Search

The main objective of this study is utilization of recurrent neural networks to categorize pressure derivative plots of well-testing data into various reservoir models. The training and test data have been generated through an analytical solution of commonly used reservoir models. The accuracy of the designed recurrent neural networks has been examined by the simulation test data and actual field data. The accuracy of the developed recurrent neural networks has been compared to a multilayer perceptron neural network. The results indicate that the recurrent neural networks can identify the correct reservoir models from test data with an accuracy of 98.39%, while multilayer perceptron neural networks represent an accuracy of 95.83%.

In this work, a reduced form of dynamic flux balance model based on artificial neural networks for batch and fed-batch fermentation of xylose-utilizing engineered Saccharomyces cerevisiae RWB 218 is developed. The intracellular description of carbon metabolism in the model is simulated by a multilayer-perceptron (MLP) network. First, this hybrid model is compared to the full mechanistic dynamic flux balance analysis (DFBA) in terms of accuracy and computational time regarding available experimental data on anaerobic batch cultivation. Afterwards, it is used in a model-based sequential dynamic optimization procedure in order to maximize ethanol productivity. The initial liquid volume charged in the bioreactor, the feed flow rates in aerobic and anaerobic conditions, the final batch time, and the switching time from aerobic to anaerobic conditions are considered as decision variables. Differential evolution (DE), as a robust and efficient optimization method, is employed to solve the problem for several glu...

An optimal temperature profile is determined for a methanol synthesis reactor of LURGI type. The temperature profile is estimated so that methanol production rate in the reactor outlet will be maximized. First, the reactor is simulated based on heterogeneous one- and two-dimensional models. The comparison of the simulation results and plant data shows that the heterogeneous one-dimensional model can reliably be used for determining optimal temperature profile. Since optimal temperature profile for reversible exothermic reaction in tubular reactors is a decreasing function of reactor length, the technique of control variable parameterization is used for determining optimal temperature profile in a methanol reactor. In this way, a third order polynomial is considered for the temperature profile and the polynomial coefficients are as decision variables. The optimization is based on a Quasi-Newton's method (BFS technique), and the objective function is methanol flow rate at the reactor outlet. The results of ...