Electrical Network Design for Offshore Wind: Analysis, Mathematical Modelling, and Optimization

Ambitious targets set by the European Commission see offshore wind power reaching 450GW by 2050. Economies of scale push offshore wind to be deployed with hundreds of Wind Turbines (WTs) in single projects. Likewise, the distance from the Onshore Connection Point (OCP) to the Offshore Substation (OSS), is also increasing as response for finding new sites with high wind power potential.

A conspicuous increment in the complexity to design (and optimize) the electrical network for modern OWFs is resulted. The export submarine cables are complex dynamic systems, exposed to time-varying conditions that need to be assessed and understood to optimally size their cross-sections. Besides, in the collection systems, the feasible set of connections between WTs increase exponentially as a function of the project size.

On the top of the technical challenges, the economics of electrical power cables play a significant role as well. Between 2020 and 2024 more than 6,750km of export cables are estimated to be needed. Meanwhile, 19,000km of submarine cables for collection systems are prognosed to be installed from 2018 to 2028, with an estimated worth of £5.36bn.

This places power cables as one of the main components of the Balance of Plant (BoP), representing at least 11% of the overall Levelised Cost Of Energy (LCOE). In order to address these challenges, several methods and mathematical optimization models are proposed in this PhD thesis. The problem of sizing optimally export cables is approached through a comprehensive single framework, supporting cost reduction and reliability.

A probabilistic lifetime estimation model is implemented to calculate the effects of cumulative damage due to electro-thermal stress. This framework is benchmarked against industrial standards, demonstrating the capability to reduce the overall LCOE by optimizing the export cable. The deterministic design of the cable layout for OWFs collection systems is studied through several proposed approaches, based on heuristics, metaheuristics, and global optimization methods.

For the latter, benchmarking results indicate the superiority of the proposed method against a state-of-the-art approach published in the scientific literature, in terms of solution quality, computing time, and optimality gap. This is continued with the proposition of a MILP program to design simultaneously the collection and transmission systems, accounting for OSSs location and forbidden areas.

Finally, a stochastic global optimization method is proposed to design closed-loop topology for OWF collection systems. A comparative analysis between radial and closed-loop topologies is performed to calculate and compare the cost benefits from each of them.