DBPapers
DOI: 10.5593/SGEM2016/HB43/S06.023

OZONE LEVEL PREDICTION MODELS USING ANT COLONY OPTIMIZATION ALGORITHM

S. Gallova
Friday 11 November 2016

References: 16th International Multidisciplinary Scientific GeoConference SGEM 2016, SGEM Vienna GREEN Extended Scientific Sessions, www.sgemviennagreen.org, SGEM2016 Conference Proceedings, ISBN 978-619-7105-82-7 / ISSN 1314-2704, 2 - 5 November, 2016, Book 4 Vol. 3, 169-180pp, DOI: 10.5593/SGEM2016/HB43/S06.023

ABSTRACT

There are many situations where human activities have significant effects on the environment. Ozone layer damage is one of them. Modeling ozone series from satelite observations, past data and its relationship with meteorological variables is an important part of proposed method. Association rules are used to explore the properties of the data, instead of predicting the class of new data. There exist many algorithms for obtaining association rules from a dataset. We use an effective technique for association rules discovery that is based on evolutionary algorithms, which have been used for the optimization and adjustment of models in solved tasks. A proposed evolutionary search algorithm uses the principles of Ant colony optimization algorithm with some important improvements. Simulation results show that the proposed algorithm in comparison with the traditional ant colony algorithms provides better performance. It achieves more balanced transmission among the node, obtains lower path costs and reduces the energy consumption of the routing. The proposed algorithm has a greater performance in comparison with other evolutionary techniques. It is an effective optimization tool for ozone control strategies implementation.

Keywords: ozone modeling, ant colony optimization, pheromone, association rule, training dataset, control, expert system

PAPER DOI: 10.5593/SGEM2016/HB43/S06.023, OZONE LEVEL PREDICTION MODELS USING ANT COLONY OPTIMIZATION ALGORITHM

35 EURO ADD TO CART


Your cart ( items)

Subtotal:
Tax cost ():
Shipping cost:
Total:
Checkout Empty