(359g) Computation and Analysis of Optimal Operating Strategies for Air Separation Units Using High Fidelity Dynamic Models

Manufacturing operations are frequently motivated and shaped by external environmental factors, such as raw material availability, demand level and government regulations. The traditional interpretation of such external variations as disturbances and the practice of decoupling a plant from its environment are not suitable for todayâ??s transient and competitive global market. This means that such plants need to switch from largely steady-state operations to more intentionally dynamic ones in response to the time-varying environment [1].  

Cryogenic air separation units (ASUs) based on distillation technology are an example of such manufacturing plants. Due to the high degree of heat and material integration of the process, ASUs typically have limited agility and low flexibility, which make process transitions challenging. Notwithstanding, the current market environment creates a significant incentive for dynamic operation of ASUs. Although the raw material, namely ambient air, is free and plentiful, ASUs consume large amounts of electricity, with industrial gas producers in the U.S. consuming in excess of $ 1 billion dollars of electricity per year [2].  Since electricity price deregulation, energy providers introduced dynamic electricity pricing strategies, such as use-of-time and real-time policies. These changes in the utility market may further amplify variations on the demand side, as now the ASUâ??s customers may also practice demand response operations.

Studies on the operations of ASUs have focused mainly on transition agility [3], short-term planning [4] and long-term scheduling [5]. It is demonstrated in several studies that when an ASU adjusts its production loads in response to the electricity price, improvements in the plantâ??s economic performance can be achieved. However, most of the studies focused on electricity price induced operation changes. Except for the work done by Pattison and co-workers [6], most of the planning and scheduling studies do not include accurate representations of the process dynamics, despite its importance in determining the actual feasibility of required transitions. In these studies, one of the commonly practiced strategies is to use liquid inventory to buffer the effects of electricity price variations.

In this paper, we conduct a more comprehensive analysis on optimal operation strategies for air separation systems through a multi-tiered dynamic optimization framework. Here, the response of the integrated plant is captured using collocation-based reduced order dynamic models that are shown to be highly accurate. The overall objective is to improve the plantâ??s economic performance under electricity price or demand variation. One of the operational issues assessed is the economic incentive for practicing a storage and evaporation strategy in response to time-varying electricity price/demand. The trade-off between profitability and feasibility (i.e. due to process dynamics) and the resulting selection of collection methods (i.e. collecting liquid product or liquefying over-produced gas product) and utilization mechanisms (i.e. vaporizing to meet gas demand or introducing as additional reflux) are systematically explored. Another issue investigated is the benefit of having cooperative supply chain subsystems. With information on the upcoming demand change, the air separation plant could potentially take pre-emptive action so that the demand of the customer can be satisfied, while also maximizing the profitability of the plant. All these optimization problems are solved using the commercially available software package, gPROMS 4.1.0.

References

[1]

T. Backx, O. Bosgra and W. Marquardt, "Towards intentional dynamics in supply chain conscious process operations," in Foundations of Computer Aided Process Operations Conference, 1998.

[2]

Z. Chen, M. Henson, P. Belanger and L. Megan, "Nonlinear model predictive control of high purity distillation columns for cryogenic air separation," IEEE Transactions on Control Systems Technology, vol. 18, no. 4, p. 811 â??821, 2010.

[3]

Y. Cao, S. C.L.E., M. Baldea and S. Blouin, "Optimization-based assessment of design limitations to air separation plant agility in demand response scenarios," Journal of Process Control, vol. 33, pp. 37-48, 2015.

[4]

Y. Zhu, S. Legg and L. C.D., "A multiperiod nonlinear programming approach for operation of air separation plants with variable power pricing," AIChE Journal, vol. 57, no. 9, pp. 2421-2430, 2011.

[5]

S. Mitra, J. Pinto and I. Grossmann, "Optimal multi-scale capacity planning for power-intensive continuous processes under time-sensitive electricity prices and demand uncertainty. Part I: Modeling," Computers & Chemical Engineering, vol. 65, pp. 89-101, 2014.

[6]

R. Pattison, C. Touretzky, T. Johansson, I. Harjunkoski and M. Baldea, "Optimal Process Operations in Fast-Changing Energy Markets: Framework for Scheduling with Low-Order Dynamic Models and an Air Separation Application," Industrial and Engineering Chemistry Research, vol. 55, no. 16, pp. 4562-4584, 2016.