Heterogeneous Unit Clustering for Efficient Operational Flexibility Modeling

Designing future capacity mixes with adequate flexibility requires capturing operating constraints through an embedded unit commitment approximation. Despite significant recent improvements, such simulations still require significant computation times. Here we propose a method, based on clustering units, for approximate unit commitment with dramatic improvements in solution time. This method speeds computation by aggregating similar but non-identical units. This replaces large numbers of binary commitment variables with fewer integers while still capturing individual unit decisions and constraints. We demonstrate the trade-off between accuracy and run-time for different levels of aggregation. A numeric example using an ERCOT-based 205-unit system illustrates that careful aggregation introduces errors of 0.05%-0.9% across several metrics while providing several orders of magnitude faster solution times (400 ×) compared to traditional binary formulations. Further aggregation increases errors slightly ( ~ 2×) with further speedup (2000 ×). We also compare other simplifications that can provide an additional order of magnitude speed-up for some problems.