Application of Artificial Neural Networks in Condition Based Predictive Maintenance

Krenek, Jiri; Kuča, Kamil; Blažek, Pavel; Krejcar, Ondřej; Jun, Daniel

doi:10.1007/978-3-319-31277-4_7

Cited by 23 publications

(5 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…), predict failures and risks or traffic flows [ 32 ]. ANN was used in this work to design a predictive maintenance model, ANNs are commonly used for this purpose due to their efficacy in this area [ 33 ]. In addition to neural networks, there are other models that can be employed in prediction systems, such as Support Vector Regression (SVR) or different linear and nonlinear models.…”

Section: Related Workmentioning

confidence: 99%

GreenVMAS: Virtual Organization Based Platform for Heating Greenhouses Using Waste Energy from Power Plants

González‐Briones

Chamoso

Yoe

et al. 2018

Sensors

View full text Add to dashboard Cite

The gradual depletion of energy resources makes it necessary to optimize their use and to reuse them. Although great advances have already been made in optimizing energy generation processes, many of these processes generate energy that inevitably gets wasted. A clear example of this are nuclear, thermal and carbon power plants, which lose a large amount of energy that could otherwise be used for different purposes, such as heating greenhouses. The role of GreenVMAS is to maintain the required temperature level in greenhouses by using the waste energy generated by power plants. It incorporates a case-based reasoning system, virtual organizations and algorithms for data analysis and for efficient interaction with sensors and actuators. The system is context aware and scalable as it incorporates an artificial neural network, this means that it can operate correctly even if the number and characteristics of the greenhouses participating in the case study change. The architecture was evaluated empirically and the results show that the user’s energy bill is greatly reduced with the implemented system.

show abstract

Section: Related Workmentioning

confidence: 99%

GreenVMAS: Virtual Organization Based Platform for Heating Greenhouses Using Waste Energy from Power Plants

González‐Briones

Chamoso

Yoe

et al. 2018

Sensors

View full text Add to dashboard Cite

show abstract

“…From a computational point of view, a work activity is a complex of data collection, processing, and consumption. Novel techniques in industrial settings are in fact focusing on the data themselves, as an advancement over traditional model-based approaches (Krenek et al, 2016). The core of databased techniques is to take full advantage of the huge amounts of available process data, and intend to provide efficient alternative solutions for different industries, with a limited need for the modeling and configuration of the systems.…”

Section: Cyber Physical Systems (Cps)mentioning

confidence: 99%

Energy Saving by Blockchaining Maintenance

Albano

Sharma

Campos

et al. 2019

JIEMS

View full text Add to dashboard Cite

The development and interest in Industry 4.0 together with rapid development of Cyber Physical Systems has created magnificent opportunities to develop maintenance to a totally new level. The Maintenance 4.0 vision considers massive exploitation of information regarding factories and machines to improve maintenance efficiency and efficacy, for example by facilitating logistics of spare parts, but on the other hand this creates other logistics issues on the data itself, which only exacerbate data management issues that emerge when distributed maintenance platforms scale up. In fact, factories can be delocalized with respect to the data centers, where data has to be transferred to be processed. Moreover, any transaction needs communication, be it related to purchase of spare parts, sales contract, and decisions making in general, and it has to be verified by remote parties. Keeping in mind the current average level of Overall Equipment Efficiency (50%) i.e. there is a hidden factory behind every factory, the potential is huge. It is expected that most of this potential can be realised based on the use

show abstract

“…Datos financieros han sido pronosticados a partir de redes neuronales (Tealab, 2018) entre otras áreas, haciendo a un lado métodos más sencillos, por lo que el futuro de los métodos predictivos debe cambiar y enfocarse a modelos más fuertes que se adapten a las necesidades actuales, y un aspecto importante del artículo citado es que su teoría fue medible y se validó en campo, demostrando resultados satisfactorios en la predicción de fallos. Las redes neuronales ofrecen una poderosa herramienta para evaluar datos y parámetros de las máquinas que pueden aprender de los datos de proceso de la simulación de fallas (Krenek et al, 2016).…”

Section: Discusionunclassified

Predicción de Fallos Mecánicos en Equipos de Envoltura

Rodas

Castrillón

2019

Inf. tecnol.

View full text Add to dashboard Cite

El objetivo de este trabajo es el pronóstico de fallos mecánicos en equipos de envoltura partiendo de datos históricos para mejorar la programación de intervenciones preventivas y disminuir el número de correctivas. Para esto se debe recolectar y clasificar datos históricos de tiempos entre fallos, realizar modelos de correlación y a partir de estos resultados definir cuál modelo de pronóstico es mejor, validando los datos mediante pruebas de hipótesis que reduzcan al mínimo la elección de este. Por último, aplicarlo, llegar a resultados y determinar tiempos aptos de programación para mantenimientos preventivos. La metodología empleada condujo al planteamiento de una ecuación que se adapta a las variables que afectan de manera directa el pronóstico del fallo. Se concluyó el impacto positivo de los paros programados en una organización, reduciendo el intervalo de tiempo de intervenciones preventivas y proyectando los tiempos de intervención para el siguiente año.

show abstract

Application of Artificial Neural Networks in Condition Based Predictive Maintenance

Cited by 23 publications

References 22 publications

GreenVMAS: Virtual Organization Based Platform for Heating Greenhouses Using Waste Energy from Power Plants

GreenVMAS: Virtual Organization Based Platform for Heating Greenhouses Using Waste Energy from Power Plants

Energy Saving by Blockchaining Maintenance

Predicción de Fallos Mecánicos en Equipos de Envoltura

Contact Info

Product

Resources

About