Hugo Gontijo

Application of image analysis in

flotation control

Hugo Montalvão GontijoHugo Montalvão Gontijo

CEMI – Process Technology and Engineering

Minas Gerais, Brazil

Objectives

Demonstrate tools applicability for advanced

process control and online image analysis in

flotation circuits

Direct Flotation

Reverse Flotation

Grade versus Recovery

Grade

Recovery

Vision Systems

There are systems that use image analysis to infer

characteristics of the froth.

OptVision Froth

Vision Systems Features

• OPC communication, allowing easy

communication with PLC, DCS, PIMS and APC

systems;

• IP Network Cameras;

• Multiple cameras processing simultaneously;

• Led Lighting.

Hardware

Network Camera with

IP 66 protectionLed lighting has a

lifetime of more than

50000 hours and low

energy consumption

Hardware

Advanced Process Control (APC)

PLC/DCS

APC

Control Room

Operator

Process

OptProcess

Control Strategic

• Objective

• Process Variables

• Manipulated Variables

• Make strategic:

IF (...)

THEN (...)

• Make strategic:

– interviews

– knowledge about process

Direct Flotation (case 1)

OptVision© = Cameramonitors froth’s speed

OptProcess© Strategy= Level setpoint is manipulated to control froth’s speed

Direct Flotation (case 1)

MetalurgicalLab Results

(concentrates) One every hour

OptProcess© Strategy= How fast the froth should flow + Mixture of reagents for

optimum recovery

Direct Flotation (case 1)OptVision© = Cameramonitors color of froth

OptProcess© Strategy= How fast the froth should flow + Mixture of reagents for

optimum recovery

color versus Lab Results (case 1)

18:00:00 6:00:00 18:00:00 6:00:00 18:00:00 6:00:00

0,58

0,6

0,62

0,64

40

42

44

46

48

50

Re

dco

lor

con

cen

tra

te(%

)

Relationship between Red color and grade in flotation froth (Raw Data)

0,5

0,52

0,54

0,56

30

32

34

36

38

18:00:00 0:00:00 6:00:00 12:00:00 18:00:00 0:00:00 6:00:00 12:00:00 18:00:00 0:00:00 6:00:00

Re

d

Gra

de

on

con

cen

tra

te

Time(h)

Grade Red Color Red Color MM15min

color versus Lab Results (case 1)

0,56

0,58

0,6

0,62

0,64

38

40

42

44

46

48

50

Grade (interpolated) and Red Moving Average (Moved forward in

time)

con

cen

tra

te(%

)

Re

dco

lor

0,5

0,52

0,54

0,56

30

32

34

36

38

18:00:00 0:00:00 6:00:00 12:00:00 18:00:00 0:00:00 6:00:00 12:00:00 18:00:00 0:00:00 6:00:00

Grade Interpolated Red Color Moving average (Moved forward)

Gra

de

on

con

cen

tra

te

Re

d

Time(h)

Direct Flotation (case 1)OptVision© = Cameramonitors color of froth

OptProcess© Strategy= Compares the color and the results: decides to either

darken or clear the froth.

MetalurgicalLab Results

(concentrates) One every hour

Results (case 1)

4

5

6

7

8

9

Fre

qu

en

ce (

%)

Grade of mineral in the concentrate

OFF

ON

0

1

2

3

4

35

35

,5 36

36

,5 37

37

,5 38

38

,5 39

39

,5 40

40

,5 41

41

,5 42

42

,5 43

43

,5 44

44

,5 45

45

,5 46

46

,5 47

47

,5 48

48

,5 49

49

,5 50

50

,5 51

51

,5 52

52

,5 53

53

,5 54

54

,5

Fre

qu

en

ce (

%)

Grade(%)

Results (case 1)

8

10

12

14

16

Fre

qu

en

ce(%

)

Grade of contaminat 1 in the concentrate

OFF

ON

0

2

4

6

1 1,1 1,2 1,3 1,4 1,5 1,6 1,7 1,8 1,9 2 2,1 2,2 2,3 2,4 2,5 2,6 2,7 2,8 2,9 3

Fre

qu

en

ce(%

)

Contaminat 1 grade(%)

Results (case 1)

10

12

14

16

18

Fre

qu

en

ce(%

)

Grade of contaminat 2 in the concentrate

OFF

ON

0

2

4

6

8

5 5,5 6 6,5 7 7,5 8 8,5 9 9,5 10 10,5 11 11,5 12 12,5 13 13,5 14 14,5 15

Fre

qu

en

ce(%

)

Contaminat 2 grade(%)

Results (case 1)

4

5

6

7

Fre

qu

en

ce(%

)

Mineral recovery

OFF

ON

0

1

2

3

Fre

qu

en

ce

Recovery (%)

ON

Results (case 1)

• Decrease standard deviation

• Increase recovery

• Increase production (10%)

Reverse Flotation (Case 2)

• Iron flotation

• Timbopeba and Fábrica Nova

• VALE (Mariana, Minas Gerais, Brazil)

Silva C. G., Miquelanti E., Silva M. C., Martins M. A. S., Santos S. C. R., Aplicação de

sistema especialista para controle das flotações de Fábrica Nova e Timbopeba, 41º

Seminário de Redução de Minério de Ferro e Matérias-primas e 12º Seminário

Brasileiro de Minério de Ferro, 12 a 26 de setembro de 2011, Vila Velha, ES.

Reverse Flotation (Case 2)

Fábrica Nova

Reverse Flotation (Case 2)

Timbopeba

Results (Case 2)

Results (Case 2)

Conclusions

• Advanced process control and online image

analysis in flotation circuits can increase recovery,

production and quality.

Projects

• Vale – Mina do Pico (ITM-I)

OptProcess/OptVision Froth (24 cameras)

• Vale – Vargem Grande

OptProcess/OptVision Froth (7 cameras)

• Vale – Timbopeba

OptProcess/OptVision Froth (19 cameras)OptProcess/OptVision Froth (19 cameras)

• Anglo American

OptProcess/OptVision Froth (9 cameras)

Projects

• Vale – Mutuca

OptProcess/OptVision Rock (4 cameras)

• Hydro – Paragominas

OptProcess/OptVision Rock (5 cameras)

• Vale – Salobo

OptVision Rock (6 cameras)OptVision Rock (6 cameras)

• Vale – TIG

OptVision Rock (1 camera)

Projects

• Votorantim Metais – Niquelândia

OptProcess

• Anglo American – Amapá

OptProcess

Projects

• Samarco – Ubu (Usina 3)

OptProcess/OptVision Pellet (3 cameras)

• Vale DIPE – Tubarão (Usinas I, II, III e IV)

OptVision Pellet (24 cameras)

• Vale DIPE – Vargem Grande

OptVision Pellet (12 cameras)OptVision Pellet (12 cameras)

• Vale DIPE – Fábrica

OptVision Pellet (7 cameras)

• Vale DIPE – São Luís

OptVision Pellet (12 cameras)