A flexible air separation process: 2. Optimal operation using economic model predictive control

Abstract

The penetration of renewable electricity promises an economic advantage for flexible operation of energy-intense processes. One way to achieve flexible operation is economic model predictive control (eNMPC), where an economic dynamic optimization problem is directly solved at controller level taking into account a process model and operational constraints. We apply eNMPC in silico to an air separation process with an integrated liquefier and liquid-assist operation. We use a mechanistic dynamic model as both controller model and plant surrogate. We conduct a closed-loop case study over a time horizon of 2 days with historical electricity prices and input disturbances. We solve the dynamic optimization problems in DyOS. Compared to the optimal steady-state operation, the eNMPC operating strategy gives a significant improvement of 14%. We further show that the eNMPC enables economic improvements similar to an idealized quasistationary scheduling. While the eNMPC provides control profiles qualitatively similar to those obtained from deterministic global optimization of quasistationary scheduling, the eNMPC satisfies the product purity constraints all the time whereas the quasistationary scheduling sometimes fails to do so. The eNMPC applies local optimization methods and achieves profiles similar to the scheduling solved using deterministic global optimization methods over the complete closed-loop simulation time horizon. K E Y W O R D S air separation unit, demand side management, direct single-shooting, economic model predictive control, integrated liquefication cycle 1 | INTRODUCTION The enhanced use of renewables, for example, onshore wind and photovoltaic, for electricity production results in fluctuating electricity prices. 1,2 Exploiting these fluctuations promises an economic advantage of flexibly operating electricity-intensive processes. There are two requirements: (a) a "flexible process", that is, a process with a wide operation range and without big losses of efficiency and (b) "flexible process operation", that is, exploiting the process flexibility in the operation. Thus, there is both economical and ecological incentive to take advantage of flexible process load adjustment as a reaction to changing market conditions. This is known as demandside management (DSM). 3 DSM has been applied to various process systems in the last decades, for example, by Ghobeity and Mitsos 4 and Brée et al. 5 Different concepts, recent advances and challenges of industrial DSM have been summarized by Zhang and Grossmann 6 and the rising need for flexible operation in context of the integration of renewable

Nonlinear scheduling with time‐variable electricity prices using sensitivity‐based truncations of wavelet transforms

“…Thus, promising future fields of application will certainly also include moving horizon applications involving nonlinear dynamic optimizations, which aim at integrating both the scheduling and the control perspective. The application of the presented algorithm for scheduling with lowdimensional dynamic surrogate models 47,54 and economic optimizations directly in the control layer using a detailed dynamic model 35,55,56 is thus left for future work.…”

“…42 For this purpose, we impose constraints on the change in production rates, as widely done in scheduling practice to ensure that load changes can be tracked without violating product requirements. 6,8,39 Revisiting the results presented in our previous work concerning the process dynamics, 35 Note that changing the control system to a more advanced one, such as nonlinear model predictive control, could involve substantially loosened ramping limits 43,44 and thereby increase potential savings, which should be thoroughly investigated in the future. Moreover, we acknowledge that numerous works argue to construct dynamic surrogate models 13,45,46 for explicitly capturing the process dynamics and demonstrate their successful application to the scheduling of ASUs.…”

“…The considered process configuration has been tailored to a high load flexibility and is described in detail our in previous works. 31,35 Most importantly, the process configuration comprises an integrated liquefaction cycle enabling a high ratio of storable liquid products. Moreover, the configuration allows for liquidassisted operation, where stored cryogenic LIN is fed back to the rectification section.…”

The Potential of Hybrid Mechanistic/Data‐Driven Approaches for Reduced Dynamic Modeling: Application to Distillation Columns

“…Herein, we provide a critical discussion of opportunities and limitations of such hybrid mechanistic/data-driven approaches for reduced dynamic modeling with a particular focus on the operation of distillation columns. Note that efforts addressing the development of efficient NMPC algorithms [24][25][26][27][28] are complementary, i.e., real-time capable control solutions will require both reduced dynamic models and efficient numerical algorithms. This discussion is, however, out of the scope of this manuscript.…”

Analysis of Capacity Potentials in Continuously Operated Chemical Processes

“…The analyzed process is the production of ethylene oxide as a representation of an important commodity chemical with a high production output. Due to the application of ultrapure oxygen within the process and the recent effort to enhance flexibility of air separation units [10][11][12], this process becomes of particular interest.…”