1,721,014 research outputs found
Health Indicator Effectiveness in Localized Fault Diagnosis: Rolling bearing elements
No full textConstant tracking of a state indicator curve derived from signal features defines the main body of the prediction in the remaining useful life of assert or its components. In this paper, the effectiveness of the pre-processing techniques to extract suitable health indicators and trace fault degradation on rolling element bearing is studied on a practical dataset. As a first step, the envelope analysis is performed to evaluate characteristic frequencies associated with inner and outer race faults. Then, its amplitude spectrum’s effectiveness as a health indicator regarding the fault evaluation is analyzed. As results show, due to the non-stationary nature of the vibration signal, this approach can be used to identify the type of fault in addition to the Root Mean Square value as the main health indicator to have an estimation on the level of degradation of bearing
State of Health analysis for Lithium-ion Batteries considering temperature effect
No full textLithium-ion batteries have become an integral component of machines and products in every field of modern life. In order to assure optimal use of the batteries, it is necessary to accurately predict their various parameters such as state-of-health (SoH), end of life (EoL) and state-of-charge (SoC). In this paper the use of the third-degree polynomial and hybrid function for SoH estimation and remaining useful life (RUL) prediction are further validated on a different dataset. Furthermore, linear interpolation is used to enlarge the dataset and achieve more accurate results. Finally, the battery state-of-health estimation in terms of temperature dependency is analyzed
The impact of Internet transmission on the uncertainty in the electric power quality estimation by means of a distributed measurement system
No full textThere is increasing evidence, in literature, that the estimation of the electric power quality requires the simultaneous measurement of several quantities and indices, in all lines connected to the same point of common coupling. The increase in the performance that the measuring systems based on digital signal processing techniques has undergone during recent years and the capability of the digital systems of interconnecting and exchanging data are making these systems more and more appealing and cost-effective for power quality applications. Moreover, the availability of a world-wide, low-cost, and public-domain interconnection system, the Internet, is pushing the evolution of the remote measurement systems, where the measurement results provided by in-field measurement systems are collected and stored by a central unit, toward the distributed measurement systems, where different systems, located in different places, share the same data in order to perform a measurement. It is known that the major drawback of these systems is the lack of synchronization of the shared data, due to the variable and unpredictable throughput of the net, which may affect the uncertainty of the result of the measurement in a quite significant way. This paper analyzes a distributed measurement system for electric power quality measurements and shows how the possible detrimental effects of data transmission over an Internet connection can be reduced by means of a suitable use of averaging techniques, thus avoiding a strict and expensive synchronization between the different units of the distributed measurement system. At last, an estimate of the effects of the possible transmission delays on the measurement uncertainty is given
Battery Remaining Useful Life Prediction Supported by Long Short-Term Memory Neural Network
No full textThe rise of renewable energy and electric vehicles has led to enormous growth and development in the field of lithium-ion batteries. Ensuring long-life and safe operation of these batteries requires accurate estimation of key parameters such as state of charge, state of health (SoH), and remaining useful life (RUL). In this paper, a long short-term memory neural network (LSTM NN) is presented for RUL prediction. Furthermore, the predictors used are discussed in detail, and a comparison between the two models is presented. The network has been trained and tested on a substantial dataset of 124 batteries, aged under various fast charging conditions, and published by the Toyota Research Institute in collaboration with MIT and Stanford. Despite their vastly different cycle lives, the proposed LSTM NN structure has performed very accurate RUL prediction for all tested cells
A Transfer Learning Approach for Remaining Useful Life Estimation of Lithium-Ion Batteries
No full text—Industry 4.0 has reimagined how businesses manufacture and distribute their products. To advance further towards
sustainability, the field of Fault Diagnosis and Prognosis (FDP)
assumes great significance. The ability to predict failures or to
know when a component will reach the end of its operational
life can significantly mitigate maintenance and replacement costs.
Machine Learning (ML) methods have exhibited the ability to
extract trends and models from complex datasets, becoming wellsuited for FDP tasks. When dealing with FDP, one of the main
problems is the difficulty of obtaining large datasets, due to
the burden of conducting extensive laboratory tests, and their
usual unbalance, for the impossibility of simulating every possible
anomaly that could ever happen. Being able to generate new
synthetic data, or to adapt a pre-trained model to another similar
task, becomes of paramount importance. In this paper, we focus
on lithium-ion batteries. Several commonly used devices are
usually powered using lithium-ion batteries, and each of these
batteries can assume a completely different behavior from its
peers based on usage, charging, and many other factors, leading
to potential harm, unreliableness, and other major potential
issues. We propose a convolutional Long Short-Term Memory
(LSTM) neural network with attention for estimating Remaining
Useful Life (RUL) of lithium-ion batteries. The model will be
trained on a source dataset, and then re-trained on a smaller
target dataset to establish the possibility of applying domain
adaptation and transfer learning to RUL estimation, allowing
for fast deployment and cost reduction in the production phase.
Results show that the use of transfer learning helps to increase
the performance of the model, obtaining on the target dataset an
accuracy similar to that of the source dataset
State of Health Estimation Procedure for Lithium-Ion Batteries Using Partial Discharge Data and Support Vector Regression
Full text linkBattery aging is a complex phenomenon, and precise state of health (SoH) monitoring
is essential for effective battery management. This paper presents a data-driven method for SoH
estimation based on support vector regression (SVR), utilizing features built from both full and partial
discharge capacity curves, as well as battery temperature data. It provides an in-depth discussion
of the novel features constructed from different voltage intervals. Moreover, three combinations of
features were analyzed, demonstrating how their efficacy changes across different voltage ranges.
Successful results were obtained using the full discharge capacity curves, built from the full interval
of 2 to 3.4 V and achieving a mean R2 value of 0.962 for the test set, thus showcasing the adequacy
of the selected SVR strategy. Finally, the features constructed from the full voltage range were
compared with ones built from 10 small voltage ranges. Similar success was observed, evidenced by
a mean R2 value ranging between 0.939 and 0.973 across different voltage ranges. This indicates the
practical applicability of the developed models in real-world scenarios. The tuning and evaluation of
the proposed models were carried outusing a substantial dataset created by Toyota, consisting of
124 lithium iron phosphate batteries
A possibilistic approach for measurement uncertainty propagation in prognostics and health management
Full text linkIn this paper, a similarity-based data-driven prognostic algorithm for the estimation of the Remaining Useful Life of a product is proposed. It is based on the exploitation of run-to-failure data of products, which are supposed to be characterized by similar operational conditions. The core of the contribution is the application of a possibilistic framework, namely a Random-Fuzzy Variable approach, for the representation and propagation of the measurement uncertainty, which is a crucial source of uncertainty in Prognostics and Health Management. The results obtained for a real application case as Medium and High Voltage Circuit Breakers, have shown a high prognostic power of the algorithm, which therefore represents a potential tool for an effective Predictive Maintenance strategy
Virtual Sensors: A Tool to Improve Reliability
No full textVirtual sensors is a powerful concept applicable in new application domain like health-care, entertainment, fitness, sport, social interaction etc. tanks its capability of integration of new components and functionality. Actually, virtual sensors can potentially use heterogeneous physical sensors in order to combine different types of data to compute a measurement. The concept 'virtual Sensor' allows to apply the principle so-called analytical redundancy. This point is important because it allows to also achieving the objective of extending reliability and availability of the application domain without adding extra hardware complexity. Starting from the previous consideration, Virtual Sensors (VS) can be candidate to offer a solution to reduce the criticality of high-risk failures increasing their detectability and predicting the related failure mode. VS and the validation procedure in the measurement process allowed by their use, offer a new tool in the Industry 4.0 scenario. In the paper, an example of virtual sensor devoted to monitoring different sections of an inverter dedicated to photovoltaic application will be analyzed. In particular, will be shown how it is possible to monitor the temperature of the CPU employed on the control board
Modern Digital Twin for Validation and Generation of Datasets for Machine Tool Spindle Modeling
No full text2024 International Symposium on Power Electronics, Electrical Drives, Automation and Motion, SPEEDAM 2024
- …
