Pasar al contenido principal

Resultados relacionados

Resultados totales (Incluyendo duplicados): 3
Encontrada(s) 1 página(s)

Digibug. Repositorio Institucional de la Universidad de Granada
Publicación Artículo científico (JournalArticle)

Physics-guided recurrent neural network trained with approximate Bayesian computation: A case study on structural response prognostics

  • Fernández Salas, Juan
  • Chiachío Ruano, Juan
  • Barros, José
  • Chiachío Ruano, Manuel
  • Kulkarni, Chetan S.
A new physics-guided Bayesian recurrent neural network is proposed in this manuscript. This hybrid algorithm benefits from the knowledge in physics-based models, the capability of recurrent neural networks to handle sequential data, and the flexibility of Bayesian methods to quantify the uncertainty. The introduction of physics in the forward pass of the neural network significantly improves the results in multistep-ahead forecasting, and the gradient-free nature of the Bayesian learning engine provides great flexibility to adapt to the observed data. The proposed algorithm has been applied to a data-driven problem about fatigue in composites, and a case study about accelerations in concrete buildings, where a comparison against the state-of-the-art algorithms is also provided. The results revealed: (1) the accuracy of the proposals, comparable to the state-of-the-art recurrent neural networks; (2) its stability during multiple runs of the algorithm, proving that it is a more reliable option; (3) its precise quantification of the uncertainty, which provides useful information for the subsequent decision-making process. As potential future applications to real world scenarios, the proposed Bayesian recurrent neural network could be used in on-board PHM systems in the aerospace industry, or as an on-site prediction tool in buildings for seismic events and/or aftershocks., ENHAnCE project, which has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 859957, Funding for open access charge: Universidad de Granada / CBUA.
Digibug. Repositorio Institucional de la Universidad de Granada
Publicación Artículo científico (JournalArticle)

Training of physics-informed Bayesian neural networks with ABC-SS for prognostic of Li-ion batteries

  • Fernández Salas, Juan
  • Corbetta, Matteo
  • Kulkarni, Chetan S.
  • Chiachío Ruano, Juan
  • Chiachío Ruano, Manuel
The current surge in the need for Li-ion batteries to power electric vehicles has also translated in a need for more advanced models that can predict their behavior, but also quantify the uncertainty in their predictions, given the amount of variables involved and the varying operating conditions. This manuscript proposes a new Bayesian physics-informed recurrent neural network, where the battery discharge curve is described using the Nernst and Butler–Volmer equations, the activity correction term within such equations is modeled with two multilayer perceptrons, and approximate Bayesian computation by subset-simulation is used to train the weights, bias and the physical parameters representing the maximum charge available and the internal resistance. The challenges found during the adaptation and implementation of the Bayesian training algorithm to the recurrent physics-informed cell are described, along with the approaches proposed to overcome them. The performance of the Bayesian hybrid model presented in this paper has also been evaluated using data from NASA Ames Prognostics Data Repository, and the results show comparable accuracy to the standard approach with backpropagation, and a flexible and realistic quantification of the uncertainty. Furthermore, the uncertainty related to the physical parameters of the hybrid model can be evaluated in semi-isolation of the weights and bias of the MLPs, providing a sensitivity tool to assess the relative importance between different parameters., ENHAnCE project, which has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 859957
Digibug. Repositorio Institucional de la Universidad de Granada
Publicación Artículo científico (JournalArticle)

Reliability-based leading edge erosion maintenance strategy selection framework

  • Contreras López, Javier
  • Kolios, Athanasios
  • Wang, Lin
  • Chiachío Ruano, Manuel
  • Dimitrov, Nikolay
Leading edge erosion has become one of the most prevailing failure modes of wind turbines. Its effects can evolve from an aerodynamic modification of the properties of the blade to a potential structural failure of the leading edge. The first produces a reduction of energy production and the second can produce a catastrophic failure of the blade. Considering the uncertainties and constraints involved in the design of optimal operation and maintenance (O&M) strategies for offshore assets and the influence of site-specific parameters on the dynamics of this particular failure mode, the task becomes complex. In this study, a framework to evaluate the influence of different maintenance strategies considering uncertainties in weather, material behaviour and repair success is presented. Monte Carlo Simulation (MCS) is used alongside a computational framework for Leading Edge Erosion (LEE) degradation to evaluate the lifetime cost distribution and probability of failure of the chosen maintenance strategies. The use of the framework is demonstrated in a case study considering a 5-MW offshore wind turbine located in the north of Germany. The influence of the modification of the maintenance interval or time between repairs and the comparison with maintenance activities executed only during months with milder weather is analysed in terms of cost and reliability. A Pareto front plot considering the probability of failure and the median of the cost is used to jointly compare strategies considering both aspects to provide a tool for risk-informed maintenance selection. Finally, the potential benefits of conditionbased maintenance and autonomous decision-making systems are discussed. The case of study shows the benefits of repairs during summer months and the importance of the relation risk/O&M cost for different maintenance strategies., ENHAnCE project that has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 859957