP

64.3

H

H

H

H

H

H

H

44.163.4 58.1

Exhaust air

Milk conc.

FB 1

FB 3

FB 2

Dryer inlet air

15

200

32

45

49

65 54

5575

55 1545.8

Eva

po

rato

r

Zo

ne

Dry

er

Zo

ne

CIP

Zo

ne

52.8

62.5

27.562.5

27.5

48.5

53.5

15

Std milk

CO

W (1

)

Cond.

Vap.

CO

W (2

)

COW (3)

H

CIP

To drain

H

CO

W

Exhaust

Air

Dryer

Inlet AirH

Std milk

CO

W (1

)

Cond.

Vap.

CO

W (2

)

COW (3)

H

CIP

To drain

H

CO

W

Exhaust

Air

Dryer

Inlet AirH

13

8

53.954

Pinch

Std milk

COW

Cond. vap.

CIP

64.3

H

H

H

H

H

H

H

59.343.563.4 57.6

Exhaust air

Milk conc.

FB 1

FB 3

FB 2

Dryer inlet air

15

200

32

45

49

65 54

5575

55 15

Eva

po

rato

r Z

on

eD

rye

r Z

on

eC

IP

Zo

ne

48.5

41.248.5

41.2

21.6

42.8

62.5

27.562.5

27.5

15

43.5

MER D: Cyclic milk matching & indirect condenser integration

13

8

53.954

Pinch

Std milk

COW

Cond. vap.

CIP

64.3

H

H

H

H

H

H

H

59.3

43.5

63.4 57.6

Exhaust air

Milk conc.

FB 1

FB 3

FB 2

Dryer inlet air

15

200

32

45

49

65 54

5575

55 15

Eva

po

rato

r Z

on

eD

rye

r Z

on

eC

IP

Zo

ne

48.5

18.548.5

18.5

21.6

42.8

62.5

27.562.5

27.5

15

MER C: Split milk & indirect condenser integration

43.5

Exhaust

Air

Std milk

CO

W (1

)

Cond.

Vap.

CO

W (2

)

COW (3)

H

CIP

To drain

Dryer

Inlet AirH

H

CO

W

FB 1,2,3

Exhaust

Air

Cond.

Vap.

CIP

Dryer

Inlet AirH

HFB 1,2,3

Std milk

CO

W (1

)

CO

W (2

)

COW (3)

H

To drain

CO

W

9 % Cost reduction, 30% Heat Recovery (HR) improvement


(1b) The Cost Derivative Method (PRES’13)

Stream

data

Cost

data

ΔTcont

Pinch

targets

HEN

structure

HE

selection

Adjust

soft

data?

Yes

No

Yes

No

Pinch Design Method

Add UE’s to all

streams

HEN design

(CDM)

De

sig

na

te a

RE

to

act a

s a

TE

Cost Derivative Method

Find θ for RE’s

to UE’s

Iteratively solve

Initialisation

No

No

Yes

Yes

HEN design(s)

using ΔTcont

HE

N s

tru

ctu

re

Relax

network?

Limiting

Tt?

Eliminate

UE?

1

C

H

x

+dTy2

-dQCx

-dQHy

+dQ1

-dTx2

y

+dA1

Hot streams

Co

ld s

tre

am

s

+dTyn

-dTxn

Other

network HE’s

-dCC,ut

-dCH,ut

Tx,s

Ty,s

Tx,t

Ty,t

TC1

TC2

TH1

TH2

-dACx

-dAHy

1

)(

1

)(

1

1

1 dA

dS

dA

dCC

dA

dCC

dA

dTC iutiut

(a) PDM

13 °C

8 °C

53.9 °C54 °C

Milk

COW

Vap.

CIP

64.3 °C1

32

H1

63.4 °C

55 °C 15 °C

(b) CDM

12748 kW

1214 kW 1198 kW

0.11

1452 kW

357 kW

12.1 °C

8 °C

53.9 °C54 °C

Milk

COW

Vap.

CIP

64.3 °C1

32

63.4 °C

55 °C 15 °C

12977 kW

916 kW 1496 kW

0.07

1521 kW

59 kW

H2

H1

H2

5 % Costreduction



+

+ -

-+

+

-

-

-

-

+dA, +dQ



(a) Non-self-interacting open loop

C1

H1

H2

C2

(b) Self-interacting closed loop

C1

H1

H2

No flow-on

C

C

H

H

C

C

H

H

(2) New Heat Recovery Loop Design Method for Improved Inter-plant Heat Integration (PRES’13)

Processing Site

Cold

storage

Process A

H1

Process B

H2

Process C

C1

Process D

H3

Hot

storage

Cold supply

Cold return

Hot supply

Hot return

Process E Process F

C3C2

Thermal

storage

C C H

C H H

Spray Dryer

Exh

0

2

4

6

8

10

12

0 25 50 75 100

C[k

W/°

C]

Time [h]

0

2

4

6

8

10

12

0% 25% 50% 75% 100%

C[k

W/°

C]

Ordered C

Time-ave C

Median C

Peak C

Walmsley et al. (2014), Energy


Stream

data

Select

Qr

Composite

curves

Calculate

Cl(h)

and Cl(c)

Calculate

ΔTmin

Select

Tlc and T

lh

DetermineT

ho and T

co

Scale Cl

CalculateHE areas

Transient

HRL model

T

ΔH

Tco

Tho

Qr

Tlc

Tlh

Cl(h)

Cl(c)

Cold Storage Pinch

T

ΔH

Tco

Tho

Qr

Tlc

Tlh

Cl(h)

Cl(c)

Hot Storage Pinch

ΔTmin

ΔTmin

ΔTmin

ΔTmin

ΔTadd



7

8

9

10

11

12

0 200 400 600 800 1000

Ave

rage

he

at r

eco

very

an

d s

ola

r h

eat

ing

[MW

]Storage tank volume [m3]

CTS

VTS

VTS with solar

CTS with solar

New HRL Design Method

12.3 MW

Conventional HRL Design Method

9.2 MW

SOLAR

SOLAR


0

20

40

60

0%

25%

50%

75%

100%

25 27 29 31 33 35 37 39

Sto

rage

te

mp

erat

ure

[°C

]

Ho

t st

ora

ge le

vel [

%]

Day

Tlh

Tlc

Storagelevel

0

20

40

60

0%

25%

50%

75%

100%

25 27 29 31 33 35 37 39

Sto

rage

te

mp

erat

ure

[°C

]

Ho

t st

ora

ge le

vel [

%]

Day

Tlh

TlcStorage

level

(3) Milk Powder Fouling of Flat Plates

Walmsley et al. (2014), J Food Eng

(3) Milk Powder Fouling Model

SMP Deposition model

0.1

1

10

100

1000

10000

20 30 40 50 60 70

γ s[J

/m2 ]

(T - Tg)crit* [°C]

This work - normal impact

This work - oblique impact

Zhao (2009), corrected

Murti et al. (2009)

Paterson et al. (2007)

Hogan et al. (2010)

Murti et al. (2010)

Hennigs et al. (2001)

Particle gunPaterson and co-workers

Fluidised bed tests

Mechanical stir test

Particle gun tests

Model

Upper bound

Lower bound

n

g

n

n

g

n

critg

G

BY

a

YvrD

G

BY

a

YvrD

TT

22525653

1

22525653

2

tan*4

*11

*2442.0log

tan*4

*11

*2442.0log

*)(

Walmsley et al. (2014), J Food Eng

(3) Milk Powder Fouling of Tubes

Front Back

Walmsley et al. (2014), HXF&C Conference

(3) Milk Powder Fouling of Tubes

Round Tube Elliptical Tube

Walmsley et al. (2014), HXF&C Conference

(3) Milk Powder Fouling of Fins

Walmsley et al. (2013), Adv Powder Tech

(3) Model Validation for Tube Fouling

0

10

20

30

40

50

60

70

80

90

30 40 50 60 70 80

Cri

tica

l im

pac

t an

gle

[°]

T – Tg* [°C]

Round tube (4.5 m/s)

Elliptical tube (4.5 m/s)

Model (4.5 m/s)

90°

30°

60°

0°

45°

45°

Air Flow

c)

Air Flow

90°

30°

60°

0°

60°

30°

b)

Air Flow

44°

71°

90°

30°

60°

0°

a)

90°

30°

60°

0°

45°

45°

Air Flow

c)

Air Flow

90°

30°

60°

0°

60°

30°

b)

Air Flow

44°

71°

90°

30°

60°

0°

a)

(3) Application of Fouling Results

0

20

40

60

80

100

40 50 60 70 80

Air

mo

istu

re c

on

ten

t (g

/kg)

Air temperature (°C)

Outlet temperature

Significant deposition

initiates

Dryer exhaust

Rear section of heat exchanger

Front section of heat exchanger

High Fouling Region

Low Fouling Region

Curve dependent on velocity

57 °C

(4) Exhaust Heat Exchanger Design Problem

Heat Recovery Savings

High Air Flow Resistance

High Fouling

High air speed

Low air speed

Few tuberows

Max. Qr

No Qr

Many tube rows

Design

Plant Specific Analysis System Integration

Design Parameters

Air Velocity (HX Face Size)

Tube Diameter

Tube Spacing◦ Transverse, Longitudinal

Fin Dimensions◦ Height, Pitch, Thickness

Number of passes

Loop flow rate

Walmsley et al. (2014), CET (PRES’14)

Spreadsheet Optimisation Model

User-Defined Inputs

Coupled HX System Design◦ Face size (i.e. air velocity)

◦ Tube & fin dimensions

◦ Pass arrangement

◦ Pump, piping & Buffer tank

◦ Fan

Economic Parameters◦ Steam & electricity price

◦ Capital cost formula

◦ Discount rate

◦ Price inflation

◦ Production hours per annum

◦ Cost to clean

Fluid Flow Specifications◦ Exhaust air temperature, flow rate,

humidity

◦ Inlet air temperature, flow rate, humidity

◦ Intermediate fluid (loop) flow rate

Fouling & Cleaning Parameters◦ Powder concentration

◦ Particle size distribution

◦ Run length

◦ Wash length

◦ Model time step

Spreadsheet Optimisation Model

Fouling Prediction Model

Milk powder

particle size

distribution

Fouling and

cleaning

parameters

Calculate critical impact

angle

Calculate tube area

coverage

Calculate probability of

sticking

Estimate probability of

impact

Determine mass

deposited

Next particle size

Next time step

Next tube row

Estimate temperature

profile in HE

Estimate Rf and ΔP

Calculate payback,

NPV and IRR

Calculate input

particle size

distribution to row

Heat

exchanger

geometry

Process

stream data

Milk powder

particle size

distribution

Fouling and

cleaning

parameters

Calculate critical impact

angle

Calculate tube area

coverage

Calculate probability of

sticking

Estimate probability of

impact

Determine mass

deposited

Next particle size

Next time step

Next tube row

Estimate temperature

profile in HE

Estimate Rf and ΔP

Calculate payback,

NPV and IRR

Calculate input

particle size

distribution to row

Heat

exchanger

geometry

Process

stream data

Key Model Outputs

Average heat exchanger duty

Temperature profiles within a heat exchanger

Fouling over time◦ Mass deposited

◦ Thermal and hydraulic resistance over time

◦ Duty over time

Cost estimations

Economic Indicators (NPV, IRR, Payback)

Fouling Model Results

0

100

200

300

400

500

0.0

0.5

1.0

1.5

2.0

2.5

3.0

0 100 200 300 400 500 600 700

Exh

aust

pre

ssu

re d

rop

[P

a]

He

at r

eco

very

[M

W]

Time in operation [h]

CF CB EB

CF CB EB

Qr :

ΔP :

Effect of Number of Tube Rows

NPV IRR

-$1.5

-$1.0

-$0.5

$0.0

$0.5

$1.0

$1.5

$2.0

$2.5

$3.0

0 10 20 30 40

Ne

t P

rese

nt

Val

ue

[$

mill

ion

s]

Number of tube rows in the exhaust heat exchanger

CF - NPV CB - NPV EB - NPV

0%

20%

40%

60%

80%

100%

0 10 20 30 40

IRR

Number of tube rows in the exhaust heat exchanger

CF - IRR CB - IRR EB - IRR

Effect of Exhaust Air Velocity

$0.0

$0.5

$1.0

$1.5

$2.0

$2.5

$3.0

0 2 4 6 8 10

Ne

t P

rese

nt

Val

ue

[$

mill

ion

s]

Exhaust heat exchanger face velocity [m/s]

CF - NPV CB - NPV

EB - NPV

0%

20%

40%

60%

80%

100%

120%

0 2 4 6 8 10

IRR

Exhaust heat exchanger face velocity [m/s]

CF - IRR CB - IRR EB - IRR

NPV IRR

Site Specific Factors Affecting Economics

◦ Inlet air temperature (outside 15°C or inside 33°C)

◦ Inlet air absolute humidity (outside or inside)

◦ Exhaust air temperature (+5° → +25% HR)

◦ Bag filters (low powder conc.)

◦ Inlet and exhaust fan capacity (reduce cost by ~25%)

◦ Existing pre-heaters using utility

◦ Existing heat recovery to dryer inlet air

Site Specific Factors Affecting Economics

◦ Re-usable existing ducting (reduce cost by ~20%)

◦ Operating and production hours

◦ Price of energy (varies by 30 – 50%)

◦ Space

◦ Inlet air heater bottleneck

◦ Good attitude to change

10

100

1,000

10,000

100 1,000 10,000 100,000

Hea

t Tr

ansf

er t

o P

ress

ure

Dro

p R

atio

Reynolds Number

Effect of velocity

Effect of number of rows

Effect of longitudinal spacing

Effect of transverse spacing

Effect of fin pitch

Effect of fin height

Effect of fin thickness

Present Work: Design Optimisation

Overall Conclusions

• New Zealand milk powder plants can economically increase HR by ~20 %; exhaust heat recovery (HR) is the single largest opportunity

• Exhaust HR can play an integral part in supplying heat to neighbouring plants via HRLs

• For SMP, a final exhaust air temperature above 55 °C can minimise particulate fouling problems

• Exhaust HR is economic for good sites (payback time of ~1.6 years, NPV of NZ$2.9 million and IRR of 71 %)

•Site selection is important

Thesis Link & Any Questions?

Heat Integrated Milk Powder Production

By Tim Walmsley

http://hdl.handle.net/10289/8767

http://hdl.handle.net/10289/8767

spray dryer exhaust heat recovery

Documents