Efficient and Rapid Optimal Chiller Loading Using a Simplified Two-Stage Algorithm

<p dir="ltr">Efficient management of chiller systems in large buildings is key to energy savings in HVAC operations, with the potential to significantly reduce greenhouse gas emissions. Traditional Optimal Chiller Loading (OCL) strategies, which keep all chillers running continuously, are not always energy-efficient and can lead to higher maintenance costs. To improve this, OCL has been approached through advanced optimization techniques, allowing for chillers to be selectively switched on or off. While metaheuristic methods currently dominate research in solving OCL problems, they often suffer from complex hyperparameter tuning, slow processing times, and inherent stochasticity in their results. To address these issues, we introduce a new algorithm within the deterministic algorithms category, employing the sequential least squares quadratic programming (SLSQP) solver, known for its proficiency in handling constrained nonlinear programming. Our approach employs a two-stage strategy to effectively overcome the inherent limitations of standard optimization solvers when dealing with the discrete decision-making elements in chiller system management. This deterministic approach not only enhances the reliability of the solutions but also ensures consistent performance. After extensive testing and validation across various case studies, our algorithm has demonstrated notable efficiency. It not only computes at a fast rate but also consistently achieves optimal results rapidly, often within a split second.</p>