Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

BDE SC3.3 Workshop - Data management in WT testing and monitoring

85 views

Published on

Data management in WT testing and monitoring (Mr. Gorka Gainza, ARESSE) at the BigDataEurope Workshop, Amsterdam, Novermber 2017

Published in: Technology
  • Be the first to comment

  • Be the first to like this

BDE SC3.3 Workshop - Data management in WT testing and monitoring

  1. 1. Big Data Europe workshop DATA MANAGMENT CHALLENGES GORKA GAINZA TECHNICAL MANAGER ggainza@aresse.com +34 659 252 212
  2. 2. ARESSE ARESSE ENGINEERING: WHO ARE WE? ENGINEERING - LABORATORY DYNAMIC - VIBRATION – ACOUSTIC WIND INDUSTRY - ROTARY MACHINERY
  3. 3. ARESSE END OF 2007 MID 2016 END 2013  MORE THAN 250 ENGINEERING PROJECTS  60 NOISE IEC REPORTS  MILESTONES 9 ENGINEERS – based on a single location Managing / Customer Analysis and Reports Tool development 4 TECHNICIANS – travelling worldwide On Site Monitoring BEG. 2010 END 2012 ARESSE ENGINEERING: LITTLE BIT OF HISTORY… GROWTH 2013 – 12% 2014 – 20% 2015 – 27% 2016 – 25%
  4. 4. ARESSE ARESSE ENGINEERING: SOME OF OUR ADVENTURES…
  5. 5. ARESSE VALIDATION CAMPAINGS  TEST RIGS  PROTOTYPES TYPE TESTING  IEC 61400/11  GEARBOX TEST CONDITION MONITORING  CMS development  Data Management DYNAMICS VIBRATION ACOUSTICS  Modal caracterization and correlation  Vibration Monitoring Campaings Failure analysis  Vibrational fatigue test SUPPORT AT DESIGN AND VALIDATION TOOLS DEVELOPMENT FOR CUSTOMIZED REQUIREMENTS TRAINNING AND EDUCATION  FEM, fatigue calculation  System identification  Event Monitoring  Vibroacoustic simulations  Noise monitoring  Noise source identification ARESSE ENGINEERING: SOME OF OUR SERVICES
  6. 6. ARESSE  How can we get immidiate access to massive amount of data?  How can we be sure that we have measured correctly everything which is needed?  How can we deal with different sources of data?  How can we get maximum information out of the data?  How can we compare the data with patterns and references? BIG DATA VOLUME VELOCITY VARIETY VERACITY ARESSE ENGINEERING: CHALLANGES TO PROMOTE OUR SERVICES BUSINESS MODEL CUSTOMER LET’S DEVELOP A PLATFORM, CUSTOMIZED TO OUR PROCESSES, AND LET’S OPEN THE DATA AND THE TOOLS TO OUR CUSTOMERS
  7. 7. ARESSE  BIG DATA APPROACH MANIPULATING DATA vs LEARNING FROM DATA  DATA INTEGRATION PROTOTYPE + OPERATION  OPEN AND MULTIPORPUSE ARESSE ENGINEERING: HOW WE DECIDED TO WORK?
  8. 8. ARESSE XDAS ACQUISITION TOOL  Programmed in Labview for windows based platforms continous monitoring  Ability to synchronize different devices such as plc’s, dataloggers, third party products and our standard NI electronic  Data analysis during monitorization. Customize algorithms based on block diagram scheme. Filtering, Virtual channels, Statistics, FFT, order analysis.  Predefined templates and calibration for IEC tests and repetive projects  Storage Strategy: Relational Database linked to Raw data ARESSE ENGINEERING: SOME OF OUR TOOLS ON THE FIELD
  9. 9. ARESSE HIVEX EXTRACTION TOOL  Programmed in C#  Defined to transfer local data into the cloud. Database information is sent using web services (Json format). Raw data stored in tdms files is sent via FTP protocol  Customer is asked to open several ports to cloud reception service IP. The communication is stablished and the data transfer is authenticated.  To be able to analyze raw data as soon as it is generated a priorization strategy is implemented, and the end user can select the level of priority of each tdms file ARESSE ENGINEERING: SOME OF OUR CONNECTION TOOLS
  10. 10. ARESSE HIVE CLOUD STORAGE AND ANALYSIS TOOL  Storage Services. Data locally postprocessed is storaged in ElasticSearch linked to a file storage service for the raw data.  A Management Service to allow the user interaction with data is defined. Its persistance is based on MySQL  Analizer Service, so far developed in JAVA, has been implemented to execute over the data some algorithms including some IEC tests and other fault / Characterization tools.  Several Access Points to the cloud are defined: Reception Service, Web Access Web browser access implemented on AngularJS ARESSE ENGINEERING: SOME OF OUR CLOUD TOOLS
  11. 11. ARESSE CAMPAIGNS – WORKSPACES – DASHBOARDS – WIDGETS  Data Viewer, data check, classification, filtering, correlation…  Usefull for Continuous and Event monitoring ARESSE ENGINEERING: SOME OF OUR CLOUD TOOLS
  12. 12. ARESSE RAW DATA VIEWER AND ANALYZER.  Raw Data Viewer. Labelling. Reproduction  Tools for time and frequency domains  Downloading ARESSE ENGINEERING: SOME OF OUR CLOUD TOOLS
  13. 13. ARESSE DATABASE VIEWER. STATISTICAL REGRESSIONS  Data grouped for visualization  Bins based statistical parameters  Reporting and comparisons ARESSE ENGINEERING: SOME OF OUR CLOUD TOOLS
  14. 14. ARESSE IEC ANALIZER  Templates for algorithm parametrization  According to standard calculation procedures  Complete analysis and reporting ARESSE ENGINEERING: SOME OF OUR CLOUD TOOLS
  15. 15. ARESSE ARESSE ENGINEERING: SOME OF OUR CLOUD TOOLS DYNAMIC ANALIZER  Spectral and Angular domain sorting and averaging tools based on binarized data
  16. 16. ARESSE ARESSE ENGINEERING: DEVELOPING CHALLENGES MACHINE LEARNING  First approach using classification and regression algorithms to:  Identify and quantify different control versions in terms of noise  Identify the main vibration parameters producing abnormal dynamic behaviours  Bringing Phyton into the cloud PHYSICAL MODELS and VIRTUAL SENSORS  Kalman filters – Introducing into the monitoring loop the information of the physical parameters and dynamic response models to enhance the data quality
  17. 17. GORKA GAINZA TECHNICAL MANAGER ggainza@aresse.com +34 659 252 212 ARESSE ENGINEERING THANK YOU !!

×