MATEC Web Conf.
Volume 157, 2018Machine Modelling and Simulations 2017 (MMS 2017)
|Number of page(s)||7|
|Section||Modelling and simulation, structural optimization|
|Published online||14 March 2018|
Missing values reconstruction in sound level monitoring station by means of intelligent computing
Kielce University of Technology, Faculty of Mechatronics and Mechanical Engineering, Aleja Tysiąclecia Państwa Polskiego 7, 25314 Kielce, Poland
2 University of Žilina, Faculty of Mechanical Engineering, Univerzitná 8215/1, 01026 Žilina, Slovakia
* Corresponding author: firstname.lastname@example.org
The aim of the paper was to reconstruct the missing data by applying the model which describes variability of sound level in the whole period from 2013 to 2016. To build the model, the computational intelligence methods, like fuzzy systems, or regression trees can be used. The latter approach was applied and we built the model with Cubist regression tree software, using equivalent sound levels recorded in 2013. For the reconstruction of sound level data in short period of time (several days), time series values and day_of_week values together should be used in the training dataset. For the reconstruction of sound level data in long period of time (several months) day_of_week values should be used in the training dataset.
Key words: reconstruction of missing sound level data / random forest / regression trees
© The Authors, published by EDP Sciences, 2018
This is an Open Access article distributed under the terms of the Creative Commons Attribution License 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. (http://creativecommons.org/licenses/by/4.0/).
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.