(119b) Industrial Demand Side Management Under Uncertainty

With deregulated electricity markets and increasing penetration of intermittent renewable energy into the electricity supply mix, the level of uncertainty in the power grid has increased tremendously. This has led to highly volatile electricity prices, which pose immense challenges to the power-intensive industries, such as air separation, aluminum, and chlor-alkali manufacturing. Demand Side Management (DSM), which refers to electric energy management on the consumers' side, has the potential of both significantly reducing the electricity cost as well as improving the efficiency and reliability of the power grid for the consumer.

The high potential impact of large industrial electricity consumers participating in DSM is widely acknowledged (Samad & Kiliccote, 2012; Merkert, et al., 2014) and has been the focus of increased research efforts in recent years. Undoubtedly, one of the major challenges in DSM is uncertainty, which does not only appear in electricity prices, but also in power availability, product demand, and the occurrence of demand response events etc. Robust and economically sound decisions can only be made when these uncertainties are appropriately taken into account in the decision-making process.

In this work, we develop optimization tools for two industrial DSM applications in which modeling uncertainty is crucial. In the first application, we consider the integrated optimization of production scheduling and electricity procurement, in which we determine how much electricity to purchase from signed bilateral contracts and how much from the spot market. Here, we apply stochastic programming (Birge & Louveaux, 2011) to account for uncertainty in spot electricity price and product demand. In the second application, we consider an air separation plant that utilizes cryogenic energy storage to store electricity and recover it later for internal use or to sell it back to the grid. By appropriately modeling the uncertainty in operating reserve demand using robust optimization techniques (Ben-Tal, et al., 2009), we can also use the energy storage capability to sell reserve capacity, which can be requested by the grid operator in times of emergency. In both cases, we apply the proposed approaches to real-world industrial case studies and demonstrate the value of accounting for uncertainty.

References

Ben-Tal, A., El Ghaoui, L. & Nemirovski, A., 2009. Robust Optimization. New Jersey: Princeton University Press.

Birge, J. R. & Louveaux, F., 2011. Introduction to Stochastic Programming. 2nd ed. New York: Springer Science+Business Media.

Merkert, L. et al., 2014. Scheduling and energy - Industrial challenges and opportunities. Computers & Chemical Engineering, Volume 72, pp. 183-198.

Samad, T. & Kiliccote, S., 2012. Smart grid technologies and applications for the industrial sector. Computers & Chemical Engineering, Volume 47, pp. 76-84.