Optimization through Automated Decision Support in District Cooling - Minimizing Electricity Use through Software-Based Start/Stop Control and Chiller Dispatch

Sammanfattning

This thesis investigates how software-based decision support can improve operationalefficiency in district cooling through automated start/stop control and chiller dispatch.The study was conducted as a design science case study at Göteborg Energi’s districtcooling network in Gothenburg, Sweden. The work addresses three main questions:identifying the main operational limitations in the current control strategy, designinga software-based artifact for improved dispatch decisions, and assessing which partsof the artifact are case-specific and which may be transferable to similar districtcooling systems. The current operation was analyzed using historical operationaldata from 2025 together with input from domain experts. This analysis showedthat the existing control strategy is only partly standardized in practice, sinceactual chiller dispatch varies under similar operating conditions and remains partlydependent on operator judgement. To address these limitations, a software-basedartifact was developed consisting of empirical chiller performance models, a MixedInteger Non-Linear Programming (MINLP) optimization engine, and an interactivedashboard for visualization and fleet management. The artifact was evaluatedthrough simulation using historical operating conditions from 2025 as a baseline. Theresults indicate that the proposed control strategy can improve operational efficiencywhile satisfying cooling demand and limiting daily start/stop events in accordancewith the defined success criteria. The largest improvements were observed duringthe summer period when the electricity use was reduced by as much as 41.69%. Thethesis thus contributes both a practical prototype and a design-oriented approachthat may be adaptable to similar district cooling systems after local recalibration ofthe chiller models.