Share this page on:

Conference programme 

Back to the programme printer.gif Print

Poster session

Lead Session Chair:
Stephan Barth, Managing Director, ForWind - Center for Wind Energy Research, Germany
Mircea Scutariu Mott MacDonald, United Kingdom
Co-authors:
Mircea Scutariu (1) F P Xiao Yi (1)
(1) Mott MacDonald, Glasgow, United Kingdom

Printer friendly version: printer.gif Print

Poster
Download poster(2.32 MB)

Presenter's biography

Biographies are supplied directly by presenters at OFFSHORE 2015 and are published here unedited

Dr. Mircea SCUTARIU has been working in the wind industry for more than eight years. He is team leader for renewables integration with Mott MacDonald in Glasgow. He received the MSc and PhD degrees in electric power engineering from University POLITEHNICA Bucharest. After his studies he was academic lecturer for nearly ten years followed by five years with an electricity distribution company in senior roles before he joined the current company. He worked on offshore wind projects optioneering and design for Race Bank, London Array, Neart na Gaoithe and East Anglia One projects focusing on concept selection and grid connections.

Abstract

Optimization of offshore wind farm collection systems

Introduction

Offshore wind is one of the most scalable renewable technologies providing future additional low carbon power
generation. Offshore wind farms use collection systems for gathering power from Wind Turbine Generators (WTG).
The procurement cost of submarine cables is typically 7% of the total capital expenditure for an offshore wind
project while cable installation costs will account for another 4%. Improved collection system design can
potentially provide up to 10% of the total capital expenditure savings for offshore wind projects. Our team
developed a unique, innovative automated tool which implements a heuristic search for the solution to the
optimization of collection systems problem.

Approach

The underlying mathematical problem of finding the configuration that collects the power from a given array of
WTGs in the most effective manner although akin to the fairly well-known optimization problem of the Travelling
Salesman has a few particular constraints that require the re-calibration of traditional solving methods.
An initial connectivity of the collection system is determined using a greedy algorithm variant that connects all
WTGs while meeting the problems’ constraints. The initial collection system is refined using genetic algorithm
processing that will retain the best performing solution from a number of refined solutions that are identified in the
process.

Main body of abstract

Collection systems design is a time-consuming labor-intensive activity due to the virtually unlimited number of
possible combinations between the collection voltage, cable cross-sectional-area sizes, number of WTGs, number
of offshore collection substations and collection system configurations. A systematic search through all these
combinations is prohibitively expensive and does not guarantee that the most effective collection system is finally
identified.
Fundamentally the problem consists in finding, for a given WTG array defined based on the analysis the
prevailing wind characteristics at the site and the geotechnical information about the seabed, the collection
system that gathers power in the most effective manner while practical constraints are taken into account.
The efficiency is usually measured by an optimization criterion that can be a single quantity, such as capital
expenditure, or a combination between capital and operational expenditure. The constraints are a complex
mixture between the numbers of collection substations, the number of strings allowed per substation, allowed
string configurations, practical cable current carrying ability, collection voltage, special crossing circumstances,
seabed install-ability, etc.
An increased efficiency of the genetic algorithm process was incorporated into the enhanced second version of
the leading-edge automated tool alongside the full calculation of the availability performance of the collection
system configuration. The availability performance is converted internally into estimates of the operational
expenditure due to non-realised energy collection that are then added to the combined capital and
operational expenditure performance. The code enables finding routes that avoid or limit crossings of seabed
zones with challenging installation conditions.

Conclusion

The paper will discuss the development and application of the leading-edge automated tool
enabling rapid and efficient determination of the collection system configuration that achieves the highest
performance measured in a combination of capital expenditure and operational expenditure accounting for
economic valuation of energy losses, energy not realised due to unavailability of collection system and the
maintenance costs during projects' lifetime.
The paper wil also described the enhanced functionality built in the tool for the identification of intelligent
routing that avoids or limits crossing of seabed zones with challenging installation conditions such as highly
mobile and/or prohibitively tall sand waves.


Learning objectives
Outline the complexity of the underlying problem of finding the collection system configuration that achieves
the highest techno-economic performance while meeting genuine practical constraints.
Provide a complete overview of a unique automated tool that enables fast and comprehensive heuristic search
through the vast solutions' space that is associated with the collection system optimisation problem.