(488e) Grid-Relevant Demand Response Modeling of Chemical Processes

Authors: 
Otashu, J. - Presenter, The University of Texas at Austin
Touretzky, C., The University of Texas at Austin
Baldea, M., The University of Texas at Austin
Manufacturing industries constitute a significant consumer of electricity ranging from about thirty three percent in developed countries[1] to twelve percent in developing countries[2]. In light of the fact that the world jointly seeks ways to improve energy efficiency, mitigate climate change, and optimize the efficiency of the grid[3]; it is beneficial that manufacturing industries, especially chemical industries, contribute to meeting this need. Several chemical processes have been identified as potential candidates owing to their high electricity consumption; including cement, pulp and paper, aluminum, and air separation. These industries can make modifications to their production schedules by increasing production during off-peak periods (i.e., times when total grid load and time-of-use prices are low) and storing excess product to satisfy customer demand during peak times when production rates are lowered. This way, the chemical industry can serve as a grid-level â??storage battery.â?

While intuitive, this strategy â?? referred to as demand response (DR) â?? requires a shift in the established operating paradigm of chemical plants which is predicated on running at steady state as much as possible. In contrast, synchronizing chemical process operations with the state of the power grid calls for frequent changes in production rate and/or product grade. From an operational point of view, this requires that production scheduling decisions become not only aware of the steady state characteristics of the process (i.e. what products can be made and at what cost) but also of the process dynamics (i.e. how long it takes to quickly switch between products, and the transient levels of critical operating constraints).

In our previous work[4] we focused precisely on developing such a scheduling framework. Specifically, we proposed the use scale-bridging models (SBMs) to represent the process dynamics in the DR scheduling calculation. SBMs are a set of low-order representations of the closed-loop dynamic behavior of a process that is relevant to scheduling calculations (including, e.g., the evolution production rate and product quality following changes in production targets). We showed that the use of SBMs presents distinct advantages compared to embedding a detailed process model in scheduling calculations, namely, leading to significant reductions in computation time and providing an incentive for real-time implementation.

Increased participation of industrial players in DR initiatives is expected to have an impact on the operation of the electric grid and, as a consequence, developing representations of DR entities that can be incorporated in power system models is highly desirable. Most existing efforts in this direction focus, however, on thermal loads[5,6]. As a consequence, in this work we propose a new representation of the dynamics of DR operations for chemical processes that is relevant from the grid perspective. Starting with the premise that the aforementioned SBMs are not amenable to sharing with grid operators due to, e.g., business confidentiality concerns related to explicitly representing the performance of a process, we propose a new representation of process dynamics as a grid-level battery. We define a specific set of states (e.g., state of charge) and constraints (e.g., rate of charge) that are meaningful both from the point of view of plant operations and from the point of view of the grid. We demonstrate that such models can be expressed in terms of linear equations (with both continuous and binary variables), and are as such readily embeddable in the existing frameworks for modeling power systems. Finally, we illustrate our results with a case study.

Reference

  1. Levin, K., Cashore, B., Bernstein, S., & Auld, G. (2008). Worldwide Trends in Energy Use and Efficiency: Key Insights from IEA Indicator Analysis.
  2. Prasad V.S.N. T (2009). Nigeriaâ??s Electricity Sector- Electricity and Gas Pricing Barriers. International Association for Energy Economics Journal, 1st Quarter, Pg. 29 â?? 34.
  3. G8 Gleaneagles (2005). Gleneagles Plan of Action: Climate Change, Clean Energy and Sustainable Development. Available online http://www.mofa.go.jp/policy/economy/summit/2005/ccc_a.pdf
  4. R. C. Pattison, C. R. Touretzky, T. Johansson, I. Harjunkoski, and M. Baldea. Optimal process operations in fast-changing electricity markets: Framework for scheduling with low-order dynamic models and an air separation application. Ind. Eng. Chem. Res., 55(16):4562â??4584, 2016.
  5. L. Goransson, J. Goop, T. Unger, M. Odenberger, and F. Johnsson. Linkages between demand-side management and congestion in the European electricity transmission system. Energy, 69:860â??872,2014.
  6. A. Zerrahn and W. P. Schill. On the representation of demand-side management in power system models. Energy, 84:840â??845, 2015.