Browsing by Author "Calvo-Rolle, Jose Luis"
Now showing 1 - 4 of 4
Results Per Page
Sort Options
- An intelligent system for harmonic distortions detection in wind generator power electronic devicesPublication . Jove, Esteban; González-Cava, Jose Manuel; Casteleiro-Roca, José-Luis; Alaiz-Moretón, Héctor; Baruque, Bruno; Leitão, Paulo; Méndez Pérez, Juan Albino; Calvo-Rolle, Jose LuisThe high concern about climate change has boosted the promotion of renewable energy systems, being the wind power one of the key generation possibilities in this field. In this context, with the aim of ensuring the energy efficiency, the present work deals with the fault detection in the power electronic circuits of a wind generator system placed in a bioclimatic house. To do so, different outliers that emulate harmonic distortion appearance are tested. To implement a system capable of detecting this anomalous situations, six different one-class techniques are used, whose performance is thoroughly analyzed, offering interesting performance.
- Artificial intelligence data-driven petri nets approach for virtualizing digital twinsPublication . Oliveira Júnior, Alexandre de; Calvo-Rolle, Jose Luis; Leitão, PauloVirtualization is one key design principle in Industry 4.0, with the modeling and simulation of the physical assets playing crucial roles in the Digital Twin context. Different approaches can be used to implement the virtual asset models, ranging from simple equations to complex mathematical models. Petri nets formalism is a suitable approach to model and simulate the physical asset operation in the Digital context, particularly those that are event-driven, taking advantage of its inherent robust mathematical foundation. Having this in mind, this paper proposes a Petri nets approach, which considers Artificial Intelligent data-driven analytics associated to timed transitions to support the execution of what-if simulation aiming the monitoring, diagnosis, prediction, and optimization. The proposed approach was tested in an experimental punching machine, allowing the early identification the performance degradation in the Digital Twin and the selection of actions to be implemented in the physical asset to improve its operation.
- A novel method for anomaly detection using beta hebbian learning and principal component analysisPublication . Zayas-Gato, Francisco; Michelena, Álvaro; Quintián, Héctor; Jove, Esteban; Casteleiro-Roca, José-Luis; Leitão, Paulo; Calvo-Rolle, Jose LuisIn this research work a novel two-step system for anomaly detection is presented and tested over several real datasets. In the first step the novel Exploratory Projection Pursuit, Beta Hebbian Learning algorithm, is applied over each dataset, either to reduce the dimensionality of the original dataset or to face nonlinear datasets by generating a new subspace of the original dataset with lower, or even higher, dimensionality selecting the right activation function. Finally, in the second step Principal Component Analysis anomaly detection is applied to the new subspace to detect the anomalies and improve its classification capabilities. This new approach has been tested over several different real datasets, in terms of number of variables, number of samples and number of anomalies. In almost all cases, the novel approach obtained better results in terms of area under the curve with similar standard deviation values. In case of computational cost, this improvement is only remarkable when complexity of the dataset in terms of number of variables is high.
- Simulation on Digital Twin: Role of Artificial Intelligence and Emergence of Industrial MetaversePublication . Júnior, Alexandre O.; Calvo-Rolle, Jose Luis; Leitão, PauloDigital Twins (DTs) are cutting-edge technological design principles of Industry 4.0. They elevate the representation level of physical systems backed up by accurate real-time data in virtual environments and empower the simulation capabilities of these systems through Artificial Intelligence (AI) for their analysis, monitoring, and optimization. This work comprehensively explores the intrinsic interaction between simulation and AI in DTs, meticulously covering the current literature status and categorizing these symbiotic interactions into three different groups that cover AI to support DT-based simulation, AI for optimization of simulation within DT, and simulation to support AI approaches in DT. In addition, a deeper look is taken at the role of simulation and AI in the emerging concept of the Industrial Metaverse, which promises to extend DTs beyond discrete virtual representation of physical systems to encompass the industrial ecosystem from end-to-end. Finally, the main research challenges for achieving the full integration of simulation and AI in DTs and at the Industrial Metaverse are discussed.