Model identification of 17h Ozone concentration over the Basse-Normandie region

Abstract

In this paper, Takagi-Sugeno (TS) approach is resorted to get a multi-model description of the ozone generation process in a specific geographical zone at a fixed time of the day. Here, attention is paid to 17 hour in the Basse-Normandie region. The focus is made on this hour because in the considered region the maximum value of the ozone occurs at this time. TS-based multi-model design necessitates the selection of influent inputs, membership functions (for the different input variables) and local models structure. The selection of influent input variables has been done according to the (hourly) correlation between the output and all input variables. It is worth noting that the wind direction plays a crucial role and needs a specific membership function. The identification scheme involves a least square type estimation algorithm to estimate the local model parameters. As the number of identified parameters grows exponentially with the number of inputs a compromise should be done between the model accuracy and complexity. The model thus obtained turns out to be satisfactory and is currently used to build-up a predictor for the ozone in the region of interest.