Fuzzy logic modeling of surface ozone concentrations

Abstract Due to the complex relationships and the necessity for forecasts in atmospheric studies, air pollution modeling is a task for which fuzzy logic methods are amicably suited. This research investigates the ability to predict surface ozone concentration with the use of an automated fuzzy logic method, termed modified learning from examples (MLFE). Hourly ozone concentrations during summer months in the city of Edmonton are predicted with MLFE models and the results are compared to models used by Environment Canada. The root mean square error, mean absolute error and scatter plots are used to compare the results of the MLFE, CHRONOS and CANFIS models. The newly developed model captures the trends in ozone concentrations, and based on the statistical comparisons, the MLFE consistently shows good agreement with the measured data. The MLFE model compares favourably with CHRONOS and CANFIS and is easier to implement.

[1]  Alex J. Cannon,et al.  Forecasting Summertime Surface-Level Ozone Concentrations in the Lower Fraser Valley of British Columbia: An Ensemble Neural Network Approach , 2000, Journal of the Air & Waste Management Association.

[2]  Aldo Cipriano,et al.  Forecasting ozone daily maximum levels at santiago, chile , 1998 .

[3]  W. R. Burrows.  Combining classification and regression trees and the neuro-fuzzy inference system for environmental data modeling , 1999, 18th International Conference of the North American Fuzzy Information Processing Society - NAFIPS (Cat. No.99TH8397).

[4]  M. Mesbah,et al.  Modelling and analysis of ozone episodes , 2000, Environ. Model. Softw..

[5]  Oleg M. Pokrovsky,et al.  Fuzzy logic approach for description of meteorological impacts on urban air pollution species: a Hong Kong case study , 2002 .

[6]  A. Comrie.  Comparing Neural Networks and Regression Models for Ozone Forecasting , 1997 .

[7]  Kazuo Tanaka,et al.  Identification and analysis of fuzzy model for air pollution-an approach to self-learning control of CO concentration , 1992, Proceedings of the 1992 International Conference on Industrial Electronics, Control, Instrumentation, and Automation.

[8]  Gilles Mourot.  Modelling of ozone concentrations using a Takagi-Sugeno model , 1999 .

[9]  Stephen Yurkovich,et al.  Fuzzy Control , 1997 .

[10]  William R. Burrows,et al.  CART Decision-Tree Statistical Analysis and Prediction of Summer Season Maximum Surface Ozone for the Vancouver, Montreal, and Atlantic Regions of Canada , 1995 .

[11]  D. Koracin,et al.  SCOS97-NARSTO 1997 southern California ozone study and aerosol study: Volume 2 -- Quality assurance plan. Final report , 1999 .

[12]  William F. Ryan,et al.  Air Quality Forecasts in the Mid-Atlantic Region: Current Practice and Benchmark Skill , 2000 .

[13]  Scott M. Robeson,et al.  Evaluation and comparison of statistical forecast models for daily maximum ozone concentrations , 1990 .

[14]  Leo Breiman,et al.  Classification and Regression Trees , 1984 .

[15]  T. Ross.  Fuzzy Logic with Engineering Applications , 1994 .