Objective:Upon completion of this course, participants will be able to:
1. Understand optimization fundamentals including problem formulation and classification.
2. Formulate architect optimization workflows to address a given problem.
3. Understand key design criteria and basic underlying physics of models used in a wind farm design process.
4. Develop highly simplified models for wind farm physics and perform optimization of simple wind farm case studies using these models.
5. Apply advanced wind farm design tools (i.e. TOPFARM) to the optimization of wind farm design.
Course Purpose: This course emphasize on optimization applications in wind energy and wind farm design in particular. The course covers fundamental aspects of optimization, including basic theory and algorithms and will then apply those tools to the analysis and design of wind farms.
Prerequisites: Participants should have advanced knowledge in engineering; previous coursework in optimization and wind energy systems will be helpful but is not required.
Materials: The course instructors will provide all necessary course materials, the following texts are recommended to participants as valuable additional resources:
Wind Energy Explained: Theory, Design and Application, 2nd Edition, Manwell, J., McGowan, J., and Rogers, A., Wiley ISBN 978-0-470-01500-1.
Optimization Concepts and Applications in Engineering, Belegundu, A. and Chandrupatla, T., Cambridge Univ. Press ISBN-13: 978-0521878463.
Software: Software required for the course is all open-source and includes: Python (http://python.org ), OpenMDAO (http://openmdao.org ), and DTU TOPFARM software (http://gitlab.windenergy.dtu.dk/TOPFARM ).
Homework assignment will consist of short projects that require participants to solve optimization problems related to material covered during classroom seminars. The assignment will involve “toy” optimization problems in wind energy that are directly based on course material.
Team mini-project: participants will work in groups of 3 to complete an extended project on layout optimization. Participants will give presentations on their projects at the end of the training.
Main topics: (We will schedule the content based on special interests of participants so we can emphasize certain topics more than others. For that reason please feel free to send us the topics that you are mostly interested in)
Introduction: Systems engineering & optimization with applications to wind energy
- Didactic situation method for the basic wind farm optimization problem
- Problem Formulation and Classification
- Inductive method using classic problem types (thief, traveling salesman, etc)
- Learning with cases using real-world examples
- Revisit didactic example and do problem formulation and classification
Optimization: from prospecting to commissioning, Activation through in-situ exercises:
- Flow & Wake models
- Convex optimization (unconstrained)
- Constrained Non-Linear Optimization
- Constrained Linear Optimization
- Gradient-Based Optimization Methods
- Gradient-Free Methods
- Infrastructure models (electrical system and roads)
- Hybrid (wind-pv-storage) renewable power plant sizing and design
- Advanced Network Optimization Methods
- Introduction to loads and surrogates
- Multi-disciplinary design optimization of wind farm design
- Activation through in-situ mathematical exercises
- Discussion of real-world projects and constraints from assignments
- A la carte method based on prior student experience with Python: Python basics, Python for Matlab users, Numpy and Scipy for numerical and scientific programming, Numpy, Xarray for gridded dand matrix algebra, Pandas and time-series data analysis, Matplotlib and visualization
- Set up and perform convex optimization for simple mathematical models
- Build a simple flow model (tutorial) and perform farm energy yield analysis
- Individual optimization of wind farm axial induction factor and sensitivity to wind turbine spacing
- Activation through identifying problem formulations and classifications
- TOPFARM tutorials and wind farm layout and collection system optimization
- Advanced topics in wind energy MDAO: adjoints, high-fidelity, optimization under uncertainty, etc - Learning by inquiry – how does this relate back to problem formulation and classification? How do they change?
Case study & industry guests (subject to availability of guests from industry)
- Review assignment 1 and collective discussion of real-world cases submitted by students
- how does this relate back to problem formulation? What types of optimization problems do we encounter?
- Design drivers and trade-offs – what trade-offs do we see?
Date: 14th June to 18th June (9am to 4pm everyday)
Key instructor: Katherine Dykes
For more information please contact: firstname.lastname@example.org