- UCAS course code
- H301
- UCAS institution code
- M20
Course unit details:
Data-driven Modelling & Simulation
Unit code | AERO40041 |
---|---|
Credit rating | 15 |
Unit level | Level 4 |
Teaching period(s) | Semester 1 |
Available as a free choice unit? | No |
Overview
Data science and machine learning have become more and more widely used in research and engineering firms and so are essential skills for a graduate engineer. This unit introduces students to advanced modelling and simulation concepts such as machine learning, data-driven optimisation, surrogate modelling and uncertainty quantification, relevant to modern design in Aerospace and Mechanical Engineering. With the help of hands-on laboratories for solving Engineering problems, the unit will build the capabilities of data-driven modelling that graduates should expect to use in their careers.
Aims
To lay the foundations of data science and machine learning for the practical solutions to engineering problems. You will expand upon prior knowledge covered in Modelling and Simulation and the Numerical Methods and Computing courses through the use of modern data-driven methods for engineering problems. You will apply this knowledge to different engineering goals. Central to this is the development of both the theoretical and computational aspects of applied data science, data-driven optimisation, machine learning and uncertainty quantification. The skills developed will be honed via hands-on laboratories aimed at solving complex engineering problems.
Syllabus
An introduction to machine learning for engineering analysis -- the motivation, notation and definitions. Fundamental Algorithms: Linear regression, KNN algorithm, Gradient decent, Neural Networks for regression tasks in engineering. Overview of optimization, terminologies and methods; Sensitivity-based optimization; Grid-search and surrogate-based optimizations; Gaussian process regression for surrogate construction; Bayesian optimization based on Gaussian processes. Cross-validation and over-fitting; An overview to uncertainty quantification. A high-level overview of advanced AI (artificial intelligence) models for engineering analysis; reinforcement learning and generative AI.
Practical skills
Assessment methods
Method | Weight |
---|---|
Written exam | 50% |
Report | 50% |
Feedback methods
Each piece of coursework is supported by timetabled computer sessions. The students will be expected to attempt the coursework in their own time and ask for feedback from teaching staff during the computer sessions, before submission. T
Study hours
Scheduled activity hours | |
---|---|
Lectures | 24 |
Practical classes & workshops | 12 |
Independent study hours | |
---|---|
Independent study | 114 |
Teaching staff
Staff member | Role |
---|---|
Saleh Rezaeiravesh | Unit coordinator |