Simultaneous mixed-integer dynamic scheduling of processes and their energy systems

TL;DR

This paper presents a mixed-integer linear programming approach for real-time, simultaneous scheduling of production processes and energy systems, effectively handling process dynamics and binary decisions amidst volatile electricity prices.

Contribution

It introduces a novel three-part scheduling formulation combining process dynamics, energy demand, and energy system models, enabling efficient real-time optimization.

Findings

Achieves 82% and 95% of nonlinear optimization improvements in case studies.

Provides a MILP formulation with fast runtimes suitable for real-time scheduling.

Demonstrates applicability to multi-input multi-output processes.

Abstract

Increasingly volatile electricity prices make simultaneous scheduling optimization desirable for production processes and their energy systems. Simultaneous scheduling needs to account for both process dynamics and binary on/off-decisions in the energy system leading to challenging mixed-integer dynamic optimization problems. We propose an efficient scheduling formulation consisting of three parts: a linear scale-bridging model for the closed-loop process output dynamics, a data-driven model for the process energy demand, and a mixed-integer linear model for the energy system. Process dynamics are discretized by collocation yielding a mixed-integer linear programming (MILP) formulation. We apply the scheduling method to three case studies: a multi-product reactor, a single-product reactor, and a single-product distillation column, demonstrating the applicability to multi-input multi-output processes. For the first two case studies, we can compare our approach to nonlinear optimization and capture 82 % and 95 % of the improvement. The MILP formulation achieves optimization runtimes sufficiently fast for real-time scheduling.