Volume 15, Issue 3

Distillation columns are among the most common fractionation systems with numerous applications in petrochemical units. Hence, the optimization of these columns is a large step in reducing energy consumption and increasing process productivity. This study was, therefore, carried out as a case study of the simulation and optimization of the parameters influencing the ethylene production of the ethylene distillation column in an olefin unit. The two defined objective functions in this research were maximum mass flow of ethylene in the upstream flow of the distillation column and the minimum energy consumption in the distillation column. The optimal operating conditions for the independent variables were estimated using the NSGAII algorithm. The sensitivity analysis of the results was, thereafter, carried out and the optimization results introduced tray no. 71 as the most suitable feed location. In addition, the optimal reflux ratio and the optimal feed flow temperature were 5.26 and −18.49 °C, respectively. In this condition, the upstream ethylene flow rate and energy consumption in the unit increased by approximately 0.74 % and 0.9 % as compared to the initial conditions, respectively.

An improved model reduction scheme is proposed here for higher-order processes and subsequently an enhanced IMC-PID controller is designed based on the obtained reduced model. In the proposed scheme, higher-order processes are estimated as first-order-plus-dead-time (FOPDT) model. Designed IMC controller includes a filter having two separate time constants with fractional order coefficients. Efficacy of the proposed model reduction scheme is verified in terms of closed loop performance evaluation for higher-order minimum and non-minimum phase process models in comparison with improved SIMC ( i SIMC) controller (Grimholt. Optimal PI and PID control of first-order plus delay processes and evaluation of the original and improved SIMC rules. J Process Control. 2018;70:36–46). Overall performance enhancement for the proposed method is demonstrated through simulation study as well as real-time experimentation on a level control loop.

At condensate stripper of a cracker plant with design control philosophy for composition control pant was facing operational difficulty. Due to disturbance in operating parameter column was becoming unstable and product was getting offspec w.r.t. desired purity. One of the applications of dynamic simulation is to troubleshoot the challenges related to control philosophy in practical application. Since steady-state simulation models cannot predict behavior with respect to time, initially steady state model and finally a dynamic model was developed in Aspen HYSYS. The model is used to study the process behavior for existing control philosophy and proposed philosophy. To avoid column puncture and without waiting for plant shut down the existing Temperature Indicator (TI) considered as Temperature Indicator Controller (TIC) for the study. A new control philosophy was developed based on the response of variables after disturbances in feed rate and composition. The revised control philosophy has been implemented and is now working satisfactorily, providing stabilized operation of the column with consistent bottom product quality. This has helped to reduce the loss of C 2 s in the bottom stream by about 700 ppm, for savings of about $100,000 USD per year.

The control design of coupled tanks is not an easy task due to the nonlinear characteristic of the valves, and the interactions between the controlled variables. Those features pose a challenge in the automatic control, so that linear controllers, such as conventional PID, might not work properly for regulating this MIMO system. Some advanced control techniques (e. g. control based on neural networks) can be used since neural networks are universal approximators which can deal with nonlinearities and interactions between process variables. In the present work, an experimental investigation was performed presenting a comparison between two neural network-based techniques and testing the feasibility of these techniques in the coupled tanks system. First principles simulations helped to find suitable parameters for the controllers. The results showed that the model predictive control based on artificial neural networks presented the best performance for supervisory tests. On the other hand, the inverse neural network needed a very accurate model and small plant-model mismatches led to undesirable offsets.

The usage of nanofluids and modification of tube geometry are the two most prominent heat transfer enhancement methods employed to improve the performance of thermal devices. In this work, the combined effect of these methods has been studied by CFD modelling of developing and Graetz laminar flow in flattened tubes with ZnO – water nanofluid. For the purpose of comparison, simulation with water and circular tube has also been carried out. Performance evaluation has been done using PEC, PER and entropy generation. Results reveal that tube flattening has more pronounced effect on both heat transfer and flow compared to that of nanofluid. An optimum tube flattening in terms of aspect ratio and nanofluid concentration has also been identified for this kind of flow. Flattened tube with aspect ratio 6 with 1 % ZnO-water nanofluid has been found to yield the highest entropy generation reduction of 13.24 %

Currently, air separation units (ASUs) have become very important in various industries, particularly oil and petrochemical industries which provide feed and utility services (oxygen, nitrogen, etc.). In this study, a new industrial ASU was evaluated by collecting operational and process information needed by the simulator by means of HYSYS software (ASPEN-ONE). The results obtained from this simulator were analyzed by ASU data and its error rate was calculated. In this research, the simulation of ASU performance was done in the presence of an expansion turbine in order to provide pressure inside the air distillation tower. Likewise, the cooling capacity of the cooling compartment and the data were analysed. The results indicated that expansion turbine is costly effective. Notably, it not only reduces the energy needed to compress air and supply power of the equipment, but also provides more cooling power and reduces air temperature. Moreover, turbines also increase the concentration of lighter gas products, namely nitrogen.

Motivated by the oxidative power of hydrogen peroxide and its environmentally attractive properties, the present study aimed to determine the optimum conditions for the production of ferric sulfate coagulant from ferrous sulfate The independent variables studied were the temperature (7.5–27.5 °C), amount of hydrogen peroxide (100–300 %) of the stoichiometric amount for the oxidation reaction, and dilution of the reaction medium using water (100–300 %) of the stoichiometric amount for the oxidation reaction. For the optimum conditions achieved, it was possible to obtain a conversion of 96.17 % of Fe +2 to Fe +3 , using a small hydrogen peroxide excess of 50 %, resulting in a product suitable for use as a coagulant in water treatment. It was found that an oxidation potential (Eh) greater than 0.7 volts corresponded to the conversion of Fe +2 to Fe +3 greater than 90 %, indicating Eh to be useful for reaction control in an industrial process.

Different control strategies have been investigated to improve nonlinear system operations. One such strategy is the use of nonlinear predictive controllers (NMPCs) based on machine learning models. These models, such as artificial neural networks (NN), support vector machines (SVMs), and neuro-fuzzy networks (NF), present satisfactory adaptability to the complexity of the processes. In this aspect, a comparative study of the models in the predictive control of a complex system, such as MIMO (multiple-input-multiple-output) process of the production process of cyclopentadiene, is of interest and is the aim of this work. In this aspect, we find, through simulations, that the NMPCs presented adequate performance, especially those based on an SVM, concerning the servo and regulatory problem scenarios, keeping the process at the optimum operating point, especially for unattainable setpoint. The instability in the use of the classical proportional-integral-derivative linear control is also shown.