Nonylphenol biodegradation kinetics estimation using neural networks

Abstract

strated that it is possible to determine intrinsic kinetics of single organic compounds by using oxygen uptake data from electrolytic Manyvmanmeceal substaes are on und hemocs respirometry. This is relatively inexpensive and gives fast, accurate and reliable data. Direct nonylphenol concentration measurements A widely used alkylphenolethoxylates (APEs) surfactant was rerequire an exhaustive steam distillation extraction procedure becently banned in Europe because scientists discovered that APE fore using high pressure liquid chromatography for analysis. This breakdown products are estrogenic and highly toxic to aquatic oris more expensive to implement.ganisms Nonylphenol is one such substance that has come under The purpose of this paper is to estimate the unknown intrinthe focus as an environmental pollutant. However, sufficient insic kinetic parameters and states through simulation based on the formation is not there to study the kinetic behavior of this toxic neural network concept. These estimating parameters are part of surfactant. The biodegradation process of nonylphenol is best dean existing mathematical biodegradation model of coupled nonlinscribed by Monod's model which is based on a coupled system of ear differential equations known as Monod's model. The model nonlinear differential equations. This model is based on set of kihas three states, namely, oxygen uptake, substrate (or pollutant) netic parameters. It is very difficult to measure the actual biodegraconcentration and bacterial cell mass concentration. Only the oxydation process of nonylphenol because of the unknown nature of gen uptake is experimentally measurable by means of electrolytic the parameters involved and expense in measuring the states. The respirometry [4]. This paper presents a neural network based apestimation of kinetic parameters of nonylphenol biodegradation is proach for high-level performance estimation, which easily adapts done by using a gradient optimization neural network estimator. to the non-linear behavior of the kinetic parameters. Through simulation the behavior of the three states can be studied using neural