Multistage robust mixed-integer optimization for industrial demand response with interruptible load

Abstract

Industrial demand response is an effective strategy for power-intensive manufacturing plants to reduce operating costs and in turn also contribute to the reliable operation of the power grid. Although industrial processes are increasingly participating in various demand response activities, the financially incentivized provision of interruptible load is still not well explored. This could be due to the uncertainty that, when providing interruptible load, one does not know in advance when load reduction will be requested. We apply an adjustable robust optimization approach to address this uncertainty in the production schedule of a continuous industrial process providing interruptible load. Piecewise linear decision rules, which can allow for both continuous and discrete recourse, are used to model the dependence of production decisions on the uncertain parameters. When applied to an industrial-scale compressor train case study, the proposed model achieves significant cost savings compared with a model that does not consider integer recourse.