Data-Driven Robust Optimization for Installation of Offshore Windfarms

Overview

This project focuses on the installation process of wind turbines in offshore wind farms. The researcher is expected to develop optimization models and algorithms to identify the optimal configuration of vessel schedules to minimise installation duration and cost. Furthermore, the project aims to achieve robust installation schedules that can handle operational changes due to weather uncertainties.

Project Details

The researcher is expected to achieve the following objectives:

1. To develop a complex deterministic model to identify the optimal configuration of vessel schedules to minimise installation duration and cost. This is achieved by using classical optimisation approaches that enable the researcher to develop a mathematical deterministic model meeting all constraints of the installation problem. The deterministic model will naturally have many physical constraints and is very difficult to solve due to the complex nature of the problem. As routes each vessel traverses can be enumerated easily, it is aimed to transform the complex deterministic model into a simpler deterministic model albeit with numerous problems to solve, each similar to classical shortest path problems.
2. To design and develop computationally efficient algorithms to solve the deterministic model that will be tested on real-world scenarios. After modelling the problem, the researcher is expected to develop efficient algorithms that can run fast for very large problems in real-world. Extensive computational testing is also essential at this stage.
3. To develop a robust, data-driven model that incorporates the uncertainties into the deterministic model and develop computationally efficient algorithms to deal with the robust model. This objective is fulfilled by incorporating the weather uncertainty into the deterministic model in novel ways. Data-driven robust optimisation, which also exploits state-of-the-art data analytics tools, will enable the researcher to build a model that can tackle all possible scenarios. Such a model will also motivate the design and development of efficient algorithms, that are to be extensively tested using years of weather data from Met Office as well as real-world offshore wind farm installation projects.

Eligibility

1st class honours/undergraduate degree (essential) and an excellent Masters-level qualification or equivalent (highly desirable), in a closely relevant subject such as computer science, operations research, mathematics and statistics, management science, and industrial engineering, from a recognised academic institution. If English is not your first language, you will also be required to provide evidence such as a recent UKVI recognised English language test (such as IELTS, minimum overall band score of 6.5 with no individual test score below 5.5) or a university degree completed in a recognized English speaking country.