Robust fault diagnosis of wind turbines based on MANFIS and zonotopic observers

Pérez-Pérez, Esvan-Jesús; Puig, Vicenç; López‐Estrada, Francisco‐Ronay; Valencia‐Palomo, Guillermo; Santos‐Ruiz, Ildeberto; Osorio-Gordillo, Gloria-Lilia

doi:10.1016/j.eswa.2023.121095

Cited by 11 publications

(2 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To face these problems digital twins (DTs) and signal based and data-driven surrogate models have recently emerged in the wind energy industry, as they can be used for health monitoring and predictive maintenance of wind turbine gearboxes [24,27]. Furthermore, they can help in providing an optimized and compact design of the WT gearboxes.…”

Section: Introductionmentioning

confidence: 99%

Digital Twin-Based Approach for a Multi-objective Optimal Design of Wind Turbine Gearboxes

Llopis-Albert,

Rubio,

Devece

et al. 2024

Preprint

View full text Add to dashboard Cite

Wind turbines (WTs) are clean a renewable energy source that has been growing in popularity in recent years. Gearboxes are complex, expensive, and critical components of WT, which are subject to high maintenance costs and several stresses including high loads and harsh environments that can lead to failure with significant downtime and financial losses. This paper focuses on the development of a digital twin-based approach for the modelling and simulation of WT gearboxes with the aim to improve their design, diagnosis, operation, and maintenance by providing insights into their behavior under different operating conditions. Pow-erful commercial computer-aided design tools (CAD) and computer-aided engineering (CAE) software are embedded into a computationally efficient multi-objective optimization framework (modeFrontier) with the aim of maximizing the power density, compactness, performance, and reliability of the WT gearbox. This entails minimizing its weight, volume, and maximum stresses and strains achieved without compromising efficiency. The 3D CAD model of the WT gearbox is carried out using SolidWorks, the Finite Element Analysis (FEA) to obtain the stresses and strains fields are modelled using Ansys Workbench, while the multibody kinematic and dynamic system is analyzed using MSC Adams. The method has been successfully applied to different case studies to find the optimal design and analyze the performance of the WT gearboxes. Simulation results can be used to identify potential failure modes and optimize gearbox design to improve performance, extend service life and relia-bility, thereby ensuring proper operation over its lifetime and reducing maintenance costs.

show abstract

Section: Introductionmentioning

confidence: 99%

Digital Twin-Based Approach for a Multi-objective Optimal Design of Wind Turbine Gearboxes

Llopis-Albert,

Rubio,

Devece

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…The authors of [14] presented an integrated design method combining active fault diagnosis and tracking control to detect incipient faults while maintaining normal closed-loop operation. The authors of [15] proposed utilizing zonotopic observers and MANFIS models for robust fault diagnosis. In [16], it was demonstrated that a failure due to breakage of two bars located at polar passage in the air gap can mask the failure harmonic and make its detection by spectral techniques unfeasible.…”

Section: Introductionmentioning

confidence: 99%

A Comparative Analysis of Deep Learning Convolutional Neural Network Architectures for Fault Diagnosis of Broken Rotor Bars in Induction Motors

Barrera-Llanga,

Burriel-Valencia,

Sapena-Bañó

et al. 2023

Sensors

View full text Add to dashboard Cite

Induction machines (IMs) play a critical role in various industrial processes but are susceptible to degenerative failures, such as broken rotor bars. Effective diagnostic techniques are essential in addressing these issues. In this study, we propose the utilization of convolutional neural networks (CNNs) for detection of broken rotor bars. To accomplish this, we generated a dataset comprising current samples versus angular position using finite element method magnetics (FEMM) software for a squirrel-cage rotor with 28 bars, including scenarios with 0 to 6 broken bars at every possible relative position. The dataset consists of a total of 16,050 samples per motor. We evaluated the performance of six different CNN architectures, namely Inception V4, NasNETMobile, ResNET152, SeNET154, VGG16, and VGG19. Our automatic classification system demonstrated an impressive 99% accuracy in detecting broken rotor bars, with VGG19 performing exceptionally well. Specifically, VGG19 exhibited high accuracy, precision, recall, and F1-Score, with values approaching 0.994 and 0.998. Notably, VGG19 exhibited crucial activations in its feature maps, particularly after domain-specific training, highlighting its effectiveness in fault detection. Comparing CNN architectures assists in selecting the most suitable one for this application based on processing time, effectiveness, and training losses. This research suggests that deep learning can detect broken bars in induction machines with accuracy comparable to that of traditional methods by analyzing current signals using CNNs.

show abstract

Fault tolerant observer design for a class of nonlinear systems with corrupted outputs

Guerra,

de Oliveira,

Berdjag

et al. 2024

Intl J Robust & Nonlinear

View full text Add to dashboard Cite

This paper presents new sufficient conditions for state and fault estimation of nonlinear systems. A new approach is proposed to deal with unmeasured premise variables and corrupted outputs. The adopted observer strategy consists of a scheduled gain that depends only on the estimated outputs provided by the observer. The design of such scheduled gain follows from a technical rewriting of the state error and uses the Lyapunov framework to derive sufficient conditions written as linear matrix inequalities. Such conditions are less conservative than existing approaches that deal with unmeasured premise variables. The effectiveness of the developed approach is assessed through a numerical single‐link robotic arm and on a real car‐based experimental setup available at the laboratory.

show abstract

Robust fault diagnosis of wind turbines based on MANFIS and zonotopic observers

Cited by 11 publications

References 25 publications

Digital Twin-Based Approach for a Multi-objective Optimal Design of Wind Turbine Gearboxes

Digital Twin-Based Approach for a Multi-objective Optimal Design of Wind Turbine Gearboxes

A Comparative Analysis of Deep Learning Convolutional Neural Network Architectures for Fault Diagnosis of Broken Rotor Bars in Induction Motors

Fault tolerant observer design for a class of nonlinear systems with corrupted outputs

Contact Info

Product

Resources

About