Operational robustness of optimized smart local energy systems: quantifying multi-criteria and multi-year performance

Abstract Optimization-based planning of smart local energy systems (SLESs), as opposed to conventional design temperature sizing, can be sensitive to interannual weather variability, often resulting in designs that are overfitted to their planning data and may thus lack robustness. This variability impacts the dimensioning and operation of technologies, including energy storage and conversion, whose optimized capacities can vary greatly depending on the data year. While this phenomenon is widely studied in the context of national-scale systems, its impact on SLESs remains largely unquantified. This paper proposes a framework for evaluating the operational performance of optimized systems within and outside their planning data periods. We optimize a Nordic SLES for single- and multi-year horizons using linear programming and assess its operational performance through rolling horizon simulations. A multi-criteria composite indicator, based on TOPSIS, integrates the performance metrics of cost, life-cycle emissions, self-sufficiency, and system reliability. The system optimized on single-year data was found to suffer from significant performance degradation when operated in other years, with one design failing to meet almost 12% of the annual heat demand in one scenario. In contrast, the system optimized using five-year data (2020–2024) maintains consistently high performance across individual years, highlighting the importance of considering multi-year conditions in SLESs. The framework helps planners to select designs that are robust over the years.