Alexandria Engineering Journal (Apr 2020)
Novel operation control strategy for conjugate high-low pressure air separation columns at different part loads
Abstract
Air separation industry can be considered as the major source provider of oxygen and nitrogen. It is an energy-intensive process that consumes a large amount of electrical energy. So, this study is concerning on doing improvements in both the process design and the process control in order to decrease the specific energy demands of the products. Therefore, a complete steady state simulation model of a heat integrated double distillation column of a cryogenic air separation unit using Aspen Plus software is presented. The simulation model results are compared with typical real process values showing a good agreement with a maximum deviation of 2.6%. Then, an optimization procedure of the main factors of the distillation columns is proposed. This optimization causing the yields of oxygen and nitrogen to increase from 95% to 99.65% and from 90% to 99.9% respectively while the specific energy demand of nitrogen (at 8 bar) is decreased from 0.25 kW/Sm3 to 0.18 kW/Sm3. After that, a sufficient control system is installed on the simulation model by using simple single input single output (SISO) PI controllers. This system links between the controlled and manipulated variables. Also, it includes a novel operation control strategy to give the plant the ability to be operated under different production modes (liquid oxygen, liquid nitrogen, or both) at wide range of partial loads (typically from 100% to 25%). A careful study of tuning of the critical controllers is provided to select the most proper gain (K) and integral time (IT) values which achieve the maximum purities and the desired product quantities in each different case while the specific energy demands of products are within range. Finally, the performance of the plant under 42 different operation cases is presented for 24 h. These cases include different production modes under different partial loads. The results show that the optimized model, with the proposed control system, can be operated under wide range of partial loads during different production modes while the products qualities are in allowable ranges. This availability of working under any part load percentage led to the provision of a large part of the energy consumed.
