Kernel-based support vector machines for automated health status assessment in monitoring sensor data

Diez-Olivan, Alberto; Pagan, Jose A.; Khoa, Nguyen Lu Dang; Sanz, Ricardo; Sierra, Basilio

doi:10.1007/s00170-017-1204-2

Cited by 28 publications

(10 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Unsupervised models are preferred in dynamic environments and when there are no labels in the data. As far as the first category (sensor signals) is concerned, the authors in Diez et al (2016), Diez-Olivan et al (2017), and Diez-Olivan et al (2018) try to model the normality using clustering algorithms or v-SVM. Specifically, in Diez et al (2016), the authors try to detect failures in bridge joints; regarding bridge joints as a fleet renders this technique applicable to our setting.…”

Section: Pdm On Top Of Continuous Datamentioning

confidence: 99%

A context-aware unsupervised predictive maintenance solution for fleet management

Giannoulidis

Gounaris

2022

J Intell Inf Syst

View full text Add to dashboard Cite

We deal with the problem of predictive maintenance (PdM) in a vehicle fleet management setting following an unsupervised streaming anomaly detection approach. We investigate a variety of unsupervised methods for anomaly detection, such as proximity-based, hybrid (statistical and proximity-based) and transformers. The proposed methods can properly model the context in which each member of the fleet operates. In our case, the context is both crucial for effective anomaly detection and volatile, which calls for streaming solutions that take into account only the recent values. We propose two novel techniques, a 2-stage proximity-based one and context-aware transformers along with advanced thresholding. In addition, to allow for testing PdM techniques for vehicle fleets in a fair and reproducible manner, we build a new fleet-like benchmarking dataset based on an existing dataset of turbofan simulations. Our evaluation results show that our proposals reduce the maintenance costs compared to existing solutions.

show abstract

Section: Pdm On Top Of Continuous Datamentioning

confidence: 99%

A context-aware unsupervised predictive maintenance solution for fleet management

Giannoulidis

Gounaris

2022

J Intell Inf Syst

View full text Add to dashboard Cite

show abstract

“…The establishment of density distribution function is to set a kernel function at each observation value, and take the bandwidth as the boundary. The kernel function within the boundary is the distribution probability of the observation value, and then the accumulation of the kernel function of the observations is the density distribution of the observations [28]. Compared with the parametric distribution method, the KDE does not need to group the original data, and it solves the influence of grouping on parametric estimation from the source [29].…”

Section: Introductionmentioning

confidence: 99%

Compilation method of CNC lathe cutting force spectrum based on kernel density estimation of G-SCE

Wang

et al. 2021

Int J Adv Manuf Technol

View full text Add to dashboard Cite

The cutting force spectrum of the CNC lathe is the basic data for the reliability design, reliability test and reliability evaluation of the CNC lathe and its components. Due to the complex and changeable turning conditions and different cutting processes, the cutting load presents multi-peak characteristic. In order to compile the CNC lathe cutting force spectrum accurately, a compilation method based on kernel density estimation (KDE) of goodnesssmoothness comprehensive evaluation (G-SCE) is proposed. According to the measured dynamic cutting force, the KDE is used to establish the dynamic cutting force distribution of the CNC lathe. For the rule of thumb method (ROT) based on the mean integrated square error and the least squares cross-validation method (LCV) based on the integrated square error does not take into account the influence of different bandwidths on the goodness estimation and the smoothness of the estimated curve. The estimation accuracy test method based on multiple goodness-of-fit tests, and the smoothness test method based on the envelope curve are proposed. The entropy method is used to comprehensively calculate the estimated goodness index and the smoothness index to determine the optimal bandwidth. The case analysis shows that the method proposed can solve the problem of too large estimation error of parameter distribution for multimodal distribution. At the same time, it can better comprehensively evaluate the KDE under different bandwidths. In short, a new method of optimal bandwidth selection is proposed in the original method.

show abstract

“…(2,3) It also hinders the appropriate operation of sensors, resulting in unexpected data. (4)(5)(6) Sensors are usually small and light but require high reliability. Thus, the vibration of ships with diesel engines is a major problem to be solved.…”

Section: Introductionmentioning

confidence: 99%

Effect of Center of Mass on Vibration-sensing Technology for Diesel Engine

Liao¹,

Yu²,

Song³

et al. 2021

Sensors and Materials

View full text Add to dashboard Cite

Different centers of mass (CMs) of a diesel engine were set to find their effects on the vertical vibration of an engine. To understand the effects and reduce vibration during sensing, we changed the location of the CM of the engine and performed a simulation to obtain data such as the vertical movements, velocity, torque, and acceleration of the vertical springs of isolators. The data showed that a 0.1 mm higher CM than the original CM of the engine reduced the amplitude of the vertical vibration by 4.35% and the velocity of the vertical vibration by 45.7% at a fixed rotation speed of the engine of 1000 revolutions per minute (rpm). Different CMs did not significantly affect the torque and acceleration. Instead, the phase of the vertical movements, the velocity, and the acceleration significantly changed with the CM in the simulation. The results imply that the CM needs to be considered in the design of the vibration isolation system of a diesel engine to increase the accuracy and lifetime of sensors used in vehicles under severe vibration conditions.

show abstract

Kernel-based support vector machines for automated health status assessment in monitoring sensor data

Cited by 28 publications

References 31 publications

A context-aware unsupervised predictive maintenance solution for fleet management

A context-aware unsupervised predictive maintenance solution for fleet management

Compilation method of CNC lathe cutting force spectrum based on kernel density estimation of G-SCE

Effect of Center of Mass on Vibration-sensing Technology for Diesel Engine

Contact Info

Product

Resources

About