Alberto Servida

Monitoring Pollutant Emissions in a 4.8 MW power Plant through Neural Network

Abstract The prime aim of this paper is to address the relevant issues associated to the development of neural-based software sensors for monitoring the pollutant emissions coming out from combustion chambers. The objective is to prove the potential of software sensors as alternative monitoring systems to conventional analytical equipment. The preliminary results refer to a 4.8 MW power pilot plant operating at the Enel Santa Gilla Research Center in Cagliari (Italy).

Monitoring of a CO oxidation reactor through a grey model-based EKF observer

Abstract Often, in real applications it is difficult to dispose of a simple, yet, representative kinetic model because of the complexity of the reactions taking place. To overcome this limitation a hybrid modelling approach is proposed for the identification of the dynamic behaviour of chemical reactors. In particular, the tools of neural network modelling have been exploited to represent the kinetic reaction data. The “neural reaction rate model” is integrated within a first principles model that constitutes the basis of a nonlinear observer extended Kalman filter (EKF) for an heterogeneous gas–solid reactor where the catalytic oxidation of carbon monoxide takes place. The outlined procedure shows that artificial neural networks (ANN) can be effectively used to formulate lumped reaction rates because of their capability in capturing the essential characteristics of the functional relationship among the state variables.

Analysis of detailed kinetic schemes for combustion processes: Application to a methane-ethane mixture

The combustion kinetics of equimolar ethane-methane mixtures have been investigated in a perfectly jet stirred reactor in the temperature range 923–1,080 K and for normalized fuel/O2 stoichiometric ratios of 0.24–1.5. The concentrations of the main molecular species have been measured by probe sampling and GC analysis. The experimental data have been compared with the model predictions of four detailed kinetic mechanisms previously reported in the literature. A simple procedure to extract the most important reactions from a detailed kinetic scheme has also been developed. With this procedure a reduced model has been identified which is able to reproduce the behavior of the original detailed model in the entire range of operating conditions investigated in the experimental part of this work. A good reduction in the number of reactions has been obtained, while the number of involved chemical species has been only slightly reduced.

A feedforward control strategy for distillation columns

Abstract The prime objective of this work is to demonstrate the potential of neural network modeling for advanced nonlinear control applications. In particular, for the case of a single composition distillation column, a model-based neural controller is developed to regulate the composition of the distillate stream. The neural controller relies on process inversion for the evaluation of the actuator action on the manipulated variable (reflux flowrate) to maintain the controlled variable (distillate composition) at the prescribed value. The performance of the neural controller is assessed and compared with that of a conventional temperature control loop and of a neural inferential control structure. The neural controller by far outperforms the other two in terms of the response speed by which the upsetting loads are compensated.

A NONLINEAR OBSERVER BASED ON HYBRID MODELLING OF CHEMICAL REACTORS

Abstract The successful design of an observer for inferring the outlet composition from a chemical reactor heavily relies on the goodness of the adopted kinetic rate model (Baratti et al., 1993). On the other hand, often, it is difficult to dispose of a simple, but, exhaustive kinetic model because of the complexity of the reaction scheme one has to deal with. In this work, we explore the possibility to represent global (lumped) reaction rate laws by the use of neural network models. The aim is to develop a nonlinear observer (extended Kalman filter, EKF) of an heterogeneous gas-solid reactor that relies on a grey model where the “neural reaction rate” law is integrated within a first principles model. The procedure is outlined for the case of the catalytic oxidation of carbon monoxide over Pt-alumina catalyst. The results show that neural networks (NN) can be effectively used in representing lumped reaction rates since NN are able to capture the essential characteristics of the functional relationship relating the state variables.

Neural DMC control strategy for a CSTR in presence of noise

Abstract The prime objective of this work is to assess a neural DMC strategy in the case of experimental data corrupted with white noise. The neural DMC structure relies on the simple and innovative dynamic neural model recently developed. (Baratti et al., 2000). A nonisothennal CSTR was considered as benchmark. The perfonnance of the DMC strategy was evaluated in tenns of set-point tracking and disturbance rejection capabilities. The results show that the inaccuracy of the dynamic neural model is overcome by simply integrating the DMC structure with the available on-line measurements even though they are corrupted with white noise.