Optimization in Chiller Loading Operation Using Genetic Algorithm

Multi-chiller systems are extensively employed in industrial and commercial buildings to effectively manage cooling loads and enhance energy efficiency. Achieving precise Part Load Ratio (PLR) configurations in these systems is crucial for optimizing energy consumption and overall system performance. Despite the utility of Genetic Algorithms (GAs) in optimizing Part Load Ratio (PLR) configurations, existing GA methodologies exhibit limitations in flexibility and computational efficiency under dynamic operational conditions. These limitations contribute to less-than-ideal energy usage and compromised system performance in multi-chiller environments. This study introduces an Improved Genetic Algorithm (I-GA) designed to overcome these constraints, promising optimized energy consumption. This study proposes the utilization of a Genetic Algorithm (GA) optimization technique to identify the optimal PLR configuration, minimizing power consumption during chiller loading operations. This research aims to investigate and discover minimal energy consumption by employing the GA approach while ensuring consistent system performance. The proposed GA optimization technique offers a systematic and intelligent approach to determine multi-chiller systems' most efficient PLR configuration. By setting important parameters, the GA algorithm can effectively explore a wide range of potential PLR combinations, identifying the optimal configuration that minimizes energy consumption. The primary objective of this study is to validate the effectiveness of the GA-based optimization technique in achieving significant energy savings compared to conventional chiller loading strategies while minimizing variability in energy consumption across different operational conditions. For cooling demands ranging from 1026RT (95%) to 756RT (70%), the GA-based optimization technique consistently outperforms the default setting, yielding reductions in total power consumption ranging from 11.47 kW (1.69%) to 40.74 kW (7.02%). The findings highlight the potential of the GA-based optimization technique in significantly reducing energy consumption with competitive savings and minimal inconsistencies in system performance, contributing valuable insights for building managers, engineers, and researchers aiming to enhance the energy efficiency of multi-chiller systems.