Increase Current Efficiency of Potline 3 (P/L-3)
Introduction Dubai Aluminium Company Limited (DUBAL) •
Based in Jebel Ali, Dubai
•
Annual Production of > 1 million tonnes of aluminium
•
>4000 Employees
•
8 Potlines consisting of 1573 aluminium cells
•
One of the few smelters in world to produce primary high purity metal for use in electronics and aerospace industries.
2
Project Background • Smelting converts alumina (ore) into aluminum metal through electrolysis process ― By using Direct Current (DC) ― Current Efficiency (CE) is the ratio of electrical direct current that results in actual metal production • Therefore, improvement in Current Efficiency remains one of the strategic objectives of any Aluminium smelter
3
Define Phase
1. Define
2. Measure
3. Analyze
4. Improve
5. Control
Problem Statement: •
Potline 3 Current Efficiency is at 92.9% for H1 2009 which is below target since increase of current amperage to 200 kA,
•
resulting in decreased plant hot metal output.
Project Target: Increase average Potline 3 current efficiency to target of 93.1% for 2010.
4
1. Define
2. Measure
3. Analyze
4. Improve
Project Scope: In Scope:
Out of Scope:
Potline 3 Process Parameters and Procedures
All other Potlines
Unit of measurement:
Potline 3 Current Efficiency
Operational Definition:
Monthly Average CE from iRPMS (Smelting Database System) 5
5. Control
1. Define
2. Measure
3. Analyze
4. Improve
5. Control
Team Charter: S.No
Name
Functional Role
Project Responsibilities
1
Daniel Whitfield
Manager – Projects D18
Team Leader.
2
Andries Louw
Snr Process Control Engineer – Potrooms
Data analysis and report making.
3
Mohamed Tawfik Boraie
H.O.D: PC-CL
Data analysis and report making.
4
Saif Mohamed
Snr. Planner – Prodn Services
Data analysis and report making.
5
Najeeba Al-Jabri
Snr Manager
Data collection and implementation of solutions.
6
Tariq Majeed
Supt Potroom Operations
Data collection and implementation of solutions.
7
Devadiga H.R.
Act, Manager, Line 3, 7 & 9
Data collection and implementation of solutions.
8
Maryam Al-Jallaf
Snr. Manager - PC PR & CL
Data collection and implementation of solutions.
9
Adam Sherrif
Snr Process Control Engineer – Potrooms
Data collection and implementation of solutions.
6
1. Define
2. Measure
3. Analyze
4. Improve
5. Control
Stake Holder Model : ARMI Chart Approver VP-Smelter Ops.
Resource
Member
Interested Party
Manager D-18
Andries Louw
VP-Casthouse
P/L-3 Superintendent
Mohamed Tawfik Boraie
VP-Marketing
P/L-3 Technicians
Saif Mohamed
VP-Finance
P/L-3 Operators
Najeeba Al Jabri
VP-Power & Desal.
P/L-3 Process Technician
Tariq Majeed Devadiga H.R. Maryam Al-Jallaf Adam Sheriff
7
1. Define
2. Measure
Project Schedule
8
3. Analyze
4. Improve
5. Control
Measure Phase
1. Define
2. Measure
3. Analyze
4. Improve
5. Control
•
Current efficiency is key measure of process performance, and is regularly reported and monitored
•
It is difficult to be measured directly; therefore, inferred from actual metal production as below:
Current Efficiency
•
Actual Hot Metal Production = -------------------------------------Theoretical Hot Metal Production
Three months average taken to ensure reasonable accuracy of the data
9
1. Define
2. Measure
3. Analyze
4. Improve
5. Control
Measurement System Analysis •
Actual Metal Production =Total Weight of Metal delivered to Casthouse Casthouse scales regularly calibrated and checked – Verified Calibration Records and is OK
•
Theoretical Metal Production = f (Amperage supplied by Power Plant) Power Plant amperage supply tested on monthly basis – Found Ok
•
Review of existing plant system for measuring and reporting current efficiency showed no significant concerns over accuracy or precision
Measurement System – Found Satisfactory
10
Analyze Phase
1. Define
2. Measure
3. Analyze
Current Efficiency – Back Reaction Flow Chart Carbon Dusting
Anode Spike
Cell Overfeeds
Excessive Sludge
5. Control
Tool (s) Applied:
Current bypasses electrolyte
Cell Becomes Unstable
4. Improve
-
Process Flow Diagram
Excess heat generation
Low Bath Temp / High AlF3
Al solubility in bath increases
High Bath Temp / Low AlF3
Cell Underfeeds
Excessive Anode Effects
Bath Height Too Low
Excessive Anode Airburn
BRSP Set too low
Cell ACD Reduced
Al mixes back into electrolyte
Oxidation of Al to Al2O3
Low Current Efficiency
Possible causes for low Current Efficiency
Bath temperature/AlF3 Age Anode Effects Alumina Feeding Metal and Bath height Base Resistance Set Point (BRSP) Noise/stability 11 Operational problems Anode Problems
1. Define
2. Measure
3. Analyze
4. Improve
5. Control
Tool (s) Applied: Multi-Variable Linear Regression • In multi-variable linear regression, there are several independent variables up to N. Yi = βo + β1xi + β2xi + … where
i = 1, …. N.
• P-value shows the significance of the correlation (p-value of 0.05 = 95% statistical significance or confidence). As much P-value closer to 0.0 as much as the parameter is statistically significant
• Strongest correlation between bath temperature and AlF3 (interrelated variables) Needs more investigation
• Age refers to life of reduction cell, and hence it is an unassignable cause
Variable Bath Temperature AlF3 Age AEF TRSP UF Duration Time Unstable Dumps BRSP Metal Height Bath Height Average Resistance Volts 12 Noise
P-value 0.000 0.034 0.038 0.066 0.094 0.275 0.305 0.451 0.537 0.554 0.583 0.608 0.736 0.746
1. Define
2. Measure
3. Analyze
4. Improve
5. Control
Bath Temperature •
Based on accumulative experience, it is proven that increase of 5oC in bath temperature can lead to 1% drop in current efficiency 976
Linear regression covers the relationship:
974
CE = (-0.2259 x bath temperature) + 309.13 (R2= 0.9543)
Bath Tem perature (C)
972 970 968 966 964 962 960 84
86
88
90
92
94
Current Efficiency (%)
96
98 13
100
1. Define
2. Measure
3. Analyze
4. Improve
5. Control
Base Resistance Set Point (BRSP) •
Relationship between BRSP/ACD and CE well established
• Initial analysis looked BRSP and CE. No big correlation above ~14.75 µ
• Some correlation < 14.75 µ 100
C u rren t E fficien cy (% )
98 96 94 92 90 88 86 84 13.5
14.0
14.5
15.0
15.5
BRSP (micro-ohms)
14
16.0
16.5
1. Define
2. Measure
3. Analyze
4. Improve
5. Control
Poor Performing Cells Tool (s) Applied: Cumulative chart 100%
Loss of Current Efficiency (% )
90% 80%
~20% of cells represent 47% of total CE loss (Actual CE – Target CE).
70% 60% 50% 40% 30% 20% 10% 0% 0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
Number of Cell (%) 15
100%
1. Define
2. Measure
3. Analyze
4. Improve
5. Control
Validated Root Causes/ Parameters
×
Root Cause 1: High Bath Temperature
Root Cause 2: Low Base Resistance Set Point (BRSP)
Age – Life of Cell: Difficult to address this cause
16
Improve Phase
1. Define
2. Measure
3. Analyze
4. Improve
5. Control
Root Cause 1: Action plan for high bath temp and low CE pots S. No.
1
Action Point
Responsib ility
Completion Date
Adjust Bath Chemistry to improve Current Efficiency
Daniel Whitfield
Dec. 2009
Current Efficiency Vs Bath Temp 95.0 y = 0.052x + 92 R2 = 0.6879
966
94.0
964
93.5 962 93.0 960 92.5 92.0 91.5
958 Bath Temperature Current Efficiency Linear (Current Efficiency)
956
91.0
954 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 Week No. (2009 - 2010)
1
3
17
Bath Temp (°C)
Current Efficiency (%)
94.5
968
1. Define
2. Measure
3. Analyze
4. Improve
5. Control
Root Cause 2: Action plan for low Base Resistance Set Point (BRSP) • •
Established Control Limits so that BRSP not to be lowered below 14.5 µ without careful monitoring of the current efficiency Critical BRSP Limit of 14.75 µ.
•
Increased the BRSP in low CE/BRSP cells
•
Example of “action plan for implementation” as a result of weekly meetings is given below
Cell 146 198 271 149 117 102
CE - 4wks CE - 16wks CE - 52wks Action Person Target Date Increase BRSP by 0.1µcheck dumpweight AS/MSW 07/01/2010 89.3 90.6 92.9 90.3 91 92 Check BFT, Cu tab and dumpweight MSW 10/01/2010 92.5 91 91.9 Improving in last 28 days, no action 20/01/2010 Increase BRSP by 0.05µcheck Cu tab 89.3 91.1 93 MSW 07/01/2010 92 91.2 90 Under Fe attack FM NA Increase BRSP by 0.1µ 93 91.3 91.5 AS 07/01/2010
• Average increase of 0.34 µ in 25 cells, average increase of 1.5% CE
18
1. Define
2. Measure
3. Analyze
4. Improve
5. Control
Current efficiency after improvement actions 95.5
201 Project Start
200
94.5
199
94.0
198
93.5
197
93.0
196 Monthly CE
92.5
3-month Running Average Target CE
92.0
195 194
Amperage 91.5 7 -0 v No
193 08 n Ju
08 c De
9 -0 l Ju
10 n Ja
10 g Au 19
11 b Fe
11 p Se
Amperage (kA)
Current Efficiency (%)
95.0
Control Phase
1. Define
2. Measure
3. Analyze
5. Control
4. Improve
System established for identifying and improving poor CE cells •
List of poor performing cells in Potline 3 developed, updated and released on weekly basis for setting up proper action plans.
•
Work Instruction was developed to diagnose and action poor performing cells Before the project
After the Project
Summary forof L3L3 CE, June 2009 Distribution CE, June 2009
Summary for 2010 Distribution ofL3 L3CE, CE,Feb Feb 2010
84
Sample size Mean STD. Dev.
87
90
93
96
86
88
227
227
Difference
92.32 %
93.70 %
+1.38%
2.47 %
1.71 %
-0.76%
90
20
1st
Quartile
91.4 %
92.7 %
+1.3%
92
94
96
98
1. Define
2. Measure
3. Analyze
4. Improve
5. Control
•
Potline 3 monitoring on daily basis by Potline Engineers and Technicians through potroom monitoring and reporting system (Smelter Analytics)
•
Fine-tuning and changes to pots operating targets
21
C u rre n t E f f ic ie n c y ( % )
Project Success & Benefits 94.2 94 93.8 93.6 93.4 93.2 93 92.8 92.6 92.4 92.2
Actual CE% Target CE%
Increased average Potline 3 current efficiency
Improved overall Potline performance 2009, 2nd Half
2010, 1st Half
2010, Full year
2011, YTD
Project yielded re-occurring financial benefits of AED 1.47 millions per annum 22
Project Closure Learnings and Roll-over: •
Documentation of the project report
•
Use of statistical tools and gained better understanding w.r.t. Smelting Process
•
Roll-over of the successful initiatives from projects – to sister Potline 1 – Achieved similar increase in current efficiency
Recognitions: •
All team members received gift and cash award
•
Project selected for Share Best Practice Session – to 200+ employees
•
Nominated for CII Symposium 23
Why this Project is an Excellent Improvement Example? •
Achieved one of the best current efficiencies in D-18 type of cell design (at higher Amperage of 200kA)
•
Combination of technical as well as statistical methods by using DMAIC approach
•
Project experiences rolled-over to Potline of similar cell design and resulted in improvements
•
Quantum contribution to company’s process performance
•
Environmentally beneficial
24
Thank you
25
