AUTOMATION
MES for predictive maintenance
Classical maintenance schedules are time-based
at regular intervals. Prediktor Solar MES includes a
Predictive maintenance module (PdM). The PdM
module predicts the future trend of equipments’
condition. This approach uses principles of
statistical process control, multivariate
technologies and model based estimation
technology to determine at what point in the future
maintenance activities will be appropriate.
Examples of models for prediction of optimal time
periods for maintenance operations are:
ring6 Registration of operating time for a wafer saw
for prediction of replacement times for pulleys
and guides.
ring6 Monitoring of band saws by combining
measurements like forces, temperatures etc.
These variables are jointly used in a predictive
model to determine optimal time slices for saw
blade replacements. The models are our Process Intelligence tools based on modeling, Fig 6: Loss categories
continuously adapted based on saw run multivariat and statistical technologies. for three different
histories. plants
ring6 Proper use of PdM technologies will decrease As part of this focus, Prediktor also participates in
the downtime of the monitored equipment, and the newly established “The Norwegian Research
will thereby increase the average production Centre for Solar Cell Technology”. The
39
yield. collaborating parties are Prediktor, NTNU, SINTEF,
University of Oslo, Elkem Solar, Fesil Sunergy,
www
Some trends for MES in the PV industry Hydro, Norsun, REC, Scatec, Solar Cell Repower
.solar
Increased interest in MES solutions that offer single and Umoe Solar.
item tracking. Few systems do this today (Prediktor
-pv-management.com
Solar MES does). Single item tracking enables The main goal of the consortium is to provide
detailed yield analyses for production optimization. current and future actors in the Norwegian solar
cell industry access to state-of-the-art technological Fig 7: Example report
Massive automated data collection will be the rule. and scientific expertise. Total Centre budget is showing multiple data
The lesson from process industry is: “It is better to about €50 mill distributed over the centre’s eight streams
measure too much than too little.” Time series data planned years of duration.
will be seamlessly integrated with relational
Issue II 2009
oriented data.
Even better tools for analysis and modeling of
relations between production parameters and
resulting yield- and quality measurements will be
developed. This will enable steadily increase in
production performance. (Process Intelligence).
Value chain tracking for data analyses and
optimisation across the value chain will be
common. This type of tracking facilitates data
analysis from polysilicon to module production.
MES companies specializing in the PV industry will
grow and become stronger. Solar dedicated
vendors will make solar dedicated products.
Prediktor focuses heavily on being a part of this
future and is participating in several research
programs directed towards the solar industry. One
of these relates to model based optimization and
control of CZ-pullers, another to further develop
Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38  |  Page 39  |  Page 40  |  Page 41  |  Page 42  |  Page 43  |  Page 44