create knowledge from your best experiences
DESCRIPTION
create knowledge from your best experiences. D ata structuration. modélisation. Data restitution. extraction. PLC. Supervision. Fichiers Excel. LAB / ERP. validation. Data extraction. Data Transfert. automatisation. D ata Transfert. files.mybraincube.com. IP transfert. - PowerPoint PPT PresentationTRANSCRIPT
create knowledge from your best experiences
Data structuration
Supervision LAB / ERPPLC Fichiers Excel
modélisation
extraction
automatisation
validation
Data extraction
Data restitution
Data Transfert
Data Transfert
PLC
Supervision
Excel
LAB / ERP
files.mybraincube.com
Convert into cs
v
(transfo
_xls_csv
)
IP transfert
SFTP on port 22 or 443
Firewall
Your jailled folder
All csv files ready to
sent
All the tool you need, IP Transfert, Transfo_xl_csv and this presentation are available on: http://www.ipleanware.com/dl/
Data restitution Real signal
2 hours
Valu
esReal variable variations
For real signal you might have Several types of
restitution
or
Valu
es
2 hours
Numerical Discrete or binary
Data restitution regular registration
Valu
es
One point every 5 minutes
Valu
es
One point every 5 minutes but filtrered
filter
Take care of the filter because all data are not extracted
High / regular frequency registration (process data)
One point every 5 minutes
Valu
es
Numerical data :
Discrete or binary data :
Valu
es
Value at precise time when status changes
Data restituted by the system between precise data musn’t be extracted
Data restitution non regular registration
Low frequency and/or non regular registration (quality data)
Valu
es
One value every 30 minutes during 2 hours
Valu
es
Data are interporlated between change (dangerous for braincube)
What do we do with these type of extraction :- the average over a period ?- to fill the values between changes?
9:1
7 9:3
2 9:5
4
Separate files for different data typeOrganization
Team…
Process
EventsBreakStop
Clothing …
Genealogy
Quality
Weather
Traceability
Separate files for different frequencies
High frequency registrationFor example : Process data
Low frequency and/or non regular registration For example : Quality data
Event dataWeather dataGenealogy dataOrganization dataTraceability data
Matrix files Advantage : small files
Disadvantage : extraction might be more complicated
File example with data every 5 minutes:
sample value every 5‘or average value over 5'
Tag name
Values
List files Advantage : easy extraction
Disadvantage: large files
File example with data every 5’:
Value every 5’ for high frequency registration (or at precise timestamp for non regular registration)
1st Column = Timestamp2nd Column = Tag name3rd Column = Value
Data files Files names : name_of_the_file_YYYYMMDD_HHMMSS
Files format: a) .csv with delimiters like « ; »b) .txt
Example : reporting_shift_20110525_134520 .csv
for the reporting_shift file of the 25th of May 2011 at 13:45:20
It is forbidden to use any special characters (é ~ ç ?...) or spaces
The frequency of extraction will be defined with production people.High frequency (ie: minute) allows a real-time jobs analysis
Mandatory No special characters or space in file names A file per type of data (process data, events, …) Each type of files contain always the same tags Never change the date format in the files You need to know if filters are applicated on the values before extraction Send files with recovery when all data don’t arrived at the same time (data will be overwrited by the last received)Never Change the file format.If the delimiter is « ; » note that you will need to eliminate the ; in text cells. Validate files over a deined period of time with IP Leanware before to run the historical extraction, this will include:
a) File nameb) File formatc) List of tagsd) Date format in the file