Combined heating and cooling networks with part-load efficiency curves: Optimization based on energy hub concept

Abstract

The equipment part-load efficiency needs to be considered in the design and control of heating and cooling networks as it strongly affects the overall performance. However, optimization models such as energy hubs usually consider constant efficiencies for their energy convertors, causing potentially erroneous predictions. Even though nonlinear programming and mixed integer linear programming formulations have been introduced to optimize hubs with part-load efficiency devices, they typically focus solely on operational costs. Furthermore, the impact of different load profiles on the performance of such hub is not well documented. In this work, the lifetime cost (operation + purchase) of such system was minimized based on an energy hub in which part-load efficiency devices (natural gas boilers, electric heaters, chillers, and heat pumps) were simulated for many different combinations of thermal loads. The model was linearized to reduce computational time and improve the capability of the solver to find solutions. To determine the impact of part-load efficiencies, each device is individually examined while the efficiency of other devices remains constant. The error resulting from assuming a constant efficiency is then calculated based on a reference hub with constant efficiencies. The results indicate a maximum error on the total cost of 1.85%, 0.6%, and 0.16% by assuming constant-efficiency for the boilers, chillers and heat pumps respectively. The loads for which these maximum errors occur are then chosen to optimize the hub with all devices modeled with a part-load efficiency curve. The errors increased to 1.9%, 0.71%, and 3%, respectively.