M. Borsotti
info
Please Note
<p>This page displays the records of the person named above and is not linked to a unique person identifier. This record may need to be merged to a profile.</p>
5 records found
1
Offshore Wind Farms (OWF) are expected to play a crucial role in mitigating climate change and promoting sustainable development, nevertheless, Operation and Maintenance (O&M) costs can reach 25–30 % of the total cost. Efficient O&M strategies reduce maintenance frequency, downtime, and improve overall performance. This paper reviews O&M decision-making for fixed-bottom OWFs according to a three-level decision-making hierarchy, strategic, tactical, and operational, reflecting how decisions vary by scope and time horizon of reference. Strategic decisions are typically focused on the overall maintenance strategy for the wind farm. Tactical decisions are focused on the selection of the fleet and the management of spare parts. Operational decisions are focused on the scheduling of individual maintenance tasks on a daily or weekly basis and the routing of the vessels. Exploring how the different decision-making layers have been addressed in the literature leads to valuable insights into open challenges and paves the way for the development of new decision-making methods. In this paper we highlight the untapped potential of prognostic-driven scheduling, we identify the lack of comprehensive models and methods that encompass all decision-making echelons holistically, and we emphasize the need for the integration of environmental considerations into the decision-making process.
...
Offshore Wind Farms (OWF) are expected to play a crucial role in mitigating climate change and promoting sustainable development, nevertheless, Operation and Maintenance (O&M) costs can reach 25–30 % of the total cost. Efficient O&M strategies reduce maintenance frequency, downtime, and improve overall performance. This paper reviews O&M decision-making for fixed-bottom OWFs according to a three-level decision-making hierarchy, strategic, tactical, and operational, reflecting how decisions vary by scope and time horizon of reference. Strategic decisions are typically focused on the overall maintenance strategy for the wind farm. Tactical decisions are focused on the selection of the fleet and the management of spare parts. Operational decisions are focused on the scheduling of individual maintenance tasks on a daily or weekly basis and the routing of the vessels. Exploring how the different decision-making layers have been addressed in the literature leads to valuable insights into open challenges and paves the way for the development of new decision-making methods. In this paper we highlight the untapped potential of prognostic-driven scheduling, we identify the lack of comprehensive models and methods that encompass all decision-making echelons holistically, and we emphasize the need for the integration of environmental considerations into the decision-making process.
Offshore wind farms face unique challenges in maintenance due to harsh weather, remote locations, and complex logistics. Traditional maintenance strategies often fail to optimize operations, leading to unplanned failures or unnecessary servicing. In recent years, deep reinforcement learning (DRL) has shown clear potential to tackle these challenges through a data-driven approach. This paper provides a critical review of representative DRL models for offshore wind farm maintenance planning, elaborating on both single- and multi-agent frameworks, diverse training algorithms, various problem formulations, and the integration of domain-specific knowledge. The review compares the benefits and limitations of these methods, identifying a significant gap in the widely adopted use of simplistic binary maintenance decisions, rather than including multi-level or imperfect repairs in the action space. In conclusion, this work suggests directions for future research to overcome current limitations and enhance the applicability of DRL methods in offshore wind maintenance.
...
Offshore wind farms face unique challenges in maintenance due to harsh weather, remote locations, and complex logistics. Traditional maintenance strategies often fail to optimize operations, leading to unplanned failures or unnecessary servicing. In recent years, deep reinforcement learning (DRL) has shown clear potential to tackle these challenges through a data-driven approach. This paper provides a critical review of representative DRL models for offshore wind farm maintenance planning, elaborating on both single- and multi-agent frameworks, diverse training algorithms, various problem formulations, and the integration of domain-specific knowledge. The review compares the benefits and limitations of these methods, identifying a significant gap in the widely adopted use of simplistic binary maintenance decisions, rather than including multi-level or imperfect repairs in the action space. In conclusion, this work suggests directions for future research to overcome current limitations and enhance the applicability of DRL methods in offshore wind maintenance.
This paper presents a stochastic optimization model for predictive maintenance scheduling in offshore wind farms. The proposed model integrates probabilistic Remaining Useful Life (RUL) prognosis with mathematical optimization and Model Predictive Control (MPC) techniques that updates RUL beliefs with new prognostic measurements at each epoch to dynamically adjust maintenance decisions. Unlike conventional scheduling methods that rely on static age thresholds, our approach uses real-time prognostics to improve cost efficiency and reduce downtime. A case study on 50 wind turbines demonstrates that dynamically adapting maintenance schedules using prognostics reduces O&M expenses by 8.7%, primarily through significant reductions in downtime, compared to traditional methods.
...
This paper presents a stochastic optimization model for predictive maintenance scheduling in offshore wind farms. The proposed model integrates probabilistic Remaining Useful Life (RUL) prognosis with mathematical optimization and Model Predictive Control (MPC) techniques that updates RUL beliefs with new prognostic measurements at each epoch to dynamically adjust maintenance decisions. Unlike conventional scheduling methods that rely on static age thresholds, our approach uses real-time prognostics to improve cost efficiency and reduce downtime. A case study on 50 wind turbines demonstrates that dynamically adapting maintenance schedules using prognostics reduces O&M expenses by 8.7%, primarily through significant reductions in downtime, compared to traditional methods.
Offshore wind maintenance requires decisions under uncertain component health, weather-driven accessibility, logistical constraints, and costly downtime. This thesis develops a framework to investigate how prognostic information can
support such decisions. It compares optimization-based maintenance planning using mixed-integer linear programming with learning-based planning using deep reinforcement learning. The results highlight their respective strengths and
limitations, supporting the development of more adaptive, reliable, and cost-effective offshore wind O&M decision-support systems. ...
support such decisions. It compares optimization-based maintenance planning using mixed-integer linear programming with learning-based planning using deep reinforcement learning. The results highlight their respective strengths and
limitations, supporting the development of more adaptive, reliable, and cost-effective offshore wind O&M decision-support systems. ...
Offshore wind maintenance requires decisions under uncertain component health, weather-driven accessibility, logistical constraints, and costly downtime. This thesis develops a framework to investigate how prognostic information can
support such decisions. It compares optimization-based maintenance planning using mixed-integer linear programming with learning-based planning using deep reinforcement learning. The results highlight their respective strengths and
limitations, supporting the development of more adaptive, reliable, and cost-effective offshore wind O&M decision-support systems.
support such decisions. It compares optimization-based maintenance planning using mixed-integer linear programming with learning-based planning using deep reinforcement learning. The results highlight their respective strengths and
limitations, supporting the development of more adaptive, reliable, and cost-effective offshore wind O&M decision-support systems.
Offshore wind farms are a promising source of renewable energy, but they face significant challenges in terms of operation and maintenance (O&M). Traditional scheduling models often overlook the potential of condition-based maintenance (CBM). Addressing this gap, this paper introduces a novel framework, incorporating principles of Model Predictive Control (MPC), to optimize the O&M scheduling of offshore wind farms using prognostic-driven maintenance. The framework integrates probabilistic remaining useful life (RUL) prognosis in a mixed-integer linear programming (MILP) optimization model with a rolling horizon approach, in alignment with MPC’s predictive and adaptive decision-making approach. The optimization model determines the optimal time to replace each component by minimizing the expected cost over the expected lifetime. This approach seeks to achieve the lowest expense while guaranteeing the highest utilization rate of each component. For the case study presented, the total O&M costs are reduced by up to 15% with respect to corrective maintenance strategies.
...
Offshore wind farms are a promising source of renewable energy, but they face significant challenges in terms of operation and maintenance (O&M). Traditional scheduling models often overlook the potential of condition-based maintenance (CBM). Addressing this gap, this paper introduces a novel framework, incorporating principles of Model Predictive Control (MPC), to optimize the O&M scheduling of offshore wind farms using prognostic-driven maintenance. The framework integrates probabilistic remaining useful life (RUL) prognosis in a mixed-integer linear programming (MILP) optimization model with a rolling horizon approach, in alignment with MPC’s predictive and adaptive decision-making approach. The optimization model determines the optimal time to replace each component by minimizing the expected cost over the expected lifetime. This approach seeks to achieve the lowest expense while guaranteeing the highest utilization rate of each component. For the case study presented, the total O&M costs are reduced by up to 15% with respect to corrective maintenance strategies.