SlideShare a Scribd company logo
1 of 38
INDUSTRY 4.0 @ GHENT UNIVERSITY
(GLOBAL CAMPUS) - IMEC:
FROM PREDICTIVE MAINTENANCE TO
PERFORMANCE MONITORING
October 23-26, 2018 @ Gangneung, Korea
The 5th Aslla Symposium
Wesley De Neve, Sofie Van Hoecke, Erik Mannens
OUTLINE
• Background
• Use cases
‒ use case 1: wind turbine condition monitoring
‒ use case 2: incipient bearing fault detection
‒ use case 3: anomaly detection in steel coil production
‒ use case 4: greenhouse condition monitoring
2
BACKGROUND
3
WESLEY DE NEVE
• Academic credentials
̶ Master’s degree in computer science (2002)
 from Ghent University, Belgium
̶ Ph.D. degree in computer science engineering (2007)
 from Ghent University, Belgium
̶ post-doctoral experience (2007-2011)
 at ICU / KAIST, Korea
• Employment: Professor
̶ IDLab, Ghent University, Belgium (2011 - current)
̶ Ghent University Global Campus, Korea (2014 - current)
4
SOFIE VAN HOECKE
• Academic credentials
̶ Master’s degree in computer science engineering (2003)
 from Ghent University, Belgium
̶ Ph.D. degree in computer science engineering (2009)
 from Ghent University, Belgium
• Employment: Professor
̶ University College West-Flanders (2011 - 2013)
̶ IDLab, Ghent University-imec, Belgium (2013 - current)
5
ERIK MANNENS
• Academic credentials
̶ Master’s degree in electro-mechanical engineering (1992)
 from KAHO Ghent, Belgium
̶ Master’s degree in computer science (1995)
 from KU Leuven, Belgium
̶ Ph.D. degree in computer science engineering (2011)
 from Ghent University, Belgium
• Employment: Professor & Research Valorization Director
̶ IDLab, Ghent University-imec, Belgium (2005 - current)
6
ORGANIZATIONAL OVERVIEW
7
GHENT UNIVERSITY
• Public university
‒ located in Ghent, Belgium
‒ established in 1817
8
GHENT UNIVERSITY
• Counts 42,000 students and 9,000 staff members
‒ about 4,000 foreign students and 800 foreign staff members
• Consists of 11 faculties, counting 117 departments
‒ campus buildings are distributed all over the city
9
Congress Center
‘Het Pand’
Faculty of Engineering
and Architecture
Aula Academia
GHENT UNIVERSITY GLOBAL CAMPUS
10
Biotech campus located in Songdo, Incheon, Korea (+250 students)
(part of the Incheon Global Campus, together with SUNY Stony Brook, GMU, & the University of Utah)
• R&D hub in nanoelectronics and digital technologies
‒ brings together 3,500 scientists and engineers from over 70 countries
• R&D groups in Belgium (headquarters: Leuven), the Netherlands,
Taiwan, USA, China, and offices in India and Japan
11
SOTA cleanroom
of 1 billion EURO
(in collaboration with
Samsung, a/o)
IMEC
IDLAB
• Internet Technologies and Data Science Lab
‒ founded on October 1, 2016
 merger of four already existing labs
‒ headcount
 40 professors
 50 post-doctoral researchers
 200 researchers
 15 support staff members
12
iGent Tower & AA Tower, Ghent, Belgium
IDLAB
• Education
‒ responsible for teaching more than 100 courses
• Fundamental and applied research
‒ machine learning and data mining
‒ semantic intelligence
‒ multimedia processing
‒ distributed intelligence for IoT
‒ cloud and Big Data infrastructures
‒ wireless and wired networking
13
Google DeepMind
(+ Ghent University)
four of our graduates are now working at
Google DeepMind and one at Google Brain
Sedol Lee
Google DeepMind Challenge Match
in March 2016 @ Seoul, Korea
verdict: 4–1 for AlphaGo
IDLAB
14
INDUSTRY 4.0 PLATFORMS
• Machines & Factories (M&F)
‒ an industrial research-oriented platform from
Ghent University linked to the manufacturing industry
‒ groups the activities of about 125 researchers from
different research groups, including 25 professors
• Flanders Make
‒ strategic research center for the manufacturing industry
in Flanders, set up by the Flemish government
‒ counts 400 researchers
15
PREDICTIVE MAINTENANCE &
PERFORMANCE MONITORING
16
MAINTENANCE STRATEGIES
• Preventive maintenance
‒ periodically inspect and replace components
• Reactive maintenance
‒ if something breaks, replace it
• Predictive maintenance
1) predict if, and possibly when, something is going to break
2) predict degradation of components
3) predict incipient faults
17
MAINTENANCE STRATEGIES
• Preventive maintenance
‒ periodically inspect and replace components
• Reactive maintenance
‒ if something breaks, replace it
• Predictive maintenance
1) predict if, and possibly when, something is going to break
2) predict degradation of components
3) predict incipient faults
18
USE CASE 1
19
WIND TURBINE CONDITION MONITORING
20
• Offshore maintenance is expensive
‒ specialized equipment is typically needed for landing and working at
offshore infrastructure
• Offshore maintenance is difficult to schedule
‒ depends on weather and wave conditions
WIND TURBINE CONDITION MONITORING
21
• Operation & maintenance actions may account for ± 30% of
the total production cost of offshore electricity in wind farms
• Early fault detection highly needed
‒ mitigates fault propagation and higher costs when detected (too) late
PERFORMANCE MONITORING
22
• Create a model that can predict the power output of a wind
turbine given several inputs
Predictive model
Predicted power
output
• Wind speed
• Wind direction
• Rotation speed
• Yaw
• Pitch
data collected from 3 wind turbines
over a time span of 3 months
(10 weeks of training data and 2 weeks of validation data)
random forests
ANOMALY DETECTION IN PRODUCTION
23
• Abnormal operation
‒ difference in predicted output
and actual output
‒ monitored through dashboard
Predictive model
Predicted power
output
• Wind speed
• Wind direction
• Rotation speed
• Yaw
• Pitch
Detection module
Actual power
output
Power production monitoring (Turbine 13)
USE CASE 2
24
INCIPIENT BEARING FAULT DETECTION
25
• Bearings are in all kinds of machines
‒ conveyer belts, wind turbines, cars, …
• Goal
‒ to detect incipient faults so to prevent faults
from escalating
INCIPIENT BEARING FAULT DETECTION
26
Time
Inadequate
lubrication
Pitting
Hard
particles Bearing
blockage
Component
failure
Machine
failure
• Goal: to detect condition of a bearing at this point in time
‒ maintenance can be scheduled accordingly and
reactive maintenance can be prevented
INCIPIENT BEARING FAULT DETECTION
27
Captured (1) vibration data
and (2) thermal imaging
data from bearings in
different conditions
Data set
Training set
Feature
extraction
Learning
algorithm
Model Predictions
INCIPIENT BEARING FAULT DETECTION
28
By using different machine learning algorithms (SVM, RDF, CNN)
for analyzing vibration data and infrared video,
we can automatically detect incipient faults with a high accuracy
Model
Inadequate
lubrication
Raceway
fault
Imbalance
Hard particles
INCIPIENT BEARING FAULT DETECTION
• Infrared imaging analysis
̶ handcrafted features + SVM: accuracy of 88.25%
• Vibration data analysis
̶ handcrafted features + RDF: accuracy of 87.25%
̶ CNN (end-to-end learning): accuracy of 91.77%
• Use of a multimodal approach (fusion)
̶ accuracy of over 99%
29
USE CASE 3
30
ANOMALY DETECTION IN STEEL MILL
31
• Any broken weld leads to
interrupts of several minutes
‒ comes with a high cost
• Fault prediction and cause
detection based on static and
dynamic parameters
‒ temperature
‒ pressure
‒ thickness
‒ speed
Metric GBRT RCN
True pos. 40 46
False neg. 10 4
False pos. 184 379
True neg. 558 363
Accuracy 77,60 % 70,46 %
USE CASE 4
32
SMART FARMING
33
? Where will we find enough food for 9 billion people by 2050?
Farming 4.0: use sensors to make farms more “intelligent” and
connected, so to be able to collect data about crop yields, soil-
mapping, fertilizer applications, weather, and plant growth
Collected data allow for improved decision-making
→
→
GREENHOUSE CONDITION MONITORING
34
condition monitoring in automated / programmable greenhouses (food computers)
plants on a conveyer belt
RGB imaging
hyperspectral imaging
adjustment of conditions
GREENHOUSE CONDITION MONITORING
35
• Drone-based phenomics
̶ measurement of physical and bio-chemical
traits of organisms as a function of genetic
and environmental changes
̶ hypothesis: gives less stress to plants
• Research questions
̶ what is (not) possible?
̶ what is the role of (deep) data analytics?
̶ how to scale to a real-world farm?
Indoor farming expert says Korea, Japan need to up investment
The Korea Herald 11/2017
36
4. Point cloud
matching
6. Determine
skeleton
5. Segmentation
organ
recognition
Structure-from-
Motion
Growing
Neural Gas
Convolutional Neural Networks
Point Feature
Histograms (PFH)
descriptors
7. Estimate
phenotypic
parameters
Conditional
Random Fields
3. Filtering
denoising
2. MVS-SFM
point cloud
1. Image
sequence
GREENHOUSE CONDITION MONITORING
TAKE-AWAY MESSAGES
• IDLab has 10-15 professors working on machine learning,
together with 50-100 pre- and post-doctoral researchers
̶ on a wide range of use cases
̶ on different types of (sensor) data
• Open to different types of R&D collaboration
̶ imec
̶ Ghent University
̶ Ghent University Global Campus
37
Thank you for your attention!
{wesley.deneve, sofie.vanhoecke, erik.mannens}@ugent.be

More Related Content

Similar to The 5th Aslla Symposium

Intefix_webinar
Intefix_webinarIntefix_webinar
Intefix_webinarI4MS_eu
 
Photovoltaic Module Energy Yield Measurements: Existing Approaches and Best P...
Photovoltaic Module Energy Yield Measurements: Existing Approaches and Best P...Photovoltaic Module Energy Yield Measurements: Existing Approaches and Best P...
Photovoltaic Module Energy Yield Measurements: Existing Approaches and Best P...Leonardo ENERGY
 
[DSC Adria 23]Hrvoje Novak & Slaven Begovic - Nowcasting: AI-based short-term...
[DSC Adria 23]Hrvoje Novak & Slaven Begovic - Nowcasting: AI-based short-term...[DSC Adria 23]Hrvoje Novak & Slaven Begovic - Nowcasting: AI-based short-term...
[DSC Adria 23]Hrvoje Novak & Slaven Begovic - Nowcasting: AI-based short-term...DataScienceConferenc1
 
Polymer Processing Cost Saving Workshop - 06 Ceramicx
Polymer Processing Cost Saving Workshop - 06 CeramicxPolymer Processing Cost Saving Workshop - 06 Ceramicx
Polymer Processing Cost Saving Workshop - 06 CeramicxInvest Northern Ireland
 
The Smart Utility Monitoring System for Industrial Safety and Enhancement on ...
The Smart Utility Monitoring System for Industrial Safety and Enhancement on ...The Smart Utility Monitoring System for Industrial Safety and Enhancement on ...
The Smart Utility Monitoring System for Industrial Safety and Enhancement on ...Amon Olimov
 
HYPERWIND Project: global and systemic monitoring of offshore renewable power...
HYPERWIND Project: global and systemic monitoring of offshore renewable power...HYPERWIND Project: global and systemic monitoring of offshore renewable power...
HYPERWIND Project: global and systemic monitoring of offshore renewable power...Jean-Claude Meteodyn
 
SURVEY ON HYDRO GENERATOR INSTRUMENTATION AND MONITORING
SURVEY ON HYDRO GENERATOR INSTRUMENTATION AND MONITORING SURVEY ON HYDRO GENERATOR INSTRUMENTATION AND MONITORING
SURVEY ON HYDRO GENERATOR INSTRUMENTATION AND MONITORING Power System Operation
 
Ghent University Global Campus 101
Ghent University Global Campus 101Ghent University Global Campus 101
Ghent University Global Campus 101Wesley De Neve
 
Yield enhancement
Yield enhancementYield enhancement
Yield enhancementLen Mei
 
Solar energy ongoing project by Magda Cayón
Solar energy ongoing project by Magda CayónSolar energy ongoing project by Magda Cayón
Solar energy ongoing project by Magda CayónASCAME
 
Designing Facility Layout and Effective Material Handling for a Ceramic Insul...
Designing Facility Layout and Effective Material Handling for a Ceramic Insul...Designing Facility Layout and Effective Material Handling for a Ceramic Insul...
Designing Facility Layout and Effective Material Handling for a Ceramic Insul...IRJET Journal
 
Show & Tell - Data & Digitalisation, Weather & Predictive Analytics.pdf
Show & Tell - Data & Digitalisation, Weather & Predictive Analytics.pdfShow & Tell - Data & Digitalisation, Weather & Predictive Analytics.pdf
Show & Tell - Data & Digitalisation, Weather & Predictive Analytics.pdfSIFOfgem
 
Dissertation Presentation
Dissertation PresentationDissertation Presentation
Dissertation PresentationShakeel Ghani
 
LEGaTO: Low-Energy Heterogeneous Computing Use of AI in the project
LEGaTO: Low-Energy Heterogeneous Computing Use of AI in the projectLEGaTO: Low-Energy Heterogeneous Computing Use of AI in the project
LEGaTO: Low-Energy Heterogeneous Computing Use of AI in the projectLEGATO project
 
COAL MINE SAFETY INTELLIGENT MONITORING BASED ON WIRELESS SENSOR NETWORK
COAL MINE SAFETY INTELLIGENT MONITORING BASED ON WIRELESS SENSOR NETWORKCOAL MINE SAFETY INTELLIGENT MONITORING BASED ON WIRELESS SENSOR NETWORK
COAL MINE SAFETY INTELLIGENT MONITORING BASED ON WIRELESS SENSOR NETWORKIRJET Journal
 
Applications of Deep Learning in Construction Industry
Applications of Deep Learning in Construction IndustryApplications of Deep Learning in Construction Industry
Applications of Deep Learning in Construction IndustryGaurav Verma
 
CD 2016 March - Advisian Showcase
CD 2016 March - Advisian ShowcaseCD 2016 March - Advisian Showcase
CD 2016 March - Advisian ShowcaseComit Projects Ltd
 

Similar to The 5th Aslla Symposium (20)

Intefix_webinar
Intefix_webinarIntefix_webinar
Intefix_webinar
 
Photovoltaic Module Energy Yield Measurements: Existing Approaches and Best P...
Photovoltaic Module Energy Yield Measurements: Existing Approaches and Best P...Photovoltaic Module Energy Yield Measurements: Existing Approaches and Best P...
Photovoltaic Module Energy Yield Measurements: Existing Approaches and Best P...
 
01-Role_of_NIWE.pdf
01-Role_of_NIWE.pdf01-Role_of_NIWE.pdf
01-Role_of_NIWE.pdf
 
[DSC Adria 23]Hrvoje Novak & Slaven Begovic - Nowcasting: AI-based short-term...
[DSC Adria 23]Hrvoje Novak & Slaven Begovic - Nowcasting: AI-based short-term...[DSC Adria 23]Hrvoje Novak & Slaven Begovic - Nowcasting: AI-based short-term...
[DSC Adria 23]Hrvoje Novak & Slaven Begovic - Nowcasting: AI-based short-term...
 
Polymer Processing Cost Saving Workshop - 06 Ceramicx
Polymer Processing Cost Saving Workshop - 06 CeramicxPolymer Processing Cost Saving Workshop - 06 Ceramicx
Polymer Processing Cost Saving Workshop - 06 Ceramicx
 
The Smart Utility Monitoring System for Industrial Safety and Enhancement on ...
The Smart Utility Monitoring System for Industrial Safety and Enhancement on ...The Smart Utility Monitoring System for Industrial Safety and Enhancement on ...
The Smart Utility Monitoring System for Industrial Safety and Enhancement on ...
 
HYPERWIND Project: global and systemic monitoring of offshore renewable power...
HYPERWIND Project: global and systemic monitoring of offshore renewable power...HYPERWIND Project: global and systemic monitoring of offshore renewable power...
HYPERWIND Project: global and systemic monitoring of offshore renewable power...
 
CIEMAT. Plataforma Solar de Almeria
CIEMAT. Plataforma Solar de AlmeriaCIEMAT. Plataforma Solar de Almeria
CIEMAT. Plataforma Solar de Almeria
 
SURVEY ON HYDRO GENERATOR INSTRUMENTATION AND MONITORING
SURVEY ON HYDRO GENERATOR INSTRUMENTATION AND MONITORING SURVEY ON HYDRO GENERATOR INSTRUMENTATION AND MONITORING
SURVEY ON HYDRO GENERATOR INSTRUMENTATION AND MONITORING
 
BATCH-7 MJT.pptx
BATCH-7 MJT.pptxBATCH-7 MJT.pptx
BATCH-7 MJT.pptx
 
Ghent University Global Campus 101
Ghent University Global Campus 101Ghent University Global Campus 101
Ghent University Global Campus 101
 
Yield enhancement
Yield enhancementYield enhancement
Yield enhancement
 
Solar energy ongoing project by Magda Cayón
Solar energy ongoing project by Magda CayónSolar energy ongoing project by Magda Cayón
Solar energy ongoing project by Magda Cayón
 
Designing Facility Layout and Effective Material Handling for a Ceramic Insul...
Designing Facility Layout and Effective Material Handling for a Ceramic Insul...Designing Facility Layout and Effective Material Handling for a Ceramic Insul...
Designing Facility Layout and Effective Material Handling for a Ceramic Insul...
 
Show & Tell - Data & Digitalisation, Weather & Predictive Analytics.pdf
Show & Tell - Data & Digitalisation, Weather & Predictive Analytics.pdfShow & Tell - Data & Digitalisation, Weather & Predictive Analytics.pdf
Show & Tell - Data & Digitalisation, Weather & Predictive Analytics.pdf
 
Dissertation Presentation
Dissertation PresentationDissertation Presentation
Dissertation Presentation
 
LEGaTO: Low-Energy Heterogeneous Computing Use of AI in the project
LEGaTO: Low-Energy Heterogeneous Computing Use of AI in the projectLEGaTO: Low-Energy Heterogeneous Computing Use of AI in the project
LEGaTO: Low-Energy Heterogeneous Computing Use of AI in the project
 
COAL MINE SAFETY INTELLIGENT MONITORING BASED ON WIRELESS SENSOR NETWORK
COAL MINE SAFETY INTELLIGENT MONITORING BASED ON WIRELESS SENSOR NETWORKCOAL MINE SAFETY INTELLIGENT MONITORING BASED ON WIRELESS SENSOR NETWORK
COAL MINE SAFETY INTELLIGENT MONITORING BASED ON WIRELESS SENSOR NETWORK
 
Applications of Deep Learning in Construction Industry
Applications of Deep Learning in Construction IndustryApplications of Deep Learning in Construction Industry
Applications of Deep Learning in Construction Industry
 
CD 2016 March - Advisian Showcase
CD 2016 March - Advisian ShowcaseCD 2016 March - Advisian Showcase
CD 2016 March - Advisian Showcase
 

More from Wesley De Neve

Towards diagnosis of rotator cuff tears in 3-D MRI using 3-D convolutional ne...
Towards diagnosis of rotator cuff tears in 3-D MRI using 3-D convolutional ne...Towards diagnosis of rotator cuff tears in 3-D MRI using 3-D convolutional ne...
Towards diagnosis of rotator cuff tears in 3-D MRI using 3-D convolutional ne...Wesley De Neve
 
Investigating the biological relevance in trained embedding representations o...
Investigating the biological relevance in trained embedding representations o...Investigating the biological relevance in trained embedding representations o...
Investigating the biological relevance in trained embedding representations o...Wesley De Neve
 
Impact of adversarial examples on deep learning models for biomedical image s...
Impact of adversarial examples on deep learning models for biomedical image s...Impact of adversarial examples on deep learning models for biomedical image s...
Impact of adversarial examples on deep learning models for biomedical image s...Wesley De Neve
 
Learning Biologically Relevant Features Using Convolutional Neural Networks f...
Learning Biologically Relevant Features Using Convolutional Neural Networks f...Learning Biologically Relevant Features Using Convolutional Neural Networks f...
Learning Biologically Relevant Features Using Convolutional Neural Networks f...Wesley De Neve
 
Booklet for the First GUGC Research Symposium
Booklet for the First GUGC Research SymposiumBooklet for the First GUGC Research Symposium
Booklet for the First GUGC Research SymposiumWesley De Neve
 
Center for Biotech Data Science at Ghent University Global Campus
Center for Biotech Data Science at Ghent University Global CampusCenter for Biotech Data Science at Ghent University Global Campus
Center for Biotech Data Science at Ghent University Global CampusWesley De Neve
 
Center for Biotech Data Science at Ghent University Global Campus
Center for Biotech Data Science at Ghent University Global CampusCenter for Biotech Data Science at Ghent University Global Campus
Center for Biotech Data Science at Ghent University Global CampusWesley De Neve
 
Learning biologically relevant features using convolutional neural networks f...
Learning biologically relevant features using convolutional neural networks f...Learning biologically relevant features using convolutional neural networks f...
Learning biologically relevant features using convolutional neural networks f...Wesley De Neve
 
Towards reading genomic data using deep learning-driven NLP techniques
Towards reading genomic data using deep learning-driven NLP techniquesTowards reading genomic data using deep learning-driven NLP techniques
Towards reading genomic data using deep learning-driven NLP techniquesWesley De Neve
 
Deep Machine Learning for Making Sense of Biotech Data - From Clean Energy to...
Deep Machine Learning for Making Sense of Biotech Data - From Clean Energy to...Deep Machine Learning for Making Sense of Biotech Data - From Clean Energy to...
Deep Machine Learning for Making Sense of Biotech Data - From Clean Energy to...Wesley De Neve
 
GUGC Info Session - Informatics and Bioinformatics
GUGC Info Session - Informatics and BioinformaticsGUGC Info Session - Informatics and Bioinformatics
GUGC Info Session - Informatics and BioinformaticsWesley De Neve
 
Ghent University Global Campus - Sungkyunkwan University: Workshop on Researc...
Ghent University Global Campus - Sungkyunkwan University: Workshop on Researc...Ghent University Global Campus - Sungkyunkwan University: Workshop on Researc...
Ghent University Global Campus - Sungkyunkwan University: Workshop on Researc...Wesley De Neve
 
Ghent University and GUGC-K: Overview of Teaching and Research Activities
Ghent University and GUGC-K: Overview of Teaching and Research ActivitiesGhent University and GUGC-K: Overview of Teaching and Research Activities
Ghent University and GUGC-K: Overview of Teaching and Research ActivitiesWesley De Neve
 
Biotech Data Science @ GUGC in Korea: Deep Learning for Prediction of Drug-Ta...
Biotech Data Science @ GUGC in Korea: Deep Learning for Prediction of Drug-Ta...Biotech Data Science @ GUGC in Korea: Deep Learning for Prediction of Drug-Ta...
Biotech Data Science @ GUGC in Korea: Deep Learning for Prediction of Drug-Ta...Wesley De Neve
 
Exploring Deep Machine Learning for Automatic Right Whale Recognition and No...
 Exploring Deep Machine Learning for Automatic Right Whale Recognition and No... Exploring Deep Machine Learning for Automatic Right Whale Recognition and No...
Exploring Deep Machine Learning for Automatic Right Whale Recognition and No...Wesley De Neve
 
Deep Machine Learning for Automating Biotech Tasks Through Self-Learning Expe...
Deep Machine Learning for Automating Biotech Tasks Through Self-Learning Expe...Deep Machine Learning for Automating Biotech Tasks Through Self-Learning Expe...
Deep Machine Learning for Automating Biotech Tasks Through Self-Learning Expe...Wesley De Neve
 
Towards using multimedia technology for biological data processing
Towards using multimedia technology for biological data processingTowards using multimedia technology for biological data processing
Towards using multimedia technology for biological data processingWesley De Neve
 
Multimedia Lab @ Ghent University - iMinds - Organizational Overview & Outlin...
Multimedia Lab @ Ghent University - iMinds - Organizational Overview & Outlin...Multimedia Lab @ Ghent University - iMinds - Organizational Overview & Outlin...
Multimedia Lab @ Ghent University - iMinds - Organizational Overview & Outlin...Wesley De Neve
 
Towards Twitter hashtag recommendation using distributed word representations...
Towards Twitter hashtag recommendation using distributed word representations...Towards Twitter hashtag recommendation using distributed word representations...
Towards Twitter hashtag recommendation using distributed word representations...Wesley De Neve
 
Orientation day at the Ghent University Global Campus in Korea: Introduction
Orientation day at the Ghent University Global Campus in Korea: IntroductionOrientation day at the Ghent University Global Campus in Korea: Introduction
Orientation day at the Ghent University Global Campus in Korea: IntroductionWesley De Neve
 

More from Wesley De Neve (20)

Towards diagnosis of rotator cuff tears in 3-D MRI using 3-D convolutional ne...
Towards diagnosis of rotator cuff tears in 3-D MRI using 3-D convolutional ne...Towards diagnosis of rotator cuff tears in 3-D MRI using 3-D convolutional ne...
Towards diagnosis of rotator cuff tears in 3-D MRI using 3-D convolutional ne...
 
Investigating the biological relevance in trained embedding representations o...
Investigating the biological relevance in trained embedding representations o...Investigating the biological relevance in trained embedding representations o...
Investigating the biological relevance in trained embedding representations o...
 
Impact of adversarial examples on deep learning models for biomedical image s...
Impact of adversarial examples on deep learning models for biomedical image s...Impact of adversarial examples on deep learning models for biomedical image s...
Impact of adversarial examples on deep learning models for biomedical image s...
 
Learning Biologically Relevant Features Using Convolutional Neural Networks f...
Learning Biologically Relevant Features Using Convolutional Neural Networks f...Learning Biologically Relevant Features Using Convolutional Neural Networks f...
Learning Biologically Relevant Features Using Convolutional Neural Networks f...
 
Booklet for the First GUGC Research Symposium
Booklet for the First GUGC Research SymposiumBooklet for the First GUGC Research Symposium
Booklet for the First GUGC Research Symposium
 
Center for Biotech Data Science at Ghent University Global Campus
Center for Biotech Data Science at Ghent University Global CampusCenter for Biotech Data Science at Ghent University Global Campus
Center for Biotech Data Science at Ghent University Global Campus
 
Center for Biotech Data Science at Ghent University Global Campus
Center for Biotech Data Science at Ghent University Global CampusCenter for Biotech Data Science at Ghent University Global Campus
Center for Biotech Data Science at Ghent University Global Campus
 
Learning biologically relevant features using convolutional neural networks f...
Learning biologically relevant features using convolutional neural networks f...Learning biologically relevant features using convolutional neural networks f...
Learning biologically relevant features using convolutional neural networks f...
 
Towards reading genomic data using deep learning-driven NLP techniques
Towards reading genomic data using deep learning-driven NLP techniquesTowards reading genomic data using deep learning-driven NLP techniques
Towards reading genomic data using deep learning-driven NLP techniques
 
Deep Machine Learning for Making Sense of Biotech Data - From Clean Energy to...
Deep Machine Learning for Making Sense of Biotech Data - From Clean Energy to...Deep Machine Learning for Making Sense of Biotech Data - From Clean Energy to...
Deep Machine Learning for Making Sense of Biotech Data - From Clean Energy to...
 
GUGC Info Session - Informatics and Bioinformatics
GUGC Info Session - Informatics and BioinformaticsGUGC Info Session - Informatics and Bioinformatics
GUGC Info Session - Informatics and Bioinformatics
 
Ghent University Global Campus - Sungkyunkwan University: Workshop on Researc...
Ghent University Global Campus - Sungkyunkwan University: Workshop on Researc...Ghent University Global Campus - Sungkyunkwan University: Workshop on Researc...
Ghent University Global Campus - Sungkyunkwan University: Workshop on Researc...
 
Ghent University and GUGC-K: Overview of Teaching and Research Activities
Ghent University and GUGC-K: Overview of Teaching and Research ActivitiesGhent University and GUGC-K: Overview of Teaching and Research Activities
Ghent University and GUGC-K: Overview of Teaching and Research Activities
 
Biotech Data Science @ GUGC in Korea: Deep Learning for Prediction of Drug-Ta...
Biotech Data Science @ GUGC in Korea: Deep Learning for Prediction of Drug-Ta...Biotech Data Science @ GUGC in Korea: Deep Learning for Prediction of Drug-Ta...
Biotech Data Science @ GUGC in Korea: Deep Learning for Prediction of Drug-Ta...
 
Exploring Deep Machine Learning for Automatic Right Whale Recognition and No...
 Exploring Deep Machine Learning for Automatic Right Whale Recognition and No... Exploring Deep Machine Learning for Automatic Right Whale Recognition and No...
Exploring Deep Machine Learning for Automatic Right Whale Recognition and No...
 
Deep Machine Learning for Automating Biotech Tasks Through Self-Learning Expe...
Deep Machine Learning for Automating Biotech Tasks Through Self-Learning Expe...Deep Machine Learning for Automating Biotech Tasks Through Self-Learning Expe...
Deep Machine Learning for Automating Biotech Tasks Through Self-Learning Expe...
 
Towards using multimedia technology for biological data processing
Towards using multimedia technology for biological data processingTowards using multimedia technology for biological data processing
Towards using multimedia technology for biological data processing
 
Multimedia Lab @ Ghent University - iMinds - Organizational Overview & Outlin...
Multimedia Lab @ Ghent University - iMinds - Organizational Overview & Outlin...Multimedia Lab @ Ghent University - iMinds - Organizational Overview & Outlin...
Multimedia Lab @ Ghent University - iMinds - Organizational Overview & Outlin...
 
Towards Twitter hashtag recommendation using distributed word representations...
Towards Twitter hashtag recommendation using distributed word representations...Towards Twitter hashtag recommendation using distributed word representations...
Towards Twitter hashtag recommendation using distributed word representations...
 
Orientation day at the Ghent University Global Campus in Korea: Introduction
Orientation day at the Ghent University Global Campus in Korea: IntroductionOrientation day at the Ghent University Global Campus in Korea: Introduction
Orientation day at the Ghent University Global Campus in Korea: Introduction
 

Recently uploaded

Cyathodium bryophyte: morphology, anatomy, reproduction etc.
Cyathodium bryophyte: morphology, anatomy, reproduction etc.Cyathodium bryophyte: morphology, anatomy, reproduction etc.
Cyathodium bryophyte: morphology, anatomy, reproduction etc.Cherry
 
POGONATUM : morphology, anatomy, reproduction etc.
POGONATUM : morphology, anatomy, reproduction etc.POGONATUM : morphology, anatomy, reproduction etc.
POGONATUM : morphology, anatomy, reproduction etc.Cherry
 
Site specific recombination and transposition.........pdf
Site specific recombination and transposition.........pdfSite specific recombination and transposition.........pdf
Site specific recombination and transposition.........pdfCherry
 
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune Waterworlds
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune WaterworldsBiogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune Waterworlds
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune WaterworldsSérgio Sacani
 
Call Girls Ahmedabad +917728919243 call me Independent Escort Service
Call Girls Ahmedabad +917728919243 call me Independent Escort ServiceCall Girls Ahmedabad +917728919243 call me Independent Escort Service
Call Girls Ahmedabad +917728919243 call me Independent Escort Serviceshivanisharma5244
 
TransientOffsetin14CAftertheCarringtonEventRecordedbyPolarTreeRings
TransientOffsetin14CAftertheCarringtonEventRecordedbyPolarTreeRingsTransientOffsetin14CAftertheCarringtonEventRecordedbyPolarTreeRings
TransientOffsetin14CAftertheCarringtonEventRecordedbyPolarTreeRingsSérgio Sacani
 
Gwalior ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Gwalior ESCORT SERVICE❤CALL GIRL
Gwalior ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Gwalior ESCORT SERVICE❤CALL GIRLGwalior ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Gwalior ESCORT SERVICE❤CALL GIRL
Gwalior ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Gwalior ESCORT SERVICE❤CALL GIRLkantirani197
 
Human genetics..........................pptx
Human genetics..........................pptxHuman genetics..........................pptx
Human genetics..........................pptxCherry
 
CYTOGENETIC MAP................ ppt.pptx
CYTOGENETIC MAP................ ppt.pptxCYTOGENETIC MAP................ ppt.pptx
CYTOGENETIC MAP................ ppt.pptxCherry
 
Cot curve, melting temperature, unique and repetitive DNA
Cot curve, melting temperature, unique and repetitive DNACot curve, melting temperature, unique and repetitive DNA
Cot curve, melting temperature, unique and repetitive DNACherry
 
biology HL practice questions IB BIOLOGY
biology HL practice questions IB BIOLOGYbiology HL practice questions IB BIOLOGY
biology HL practice questions IB BIOLOGY1301aanya
 
LUNULARIA -features, morphology, anatomy ,reproduction etc.
LUNULARIA -features, morphology, anatomy ,reproduction etc.LUNULARIA -features, morphology, anatomy ,reproduction etc.
LUNULARIA -features, morphology, anatomy ,reproduction etc.Cherry
 
PODOCARPUS...........................pptx
PODOCARPUS...........................pptxPODOCARPUS...........................pptx
PODOCARPUS...........................pptxCherry
 
FS P2 COMBO MSTA LAST PUSH past exam papers.
FS P2 COMBO MSTA LAST PUSH past exam papers.FS P2 COMBO MSTA LAST PUSH past exam papers.
FS P2 COMBO MSTA LAST PUSH past exam papers.takadzanijustinmaime
 
Kanchipuram Escorts 🥰 8617370543 Call Girls Offer VIP Hot Girls
Kanchipuram Escorts 🥰 8617370543 Call Girls Offer VIP Hot GirlsKanchipuram Escorts 🥰 8617370543 Call Girls Offer VIP Hot Girls
Kanchipuram Escorts 🥰 8617370543 Call Girls Offer VIP Hot GirlsDeepika Singh
 
Digital Dentistry.Digital Dentistryvv.pptx
Digital Dentistry.Digital Dentistryvv.pptxDigital Dentistry.Digital Dentistryvv.pptx
Digital Dentistry.Digital Dentistryvv.pptxMohamedFarag457087
 
The Mariana Trench remarkable geological features on Earth.pptx
The Mariana Trench remarkable geological features on Earth.pptxThe Mariana Trench remarkable geological features on Earth.pptx
The Mariana Trench remarkable geological features on Earth.pptxseri bangash
 
Bhiwandi Bhiwandi ❤CALL GIRL 7870993772 ❤CALL GIRLS ESCORT SERVICE In Bhiwan...
Bhiwandi Bhiwandi ❤CALL GIRL 7870993772 ❤CALL GIRLS  ESCORT SERVICE In Bhiwan...Bhiwandi Bhiwandi ❤CALL GIRL 7870993772 ❤CALL GIRLS  ESCORT SERVICE In Bhiwan...
Bhiwandi Bhiwandi ❤CALL GIRL 7870993772 ❤CALL GIRLS ESCORT SERVICE In Bhiwan...Monika Rani
 
Concept of gene and Complementation test.pdf
Concept of gene and Complementation test.pdfConcept of gene and Complementation test.pdf
Concept of gene and Complementation test.pdfCherry
 
THE ROLE OF BIOTECHNOLOGY IN THE ECONOMIC UPLIFT.pptx
THE ROLE OF BIOTECHNOLOGY IN THE ECONOMIC UPLIFT.pptxTHE ROLE OF BIOTECHNOLOGY IN THE ECONOMIC UPLIFT.pptx
THE ROLE OF BIOTECHNOLOGY IN THE ECONOMIC UPLIFT.pptxANSARKHAN96
 

Recently uploaded (20)

Cyathodium bryophyte: morphology, anatomy, reproduction etc.
Cyathodium bryophyte: morphology, anatomy, reproduction etc.Cyathodium bryophyte: morphology, anatomy, reproduction etc.
Cyathodium bryophyte: morphology, anatomy, reproduction etc.
 
POGONATUM : morphology, anatomy, reproduction etc.
POGONATUM : morphology, anatomy, reproduction etc.POGONATUM : morphology, anatomy, reproduction etc.
POGONATUM : morphology, anatomy, reproduction etc.
 
Site specific recombination and transposition.........pdf
Site specific recombination and transposition.........pdfSite specific recombination and transposition.........pdf
Site specific recombination and transposition.........pdf
 
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune Waterworlds
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune WaterworldsBiogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune Waterworlds
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune Waterworlds
 
Call Girls Ahmedabad +917728919243 call me Independent Escort Service
Call Girls Ahmedabad +917728919243 call me Independent Escort ServiceCall Girls Ahmedabad +917728919243 call me Independent Escort Service
Call Girls Ahmedabad +917728919243 call me Independent Escort Service
 
TransientOffsetin14CAftertheCarringtonEventRecordedbyPolarTreeRings
TransientOffsetin14CAftertheCarringtonEventRecordedbyPolarTreeRingsTransientOffsetin14CAftertheCarringtonEventRecordedbyPolarTreeRings
TransientOffsetin14CAftertheCarringtonEventRecordedbyPolarTreeRings
 
Gwalior ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Gwalior ESCORT SERVICE❤CALL GIRL
Gwalior ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Gwalior ESCORT SERVICE❤CALL GIRLGwalior ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Gwalior ESCORT SERVICE❤CALL GIRL
Gwalior ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Gwalior ESCORT SERVICE❤CALL GIRL
 
Human genetics..........................pptx
Human genetics..........................pptxHuman genetics..........................pptx
Human genetics..........................pptx
 
CYTOGENETIC MAP................ ppt.pptx
CYTOGENETIC MAP................ ppt.pptxCYTOGENETIC MAP................ ppt.pptx
CYTOGENETIC MAP................ ppt.pptx
 
Cot curve, melting temperature, unique and repetitive DNA
Cot curve, melting temperature, unique and repetitive DNACot curve, melting temperature, unique and repetitive DNA
Cot curve, melting temperature, unique and repetitive DNA
 
biology HL practice questions IB BIOLOGY
biology HL practice questions IB BIOLOGYbiology HL practice questions IB BIOLOGY
biology HL practice questions IB BIOLOGY
 
LUNULARIA -features, morphology, anatomy ,reproduction etc.
LUNULARIA -features, morphology, anatomy ,reproduction etc.LUNULARIA -features, morphology, anatomy ,reproduction etc.
LUNULARIA -features, morphology, anatomy ,reproduction etc.
 
PODOCARPUS...........................pptx
PODOCARPUS...........................pptxPODOCARPUS...........................pptx
PODOCARPUS...........................pptx
 
FS P2 COMBO MSTA LAST PUSH past exam papers.
FS P2 COMBO MSTA LAST PUSH past exam papers.FS P2 COMBO MSTA LAST PUSH past exam papers.
FS P2 COMBO MSTA LAST PUSH past exam papers.
 
Kanchipuram Escorts 🥰 8617370543 Call Girls Offer VIP Hot Girls
Kanchipuram Escorts 🥰 8617370543 Call Girls Offer VIP Hot GirlsKanchipuram Escorts 🥰 8617370543 Call Girls Offer VIP Hot Girls
Kanchipuram Escorts 🥰 8617370543 Call Girls Offer VIP Hot Girls
 
Digital Dentistry.Digital Dentistryvv.pptx
Digital Dentistry.Digital Dentistryvv.pptxDigital Dentistry.Digital Dentistryvv.pptx
Digital Dentistry.Digital Dentistryvv.pptx
 
The Mariana Trench remarkable geological features on Earth.pptx
The Mariana Trench remarkable geological features on Earth.pptxThe Mariana Trench remarkable geological features on Earth.pptx
The Mariana Trench remarkable geological features on Earth.pptx
 
Bhiwandi Bhiwandi ❤CALL GIRL 7870993772 ❤CALL GIRLS ESCORT SERVICE In Bhiwan...
Bhiwandi Bhiwandi ❤CALL GIRL 7870993772 ❤CALL GIRLS  ESCORT SERVICE In Bhiwan...Bhiwandi Bhiwandi ❤CALL GIRL 7870993772 ❤CALL GIRLS  ESCORT SERVICE In Bhiwan...
Bhiwandi Bhiwandi ❤CALL GIRL 7870993772 ❤CALL GIRLS ESCORT SERVICE In Bhiwan...
 
Concept of gene and Complementation test.pdf
Concept of gene and Complementation test.pdfConcept of gene and Complementation test.pdf
Concept of gene and Complementation test.pdf
 
THE ROLE OF BIOTECHNOLOGY IN THE ECONOMIC UPLIFT.pptx
THE ROLE OF BIOTECHNOLOGY IN THE ECONOMIC UPLIFT.pptxTHE ROLE OF BIOTECHNOLOGY IN THE ECONOMIC UPLIFT.pptx
THE ROLE OF BIOTECHNOLOGY IN THE ECONOMIC UPLIFT.pptx
 

The 5th Aslla Symposium

  • 1. INDUSTRY 4.0 @ GHENT UNIVERSITY (GLOBAL CAMPUS) - IMEC: FROM PREDICTIVE MAINTENANCE TO PERFORMANCE MONITORING October 23-26, 2018 @ Gangneung, Korea The 5th Aslla Symposium Wesley De Neve, Sofie Van Hoecke, Erik Mannens
  • 2. OUTLINE • Background • Use cases ‒ use case 1: wind turbine condition monitoring ‒ use case 2: incipient bearing fault detection ‒ use case 3: anomaly detection in steel coil production ‒ use case 4: greenhouse condition monitoring 2
  • 4. WESLEY DE NEVE • Academic credentials ̶ Master’s degree in computer science (2002)  from Ghent University, Belgium ̶ Ph.D. degree in computer science engineering (2007)  from Ghent University, Belgium ̶ post-doctoral experience (2007-2011)  at ICU / KAIST, Korea • Employment: Professor ̶ IDLab, Ghent University, Belgium (2011 - current) ̶ Ghent University Global Campus, Korea (2014 - current) 4
  • 5. SOFIE VAN HOECKE • Academic credentials ̶ Master’s degree in computer science engineering (2003)  from Ghent University, Belgium ̶ Ph.D. degree in computer science engineering (2009)  from Ghent University, Belgium • Employment: Professor ̶ University College West-Flanders (2011 - 2013) ̶ IDLab, Ghent University-imec, Belgium (2013 - current) 5
  • 6. ERIK MANNENS • Academic credentials ̶ Master’s degree in electro-mechanical engineering (1992)  from KAHO Ghent, Belgium ̶ Master’s degree in computer science (1995)  from KU Leuven, Belgium ̶ Ph.D. degree in computer science engineering (2011)  from Ghent University, Belgium • Employment: Professor & Research Valorization Director ̶ IDLab, Ghent University-imec, Belgium (2005 - current) 6
  • 8. GHENT UNIVERSITY • Public university ‒ located in Ghent, Belgium ‒ established in 1817 8
  • 9. GHENT UNIVERSITY • Counts 42,000 students and 9,000 staff members ‒ about 4,000 foreign students and 800 foreign staff members • Consists of 11 faculties, counting 117 departments ‒ campus buildings are distributed all over the city 9 Congress Center ‘Het Pand’ Faculty of Engineering and Architecture Aula Academia
  • 10. GHENT UNIVERSITY GLOBAL CAMPUS 10 Biotech campus located in Songdo, Incheon, Korea (+250 students) (part of the Incheon Global Campus, together with SUNY Stony Brook, GMU, & the University of Utah)
  • 11. • R&D hub in nanoelectronics and digital technologies ‒ brings together 3,500 scientists and engineers from over 70 countries • R&D groups in Belgium (headquarters: Leuven), the Netherlands, Taiwan, USA, China, and offices in India and Japan 11 SOTA cleanroom of 1 billion EURO (in collaboration with Samsung, a/o) IMEC
  • 12. IDLAB • Internet Technologies and Data Science Lab ‒ founded on October 1, 2016  merger of four already existing labs ‒ headcount  40 professors  50 post-doctoral researchers  200 researchers  15 support staff members 12 iGent Tower & AA Tower, Ghent, Belgium
  • 13. IDLAB • Education ‒ responsible for teaching more than 100 courses • Fundamental and applied research ‒ machine learning and data mining ‒ semantic intelligence ‒ multimedia processing ‒ distributed intelligence for IoT ‒ cloud and Big Data infrastructures ‒ wireless and wired networking 13
  • 14. Google DeepMind (+ Ghent University) four of our graduates are now working at Google DeepMind and one at Google Brain Sedol Lee Google DeepMind Challenge Match in March 2016 @ Seoul, Korea verdict: 4–1 for AlphaGo IDLAB 14
  • 15. INDUSTRY 4.0 PLATFORMS • Machines & Factories (M&F) ‒ an industrial research-oriented platform from Ghent University linked to the manufacturing industry ‒ groups the activities of about 125 researchers from different research groups, including 25 professors • Flanders Make ‒ strategic research center for the manufacturing industry in Flanders, set up by the Flemish government ‒ counts 400 researchers 15
  • 17. MAINTENANCE STRATEGIES • Preventive maintenance ‒ periodically inspect and replace components • Reactive maintenance ‒ if something breaks, replace it • Predictive maintenance 1) predict if, and possibly when, something is going to break 2) predict degradation of components 3) predict incipient faults 17
  • 18. MAINTENANCE STRATEGIES • Preventive maintenance ‒ periodically inspect and replace components • Reactive maintenance ‒ if something breaks, replace it • Predictive maintenance 1) predict if, and possibly when, something is going to break 2) predict degradation of components 3) predict incipient faults 18
  • 20. WIND TURBINE CONDITION MONITORING 20 • Offshore maintenance is expensive ‒ specialized equipment is typically needed for landing and working at offshore infrastructure • Offshore maintenance is difficult to schedule ‒ depends on weather and wave conditions
  • 21. WIND TURBINE CONDITION MONITORING 21 • Operation & maintenance actions may account for ± 30% of the total production cost of offshore electricity in wind farms • Early fault detection highly needed ‒ mitigates fault propagation and higher costs when detected (too) late
  • 22. PERFORMANCE MONITORING 22 • Create a model that can predict the power output of a wind turbine given several inputs Predictive model Predicted power output • Wind speed • Wind direction • Rotation speed • Yaw • Pitch data collected from 3 wind turbines over a time span of 3 months (10 weeks of training data and 2 weeks of validation data) random forests
  • 23. ANOMALY DETECTION IN PRODUCTION 23 • Abnormal operation ‒ difference in predicted output and actual output ‒ monitored through dashboard Predictive model Predicted power output • Wind speed • Wind direction • Rotation speed • Yaw • Pitch Detection module Actual power output Power production monitoring (Turbine 13)
  • 25. INCIPIENT BEARING FAULT DETECTION 25 • Bearings are in all kinds of machines ‒ conveyer belts, wind turbines, cars, … • Goal ‒ to detect incipient faults so to prevent faults from escalating
  • 26. INCIPIENT BEARING FAULT DETECTION 26 Time Inadequate lubrication Pitting Hard particles Bearing blockage Component failure Machine failure • Goal: to detect condition of a bearing at this point in time ‒ maintenance can be scheduled accordingly and reactive maintenance can be prevented
  • 27. INCIPIENT BEARING FAULT DETECTION 27 Captured (1) vibration data and (2) thermal imaging data from bearings in different conditions Data set Training set Feature extraction Learning algorithm Model Predictions
  • 28. INCIPIENT BEARING FAULT DETECTION 28 By using different machine learning algorithms (SVM, RDF, CNN) for analyzing vibration data and infrared video, we can automatically detect incipient faults with a high accuracy Model Inadequate lubrication Raceway fault Imbalance Hard particles
  • 29. INCIPIENT BEARING FAULT DETECTION • Infrared imaging analysis ̶ handcrafted features + SVM: accuracy of 88.25% • Vibration data analysis ̶ handcrafted features + RDF: accuracy of 87.25% ̶ CNN (end-to-end learning): accuracy of 91.77% • Use of a multimodal approach (fusion) ̶ accuracy of over 99% 29
  • 31. ANOMALY DETECTION IN STEEL MILL 31 • Any broken weld leads to interrupts of several minutes ‒ comes with a high cost • Fault prediction and cause detection based on static and dynamic parameters ‒ temperature ‒ pressure ‒ thickness ‒ speed Metric GBRT RCN True pos. 40 46 False neg. 10 4 False pos. 184 379 True neg. 558 363 Accuracy 77,60 % 70,46 %
  • 33. SMART FARMING 33 ? Where will we find enough food for 9 billion people by 2050? Farming 4.0: use sensors to make farms more “intelligent” and connected, so to be able to collect data about crop yields, soil- mapping, fertilizer applications, weather, and plant growth Collected data allow for improved decision-making → →
  • 34. GREENHOUSE CONDITION MONITORING 34 condition monitoring in automated / programmable greenhouses (food computers) plants on a conveyer belt RGB imaging hyperspectral imaging adjustment of conditions
  • 35. GREENHOUSE CONDITION MONITORING 35 • Drone-based phenomics ̶ measurement of physical and bio-chemical traits of organisms as a function of genetic and environmental changes ̶ hypothesis: gives less stress to plants • Research questions ̶ what is (not) possible? ̶ what is the role of (deep) data analytics? ̶ how to scale to a real-world farm? Indoor farming expert says Korea, Japan need to up investment The Korea Herald 11/2017
  • 36. 36 4. Point cloud matching 6. Determine skeleton 5. Segmentation organ recognition Structure-from- Motion Growing Neural Gas Convolutional Neural Networks Point Feature Histograms (PFH) descriptors 7. Estimate phenotypic parameters Conditional Random Fields 3. Filtering denoising 2. MVS-SFM point cloud 1. Image sequence GREENHOUSE CONDITION MONITORING
  • 37. TAKE-AWAY MESSAGES • IDLab has 10-15 professors working on machine learning, together with 50-100 pre- and post-doctoral researchers ̶ on a wide range of use cases ̶ on different types of (sensor) data • Open to different types of R&D collaboration ̶ imec ̶ Ghent University ̶ Ghent University Global Campus 37
  • 38. Thank you for your attention! {wesley.deneve, sofie.vanhoecke, erik.mannens}@ugent.be