Course info

Description:
Machine learning (ML) and advanced predictive statistical techniques are gaining widespread use in the field of electrical engineering as a whole, and for state-of-health modelling of Lithium-ion batteries in particular. The introduction of ML and statistics in electrical engineering is a consequence of the field slowly subsidising some of the more expensive laboratory testing by using data collected during real-life operating conditions. The upside of using ML and predictive statistics is that, in many instances, these methods can achieve an acceptable precision using a reduced amount of laboratory testing. However, it comes at the cost of added model complexity and the loss of some of the explanatory power when compared to the physics based state-of-health models.

This two-day course introduces key aspects of machine learning, predictive modelling, and model validation. Focusing on quantitative predictive models for Lithium-ion battery state-of-health modelling. The models will include sequential and non-sequential approaches, univariate and multivariate outcomes, Bayesian and frequentist frameworks, as well as the reduction and selection of features for modelling. The general aim of these methods is to predict capacity degradation, resistance increase, and remaining useful life based on a combination of laboratory and field data.

Exemplifications of some of the discussed topics will be made through exercises in R and Matlab.

Day 1: Introduction to Machine Learning and Bayesian statistics – Søren B. Vilsen & Daniel Stroe (8 hours)

· Overview of machine learning methods, the bias-variance trade-off, and cross-validation.

· Bayesian statistics, sequential model updating, and remaining useful life estimation.

· Auto-regressive models, Kalman-filtering, and state intervention for state-of-charge and state-of health estimation.

· Battery performance parameters for state-of-health estimation

Day 2: Machine learning for battery SOH estimation – Søren B. Vilsen & Daniel Stroe (8 hours)

· Feature reduction using principle components analysis and auto-encoders.

· Feature selection using subset selection, partial least squares, and shrinkage methods.

· Support vector regression, Gaussian process regression, and neural networks.

· Boosting, ensemble learning, and interpreting black-box methods.

Prerequisites: Fundamental understanding of probability, statistics, and is recommended, as is basic knowledge of either R or Matlab. Note: the course language is English.

Form of evaluation: Students are expected to solve a number of exercises and deliver an individual report with solutions and comments.

Organizer: Assoc. Prof. Daniel-Ioan Stroe, dis@energy.aau.dk

Lecturers:
Postdoc. Søren B. Vilsen, AAU-MATH
Assoc. Prof. Daniel-Ioan Stroe, AAU Energy

ECTS: 2.0

Time: 24-25 May 2022

Place: AAU Energy, Aalborg (hybrid)

Price: 6000 DKK for PhD students outside of Denmark and 8000 DKK for the Industry excl. VAT

Number of seats: 30

Deadline: 3 May 2022

Important information concerning PhD courses:
We have over some time experienced problems with no-show for both project and general courses. It has now reached a point where we are forced to take action. Therefore, the Doctoral School has decided to introduce a no-show fee of DKK 3.000 for each course where the student does not show up. Cancellations are accepted no later than 2 weeks before start of the course. Registered illness is of course an acceptable reason for not showing up on those days. Furthermore, all courses open for registration approximately four months before start. This can hopefully also provide new students a chance to register for courses during the year. We look forward to your registrations.