Virtual Reality in Context of Industry 4.0

Virtual Reality in Context of Industry 4.0

článek ve sborníku ve WoS nebo Scopus

en

In industrial environments, it is often difficult and expensive to collect a good amount of data to adequately train expert systems for regression purposes. Therefore the usage of already available data, related to environments showing similar characteristics, could represent an effective approach to find a good balance between regression performance and the amount of data to gather for training. In this paper, the authors propose two alternative strategies for improving the regression performance by using heterogeneous data, i.e. data coming from diverse environments with respect to the one taken as reference for testing. These strategies are based on a standard machine learning algorithm, i.e. the Artificial Neural Network (ANN). The employed data came from measurements in industrial plants for energy production through the combustion of coal powder. The powder is transported in air within ducts and its size is detected by means of Acoustic Emissions (AE) produced by the impact of powder on the inner surface of the duct. The estimation of powder size distribution from AE signals is the task addressed in this work. Computer simulations show how the proposed strategies achieve a relevant improvement of regression performance with respect to the standard approach, using ANN directly on the dataset related to the reference plant.

V průmyslovém prostředí, je často obtížné a drahé nasbírat odpovídající množství dat, pro expertní systémy pro regresní účely. Proto použití již dostupných údajů, týkajících se prostředí, které zobrazují podobné charakteristiky, může představovat efektivní přístup k nalezení dobrého poměru mezi výkonem a regresním množstvím shromažěných dat. V tomto článku jsou navrženy dvě alternativní strategie pro zlepšení regresních výstupů použitím heterogenních dat, tj data přicházející z různorodých prostředí s ohledem na jednu referenci pro testování.

In industrial environments, it is often difficult and expensive to collect a good amount of data to adequately train expert systems for regression purposes. Therefore the usage of already available data, related to environments showing similar characteristics, could represent an effective approach to find a good balance between regression performance and the amount of data to gather for training. In this paper, the authors propose two alternative strategies for improving the regression performance by using heterogeneous data, i.e. data coming from diverse environments with respect to the one taken as reference for testing. These strategies are based on a standard machine learning algorithm, i.e. the Artificial Neural Network (ANN). The employed data came from measurements in industrial plants for energy production through the combustion of coal powder. The powder is transported in air within ducts and its size is detected by means of Acoustic Emissions (AE) produced by the impact of powder on the inner surface of the duct. The estimation of powder size distribution from AE signals is the task addressed in this work. Computer simulations show how the proposed strategies achieve a relevant improvement of regression performance with respect to the standard approach, using ANN directly on the dataset related to the reference plant.

Virtual Reality; Industry 4.0; Factory of Future; Dynamics; FEM; Mikroskop atomárních sil; Mikrostruktura

Virtual Reality; Industry 4.0; Factory of Future; Dynamics; FEM; Atomic Force Microscopy; Microstructures

07.12.2016

Czech Technical University in Prague, Faculty of Electrical Engineering

Praha

978-80-01-05882-4

Mechatronika 2016

1–7

8