A guided operational optimization framework for district cooling systems integrating energy performance indicator and real-coded genetic algorithm

Purpose This study proposes an AI-driven optimization framework for district cooling systems, targeting reductions in operational energy consumption and contributing to net-zero goals while supporting smart and sustainable urban development. Design/methodology/approach This study introduces a systematic framework to minimize energy consumption in district cooling systems, leveraging a novel energy performance indicator (EPI) as a global efficiency metric. Adaptive, data-driven models capture thermohydraulic interactions and hourly dynamics, while a real coded genetic algorithm (RCGA) conducts guided optimization. Robust component models are integrated into the framework, with the indicator serving as the objective function. Extensive validation, calibration, and testing ensure methodological rigor, delivering reliable, precise, and reproducible results aligned with sustainability objectives. Findings Model validation demonstrates strong predictive consistency across the components, with derived accuracy ranging from 96% to 99% and R2 values approaching 0.999. Sensitivity analyses suggest the optimization algorithm operates efficiently, converging around 990 iterations and 4.5 s of computation time. Results from test cases performed over 24-h periods at hourly intervals indicate that the optimization framework consistently delivers lower EPI values compared to baseline operations. The framework achieved reductions in overall energy consumption ranging from 16% to 30.4%, with outcomes that were both stable and dependable. Research limitations/implications The study is limited by its single-objective EPI formulation, reliance on historical data, and lack of analysis linking chilled-water temperature to indoor comfort. Scalability and computational demands pose challenges for larger systems, and real-time deployment may be constrained by latency and integration requirements. Future work should explore multi-objective models and broader operational datasets. Practical implications The framework is intended for integration with smart city infrastructure and energy management systems to enhance energy efficiency. Its adaptable architecture supports scalable optimization of urban cooling networks within the district cooling technology, aligning with environmental policy objectives and net-zero initiatives to advance sustainability across the built environment. Originality/value The proposed framework leverages EPI as a novel global metric, embedded within the AI-based algorithm to secure scalable and reproducible energy savings. Integrated across smart urban cooling networks, it supports systems to adapt in real time, improves cooling efficiency, and advances sustainability goals.