Electrical Insulator Fault Forecasting Based on a Wavelet Neuro-Fuzzy System

Stefenon, Stéfano Frizzo; Freire, Roberto Zanetti; Coelho, Leandro dos Santos; Meyer, Luiz Henrique; Grebogi, Rafael Bartnik; Buratto, William Gouvêa; Nied, Ademir

doi:10.3390/en13020484

Cited by 52 publications

(31 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The signal is decomposed by dividing the spectrum into separate halves with a low-pass filter and a high-pass filter [ 29 ]. At each new decomposition, the coefficients of the previous iteration are assumed [ 30 ]. Figure 1 shows the scheme for a WPT decomposition level 3, where 8 packets of the same frequency are obtained, where

represents the correlation coefficients and k is the decomposition level.…”

Section: Methodsmentioning

confidence: 99%

Railway Axle Condition Monitoring Technique Based on Wavelet Packet Transform Features and Support Vector Machines

Gómez

Castejón

Corral

et al. 2020

Sensors

View full text Add to dashboard Cite

Railway axles are critical to the safety of railway vehicles. However, railway axle maintenance is currently based on scheduled preventive maintenance using Nondestructive Testing. The use of condition monitoring techniques would provide information about the status of the axle between periodical inspections, and it would be very valuable in the prevention of catastrophic failures. Nevertheless, in the literature, there are not many studies focusing on this area and there is a lack of experimental data. In this work, a reliable real-time condition-monitoring technique for railway axles is proposed. The technique was validated using vibration measurements obtained at the axle boxes of a full bogie installed on a rig, where four different cracked railway axles were tested. The technique is based on vibration analysis by means of the Wavelet Packet Transform (WPT) energy, combined with a Support Vector Machine (SVM) diagnosis model. In all cases, it was observed that the WPT energy of the vibration signals at the first natural frequency of the axle when the wheelset is first installed (the healthy condition) increases when a crack is artificially created. An SVM diagnosis model based on the WPT energy at this frequency demonstrates good reliability, with a false alarm rate of lower than 10% and defect detection for damage occurring in more than 6.5% of the section in more than 90% of the cases. The minimum number of wheelsets required to build a general model to avoid mounting effects, among others things, is also discussed.

show abstract

represents the correlation coefficients and k is the decomposition level.…”

Section: Methodsmentioning

confidence: 99%

Railway Axle Condition Monitoring Technique Based on Wavelet Packet Transform Features and Support Vector Machines

Gómez

Castejón

Corral

et al. 2020

Sensors

View full text Add to dashboard Cite

show abstract

“…Adaptative Neuro-Fuzzy Inference System (ANFIS) aggregates the Artificial Neural Networks (ANN) concept of robustness, adaptation, non-linear mapping, and at the same time as it is based on the Takagi-Sugeno-Kang inference model [14]. The Fuzzy Inference System (FIS), or fuzzy logic, can handle uncertainties and imprecise pieces of information.…”

Section: Meta-algorithm Inducers Descriptionmentioning

confidence: 99%

“…ANFIS have been applied in many recent works to FDI, e.g., [14][15][16]. All of them are in an ad-hoc approach since such works have used ANFIS straightforwardly to analyze system output data.…”

Section: Meta-algorithm Inducers Descriptionmentioning

confidence: 99%

Empowering Predictive Maintenance: A Hybrid Method to Diagnose Abnormal Situations

et al. 2020

View full text Add to dashboard Cite

Aerospace systems are composed of hundreds or thousands of components and complex subsystems which need an appropriate health monitoring capability to enable safe operation in various conditions. In terms of monitoring systems, it is possible to find a considerable number of state-of-the-art works in the literature related to ad-hoc solutions. Still, it is challenging to reuse them even with subtle differences in analogous subsystems or components. This paper proposes the Generic Anomaly Detection Hybridization Algorithm (GADHA) aiming to build a more reusable algorithm to support anomaly detection. The solution consists of analyzing different supervised machine learning classification algorithms combined in ensemble techniques, with a physical model when available, and two levels of a decision to estimate the current state of the monitored system. Finally, the proposed algorithm assures at least equal, or, more typically, better, overall accuracy in fault detection and isolation than the application of such algorithms alone, through few adaptations.

show abstract

“…To develop accurate forecasting models, there is a trend that involves using combined or hybrid forecasting methods such as pre-processing (decomposition), optimization (single and multi-objective approaches), and artificial intelligence models [10][11][12][13]. Within this context, each methodology can add to the forecasting model its own expertise to deal with different signals characteristics.…”

Section: Introductionmentioning

confidence: 99%

Electricity Price Forecasting Based on Self-Adaptive Decomposition and Heterogeneous Ensemble Learning

et al. 2020

Self Cite

View full text Add to dashboard Cite

Electricity price forecasting plays a vital role in the financial markets. This paper proposes a self-adaptive, decomposed, heterogeneous, and ensemble learning model for short-term electricity price forecasting one, two, and three-months-ahead in the Brazilian market. Exogenous variables, such as supply, lagged prices and demand are considered as inputs signals of the forecasting model. Firstly, the coyote optimization algorithm is adopted to tune the hyperparameters of complementary ensemble empirical mode decomposition in the pre-processing phase. Next, three machine learning models, including extreme learning machine, gradient boosting machine, and support vector regression models, as well as Gaussian process, are designed with the intent of handling the components obtained through the signal decomposition approach with focus on time series forecasting. The individual forecasting models are directly integrated in order to obtain the final forecasting prices one to three-months-ahead. In this case, a grid of forecasting models is obtained. The best forecasting model is the one that has better generalization out-of-sample. The empirical results show the efficiency of the proposed model. Additionally, it can achieve forecasting errors lower than 4.2% in terms of symmetric mean absolute percentage error. The ranking of importance of the variables, from the smallest to the largest is, lagged prices, demand, and supply. This paper provided useful insights for multi-step-ahead forecasting in the electrical market, once the proposed model can enhance forecasting accuracy and stability.

show abstract

Electrical Insulator Fault Forecasting Based on a Wavelet Neuro-Fuzzy System

Cited by 52 publications

References 48 publications

Railway Axle Condition Monitoring Technique Based on Wavelet Packet Transform Features and Support Vector Machines

Railway Axle Condition Monitoring Technique Based on Wavelet Packet Transform Features and Support Vector Machines

Empowering Predictive Maintenance: A Hybrid Method to Diagnose Abnormal Situations

Electricity Price Forecasting Based on Self-Adaptive Decomposition and Heterogeneous Ensemble Learning

Contact Info

Product

Resources

About