Identification and Estimation of temperature in Nitration process

The nitration of organic compounds is among the basic reactions used in synthesis of many compounds used in agrochemicals industries, pharmaceuticals, dyes and agriculture. The Nitration process are usually carried out as batch process in aqueous phase as they are very sensitive and liable to explode. A very close control of heat and mass transfer is necessary as it leads to selectivity of reaction under safe condition. This prevents the thermal loss of control of the system, so that the reaction temperature rise can be limited below a critical value. The Present work carries out the Identification and Estimation of model states of a runaway reaction. The temperature data are collected from exothermic reactor and the optimal model is identified using NARX method. The Kalman, Extended Kalman and Unscented Kalman filter dynamics are studied and implementation and design of the optimal filters are carried out. The filter algorithms are used to estimate the concentration and temperature in the presence of random noise in measurement. The nonlinear estimators are compared for selection of best estimator. The results are found motivating for development of control.