101-0521-10L  Machine Learning for Predictive Maintenance Applications

SemesterFrühjahrssemester 2020
DozierendeO. Fink
Periodizitätjährlich wiederkehrende Veranstaltung
LehrspracheEnglisch
KommentarThe number of participants in the course is limited to 25 students.

Students interested in attending the lecture are requested to upload their transcript and a short motivation responding the following two questions (max. 200 words):
-How does this course fit to the other courses you have attended so far?
-How does the course support you in achieving your goal?
The following link can be used to upload the documents.
Link



Lehrveranstaltungen

NummerTitelUmfangDozierende
101-0521-10 GMachine Learning for Predictive Maintenance Applications4 Std.
Di09:45-11:30HIL C 10.2 »
10:00-12:00ER SA TZ »
Do12:45-14:30HIL C 10.2 »
13:00-15:00ER SA TZ »
O. Fink

Katalogdaten

KurzbeschreibungThe course aims at developing machine learning algorithms that are able to use condition monitoring data efficiently and detect occurring faults in complex industrial assets, isolate their root cause and ultimately predict the remaining useful lifetime.
LernzielStudents will
- be able to understand the main challenges faced by predictive maintenance systems
- learn to extract relevant features from condition monitoring data
-learn to select appropriate machine learning algorithms for fault detection, diagnostics and prognostics
-learn to define the learning problem in way that allows its solution based on existing constrains such as lack of fault samples.
- learn to design end-to-end machine learning algorithms for fault detection and diagnostics
-be able to evaluate the performance of the applied algorithms.

At the end of the course, the students will be able to design data-driven predictive maintenance applications for complex engineered systems from raw condition monitoring data.
InhaltEarly and reliable detection, isolation and prediction of faulty system conditions enables the operators to take recovery actions to prevent critical system failures and ensure a high level of availability and safety. This is particularly crucial for complex systems such as infrastructures, power plants and aircraft engines. Therefore, their system condition is increasingly tightly monitored by a large number of diverse condition monitoring sensors. With the increased availability of data on system condition on the one hand, and the increased complexity of explicit system physics-based models on the other hand, the application of data-driven approaches for predictive maintenance has been recently increasing.
This course provides insights and hands-on experience in selecting, designing, optimizing and evaluating machine learning algorithms to tackle the challenges faced by maintenance systems of complex engineered systems.

Specific topics include:

-Introduction to condition monitoring and predictive maintenance systems
-Feature extraction and selection methodology
-Machine learning algorithms for fault detection and fault isolation
-End-to-end learning architectures (including feature learning) for fault detection and fault isolation
-Unsupervised and semi-supervised learning algorithms for predictive maintenance
-Machine learning algorithms for prediction of the remaining useful life
-Performance evaluation
-Predictive maintenance systems at fleet level
-Domain adaptation for fault diagnostics
-Introduction to decision support systems for maintenance applications
SkriptSlides and other materials will be available online.
LiteraturRelevant scientific papers will be discussed in the course.
Voraussetzungen / BesonderesStrong analytical skills.
Programming skills in python are strongly recommended.

Leistungskontrolle

Information zur Leistungskontrolle (gültig bis die Lerneinheit neu gelesen wird)
Leistungskontrolle als Semesterkurs
ECTS Kreditpunkte8 KP
PrüfendeO. Fink
Formbenotete Semesterleistung
PrüfungsspracheEnglisch
RepetitionRepetition nur nach erneuter Belegung der Lerneinheit möglich.
Zusatzinformation zum PrüfungsmodusPerformance will be assessed during the semester based on
-5 exercises, requiring the students to perform defined sub-tasks for designing a predictive maintenance system (60% of the final grade in total)
-Presentation and reimplementation of a scientific journal paper recently published in the field of predictive maintenance (15%).
-Report (including the implementation) and presentation of a real case study of designing a predictive maintenance system based on raw condition monitoring signals of a complex engineered system (25%)

Lernmaterialien

Keine öffentlichen Lernmaterialien verfügbar.
Es werden nur die öffentlichen Lernmaterialien aufgeführt.

Gruppen

Keine Informationen zu Gruppen vorhanden.

Einschränkungen

PlätzePlätze beschränkt. Spezielles Auswahlverfahren.
WartelisteBis 03.03.2020
BelegungsendeBelegung nur bis 17.02.2020 möglich

Angeboten in

StudiengangBereichTyp
Bauingenieurwissenschaften MasterVertiefung in Bau- und ErhaltungsmanagementWInformation
Maschineningenieurwissenschaften MasterRobotics, Systems and ControlWInformation
Raumentwicklung und Infrastruktursysteme MasterVertiefungsfächer für alle VertiefungenWInformation
Robotics, Systems and Control MasterKernfächerWInformation