Due to high global warming potential (GWP) of hydrofluorocarbons (HFCs), the separation and recovery of HFCs from different refrigerant mixtures is an important issue. Most HFC mixtures are azeotropic in nature, thereby rendering the conventional distillation-based separation difficult and energy intensive. Extractive distillation (ED) with ionic liquid (IL) as solvent provides an attractive strategy for selective separation of HFC mixtures. However, systematic design and optimization of ED-based separation processes is nontrivial. In this work, we present SPICE_ED which is a software framework for the detailed design, synthesis, and techno-economic analysis of ED-based separation processes. The framework employs a building block representation followed by superstructure optimization that is able to automatically generate numerous design solutions and screen the best without requiring prior expert knowledge of candidate configurations. For a given IL as solvent and a set of design specifications, one can automatically determine the feasibility of the solvent and obtain the optimal process flowsheets that correspond to minimum energy consumption, minimum separation cost, or minimum emission/waste. We demonstrate the capability of SPICE_ED for the separation of R-410A (50 wt % R-32 and 50 wt % R-125) using [bmim][PF 6 ], a commonly used IL. Our optimized design requires an equivalent work of only 338.2 kJ/kg R-410A, which is about 48% less than the previously reported value of 656 kJ/kg. The newly identified design also achieves more than 47% and 27% reductions in CO 2 -equivalent emission (sustainability) and cost, respectively. Through multiobjective optimization, we further identify an operating regime to separate R-410A at a near-minimum cost without significantly increasing the energy consumption and emission. The processes obtained from SPICE_ED show excellent agreement with the key performance metrics when simulated in Aspen Plus, thereby establishing confidence in our designs as realistic and implementable.

Superior controllability of reactive distillation (RD) systems, designed at the maximum driving force (design-control solution) is demonstrated in this article. Binary or multielement single or double feed RD systems are considered. Reactive phase equilibrium data, needed for driving force analysis and design of the RD system, is generated through an in-house property prediction tool. Rigorous steady-state simulation is carried out in ASPEN plus in order to verify that the predefined design targets and dynamics are met. A multiobjective performance function is employed to evaluate the performance of the RD system in terms of energy consumption, sustainability metrics (total CO 2 footprint), and control performance. Controllability of the designed system is evaluated using indices like the relative gain array (RGA) and Niederlinski index (N I ), to evaluate the degree of loop interaction, as well as through dynamic simulations using proportional-integral (PI) controllers and model predictive controllers (MPC). The design-control of the RD systems corresponding to other alternative designs that do not take advantage of the maximum driving force is also investigated. The analysis shows that the RD designs at the maximum driving force exhibit enhanced controllability and lower carbon footprint than the alternative RD designs.

Valve stiction is a hidden menace in process control loops. The presence of stiction in control valves limits the control loop performance. Compensation of its effect is beneficial before the sticky valve can be sent for maintenance. This work is the first of a two part study on control valve stiction compensation. This part proposes a novel stiction compensation method, while the second part compares the performance of this proposed stiction compensation method with some of other compensation methods appeared in the literature. The proposed compensator is developed based on reduction of control action and addition of an extra pulse of finite energy as required. A method for estimating an appropriate parameter for reducing controller action has been developed. The proposed stiction compensator has been extensively evaluated using the MATLAB Simulink environment. The compensator has also been implemented in a pilot plant experimental setup. It has been found to be successful in removing valve stiction-induced oscillations from process variables both in simulation and pilot plant experimentation. The compensator developed in this study has the capability of reducing process variability with a minimum number of valve reversals.

No abstract

No abstract