A novel method based on artificial neural networks for selecting the most appropriate locations of the offshore wind farms

In order to obtain the most efficient solution in harvesting offshore wind energy, the converting devices must be located in places where the wind provides enough power, according to the used device specifications and, as possible, with constant value during turbine lifetime. On the other hand, the device must be chosen for a specific place - imposed by other reasons, like water depth, shore distance etc. The paper presents an analysis and prediction methodology for the wind speed based on the artificial neural networks modeling. The proposed approach uses as input data the previous recorded and forecasted wind speed values and the GPS coordinates for several places in Black Sea allowing the identification of the location with less wind speed modification during analyzed period of time. Also, the model provides useful information about inputs importance on output evolution. The obtained results are used for the identification of optimal placement of wind energy converting devices, leading to an improved efficiency. (C) 2022 The Author(s). Published by Elsevier Ltd.

Automated summary

This paper proposes an analysis and prediction methodology for wind speed based on artificial neural networks modeling. By utilizing previous recorded and forecasted wind speed values and GPS coordinates as input data, it identifies the location with the least wind speed variation in the Black Sea. This provides useful information for determining the optimal placement of wind energy converting devices.