(417e) Assessing the Demand Response Potential of Power-Intensive Processes By Stochastic Scheduling Optimization

Authors: 
Germscheid, S. - Presenter, Forschungszentrum Jülich GmbH
Dahmen, M., FZ Jülich
Mitsos, A., RWTH Aachen University
Electricity for power-intensive production processes can be purchased on the day-ahead and intraday electricity market 24 hours in advance and during the day of commitment, respectively. Flexible processes can adjust their load to time-varying electricity price signals. As part of this so-called demand response, scheduling optimization can be used to determine cost-optimal process operation and electricity purchases on the two markets [1–3]. Besides possible economic advantages for the plant operator, demand response helps balancing the electricity grid [4,5]. In particular, short-term imbalances caused by volatile renewable energy can be reduced by quickly adjusting the load of large consumers in response to intraday market price movements. Intraday prices exhibit a high uncertainty, which can be explicitly taken into account in scheduling optimization by means of stochastic programming [6,7]. Stochastic scheduling optimization considering price uncertainty has proven to be economically promising [8,9], however, is associated with a high computational burden. Simkoff and Baldea [3] therefore employ a reduced-order model in their two-stage stochastic scheduling of a chlor-alkali electrolysis plant and Zhang et al. [9] and Leo et al. [10] rely on tailored solution methods such as progressive hedging or Benders decomposition. Furthermore, Zhang et al. [9] and Leo et al. [10] propose to use risk-averse scheduling optimization, i.e., minimization of both the expected cost and the conditional value-at-risk, to greatly reduce the financial risk of electricity trading while still achieving high expected electricity procurement savings. The question arises which processes are primary candidates for demand response addressing both day-ahead and intraday electricity markets. We approach this question by means of stochastic scheduling optimization of a generic process model [7] that allows to describe flexible electricity demand based on few characteristic process parameters such as utilization rate, ramping limits, or storage capacity. Following the market assumptions made by Simkoff and Baldea [3], we optimize the process operation for the next day assuming known day-ahead prices and employing scenarios as possible realizations of the uncertain intraday prices. We integrate the risk-averse formulation proposed by Zhang et al. [9], i.e., we weigh the expected electricity cost and the conditional value-at-risk evenly in the objective function. By systematically varying the generic process parameters, we study the relative impact of different process characteristics on the additional savings achieved by flexible operation and stochastic scheduling. Our analysis reveals which process characteristics are most important for reducing energy procurement costs through trading flexibility on both markets and thus can be used to shortlist concrete process candidates suitable for scheduling under price uncertainty.

References:

[1] B. Daryanian, R.E. Bohn, R.D. Tabors, Optimal Demand-Side Response to Electricity Spot Prices for Storage-Type Customers, IEEE Power Eng. Rev. 9 (1989) 36.

[2] G. Dalle Ave, I. Harjunkoski, S. Engell, A non-uniform grid approach for scheduling considering electricity load tracking and future load prediction, Comput. Chem. Eng. 129 (2019) 106506.

[3] J.M. Simkoff, M. Baldea, Stochastic scheduling and control using data-driven nonlinear dynamic models: Application to demand response operation of a chlor-alkali plant, Ind. Eng. Chem. Res. 59 (2020) 10031–10042.

[4] Q. Zhang, I.E. Grossmann, Planning and Scheduling for Industrial Demand Side Management: Advances and Challenges, in: Alternative Energy Sources and Technologies, Springer, Cham, 2016: pp. 383–414.

[5] J. Burre, D. Bongartz, L. Brée, K. Roh, A. Mitsos, Power-to-X: Between electricity storage, e-production, and demand side management, Chemie Ing. Tech. 92 (2020) 74–84.

[6] I.E. Grossmann, R.M. Apap, B.A. Calfa, P. García-Herreros, Q. Zhang, Recent advances in mathematical programming techniques for the optimization of process systems under uncertainty, Comput. Chem. Eng. 91 (2016) 3–14.

[7] Z. Chen, L. Wu, Y. Fu, Real-time price-based demand response management for residential appliances via stochastic optimization and robust optimization, IEEE Trans. Smart Grid. 3 (2012) 1822–1831.

[8] M.G. Ierapetritou, D. Wu, J. Vin, P. Sweeney, M. Chigirinskiy, Cost Minimization in an Energy-Intensive Plant Using Mathematical Programming Approaches, Ind. Eng. Chem. Res. 41 (2002) 5262–5277.

[9] Q. Zhang, J.L. Cremer, I.E. Grossmann, A. Sundaramoorthy, J.M. Pinto, Risk-based integrated production scheduling and electricity procurement for continuous power-intensive processes, Comput. Chem. Eng. 86 (2016) 90–105.

[10] E. Leo, G. Dalle Ave, I. Harjunkoski, S. Engell, Stochastic short-term integrated electricity procurement and production scheduling for a large consumer, Comput. Chem. Eng. 145 (2020) 107191.