Deep Reinforcement Learning Based on Proximal Policy Optimization for the Maintenance of a Wind Farm with Multiple Crews

Pinciroli, Luca; Baraldi, Piero; Ballabio, Guido; Compare, Michele; Zio, Enrico

doi:10.3390/en14206743

Cited by 21 publications

(7 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…By employing a clipping surrogate objective function, PPO adeptly balances exploration and exploitation, effectively mitigating the risk of suboptimal policy adjustments. Its versatility is underscored by its successful applications across diverse domains, including robotics Heess et al (2017) , gaming (notably achieving exceptional performance in Dota 2) Berner et al (2019) , finance Lin & Beling (2020) , and wind farm management Pinciroli et al (2021) .…”

Section: Methodsmentioning

confidence: 99%

Management of investment portfolios employing reinforcement learning

Santos,

Garruti,

Barboza

et al. 2023

PeerJ Computer Science

View full text Add to dashboard Cite

Investors are presented with a multitude of options and markets for pursuing higher returns, a task that often proves complex and challenging. This study examines the effectiveness of reinforcement learning (RL) algorithms in optimizing investment portfolios, comparing their performance with traditional strategies and benchmarking against American and Brazilian indices. Additionally, it was explore the impact of incorporating commodity derivatives into portfolios and the associated transaction costs. The results indicate that the inclusion of derivatives can significantly enhance portfolio performance while reducing volatility, presenting an attractive opportunity for investors. RL techniques also demonstrate superior effectiveness in portfolio optimization, resulting in an average increase of 12% in returns without a commensurate increase in risk. Consequently, this research makes a substantial contribution to the field of finance. It not only sheds light on the application of RL but also provides valuable insights for academia. Furthermore, it challenges conventional notions of market efficiency and modern portfolio theory, offering practical implications. It suggests that data-driven investment management holds the potential to enhance efficiency, mitigate conflicts of interest, and reduce biased decision-making, thereby transforming the landscape of financial investment.

show abstract

Section: Methodsmentioning

confidence: 99%

Management of investment portfolios employing reinforcement learning

Santos,

Garruti,

Barboza

et al. 2023

PeerJ Computer Science

View full text Add to dashboard Cite

show abstract

“…The International Energy Agency expects wind energy to be the main energy source used for producing electricity in 2027 [4]. Wind power growth will be driven by future improvement in the design and management of wind farms [5,6].…”

Section: Background and Scopementioning

confidence: 99%

Heat Recovery from a PtSNG Plant Coupled with Wind Energy

et al. 2021

View full text Add to dashboard Cite

Power to substitute natural gas (PtSNG) is a promising technology to store intermittent renewable electricity as synthetic fuel. Power surplus on the electric grid is converted to hydrogen via water electrolysis and then to SNG via CO2 methanation. The SNG produced can be directly injected into the natural gas infrastructure for long-term and large-scale energy storage. Because of the fluctuating behaviour of the input energy source, the overall annual plant efficiency and SNG production are affected by the plant operation time and the standby strategy chosen. The re-use of internal (waste) heat for satisfying the energy requirements during critical moments can be crucial to achieving high annual efficiencies. In this study, the heat recovery from a PtSNG plant coupled with wind energy, based on proton exchange membrane electrolysis, adiabatic fixed bed methanation and membrane technology for SNG upgrading, is investigated. The proposed thermal recovery strategy involves the waste heat available from the methanation unit during the operation hours being accumulated by means of a two-tanks diathermic oil circuit. The stored heat is used to compensate for the heat losses of methanation reactors, during the hot-standby state. Two options to maintain the reactors at operating temperature have been assessed. The first requires that the diathermic oil transfers heat to a hydrogen stream, which is used to flush the reactors in order to guarantee the hot-standby conditions. The second option entails that the stored heat being recovered for electricity production through an Organic Rankine Cycle. The electricity produced is used to compensate the reactors heat losses by using electrical trace heating during the hot-standby hours, as well as to supply energy to ancillary equipment. The aim of the paper is to evaluate the technical feasibility of the proposed heat recovery strategies and how they impact on the annual plant performances. The results showed that the annual efficiencies on an LHV basis were found to be 44.0% and 44.3% for the thermal storage and electrical storage configurations, respectively.

show abstract

“…Researchers address many challenges to optimize the performance of such systems and the reliability of operation. In [11], the authors considered the optimization of operation and maintenance of wind turbines in multi-user wind farms. They proposed a new approach to the problem as a sequential decision problem over a long time horizon and used deep-learning mechanisms with reinforcement based on proximal policy optimization to solve it.…”

Section: Literature Reviewmentioning

confidence: 99%

“…What is more, Google already announced in 2010 that the loading speed factor of websites will be taken into account during their positioning in the search results on desktop browsers. For mobile devices, a fix called "speed update" was implemented in July 2018 [11]. Studies on this issue have been ongoing for years.…”

Section: Literature Reviewmentioning

confidence: 99%

Comparative Analysis of Selected Open-Source Solutions for Traffic Balancing in Server Infrastructures Providing WWW Service

2021

View full text Add to dashboard Cite

As the number of users increased over the years, pioneering technologies and solutions in given areas ceased to be sufficient even in terms of performance. Therefore, there was a need for their development or even redesign and redefinition. One of the issues that undoubtedly has a huge impact on the current shape of the global network and the way information is processed in it is the issue of traffic balancing, especially the one in the server infrastructure related to the WWW service, providing users with the possibility of efficient and reliable web browsing. The paper presents a comparative analysis of selected open-source solutions used for traffic balancing in server infrastructures providing WWW service based on selected criteria. The designed architecture of the test environment and the test results of selected tools implementing the traffic-balancing functionality are presented. Methodologies, test plans, and comparison criteria are proposed. A comparative analysis of results based on specific criteria was performed. The balance between network traffic optimization and load balancing distribution among servers is crucial for the development of energy-efficient data processing centers.

show abstract

Deep Reinforcement Learning Based on Proximal Policy Optimization for the Maintenance of a Wind Farm with Multiple Crews

Cited by 21 publications

References 43 publications

Management of investment portfolios employing reinforcement learning

Management of investment portfolios employing reinforcement learning

Heat Recovery from a PtSNG Plant Coupled with Wind Energy

Comparative Analysis of Selected Open-Source Solutions for Traffic Balancing in Server Infrastructures Providing WWW Service

Contact Info

Product

Resources

About