INTERVAL OBSERVERS WITH GUARANTEED CONFIDENCE LEVELS APPLICATION TO THE ACTIVATED SLUDGE PROCESS

摘要

This paper deals with the design of robust observers for uncertain models, with application to the activated sludge process. We assume that the model is such that the bacterial growth rate is unknown, and the measurements and influent flow rates are disturbed. However we suppose that the upper and lower bounds of the uncertainties are known. Under appropriate hypotheses, we are able to build interval observers giving dynamic bounds containing the variables to estimate. Besides, if we know a priori the probability density of each uncertain parameter, we can synthesize interval observers with guaranteed confidence levels and provide a confidence density for the state. We apply this approach to the sludge process and compare the results to the probability density estimation obtained with Monte-Carlo simulations.