Bayesian Network Modelling for the Wind Energy Industry: An Overview

Abstract

Wind energy farms are moving into deeper and more remote waters to benefit from availability of more space for the installation of wind turbines as well as higher wind speed for the production of electricity. Wind farm asset managers must ensure availability of adequate power supply as well as reliability of wind turbines throughout their lifetime. The operating conditions in deep water environments often change very rapidly and, therefore the decision metrics used in different phases of a wind energy project's lifecycle will have to be updated on a very frequent basis, to guarantee higher wind energy system reliability levels. For this reason, there is a crucial need for the wind energy industry to develop advanced computational tools/techniques that are capable of modelling the possible scenarios in (near) real-time and provide a prompt response to any changes in operational/environmental conditions. Bayesian network (BN) is a popular machine learning (ML) method used for system modelling and decision-making under uncertainty. This paper provides a systematic review and evaluation of existing research on the use of BN models in the wind energy sector. To conduct this literature review, all relevant databases from inception to date were searched, and a total of 70 sources (including journal publications, conference proceedings, PhD dissertations, industry reports, best practice documents and software user guides) which met the inclusion criteria were identified, excluding references used in other sections of the text for discussion. Our review findings reveal that the applications of BNs in the wind energy industry are quite diverse, ranging from wind power and weather forecasting to risk management, fault diagnosis and prognosis, structural analysis, reliability assessment, and maintenance planning and updating. Furthermore, a number of case studies are presented to illustrate the applicability of BNs in practice. Although the paper details information applicable to the wind energy industry, knowledge can be transferred to many other sectors.

Unmanned Aerial Drones for Inspection of Offshore Wind Turbines: A Mission-Critical Failure Analysis

“…FTA is usually represented by a logic diagram beginning with an undesired top event (which, in this case, is the inability of the drone to accomplish the mission) and then working backward through intermediate events until all possible root causes at the bottom are determined. Pathways are used to interconnect events that contribute to the top event and/or intermediate events [28]. These pathways use standard logic symbols, such as AND, OR, and others.…”

5G Wireless Networks in the Future Renewable Energy Systems

“…It analyses fluctuating demand and changing weather conditions in order not only to predict energy demand, but also to anticipate potential problems in the grid and to propose ways to improve weaknesses in distribution lines. In addition, similar to the German example, AI can be used to analyze data from various sources such as wind, solar, wind turbines and wind farms to make informed real-time decisions to maximize demand flexibility (Adedipe et al, 2020;Sweeney et al, 2020). Other ways in which artificial intelligence can facilitate demand-side response and demand flexibility include using game-theory algorithms to create incentives to improve aggregate participation and using blockchain and other distributed ledger technologies to protect data (Ahl et al, 2019;Tsao et al, 2020).…”

A Stochastic Petri Net Model for O&M Planning of Floating Offshore Wind Turbines

“…Bayesian network (BN) models have been used in several studies for reliability analysis as well as fault diagnosis of complex engineering systems. They have been applied to offshore wind turbines as well, but the model is yet to be adapted for O&M optimization of floating offshore wind farms [24]. BNs are useful for the dynamic risk analysis and reliability assessment of offshore wind farms, allowing for updating the O&M schedules when additional data becomes available.…”