Hydraulic–thermal dynamic model of meshed district heating network based on discrete event simulation

Online planning and optimization of district heating (DH) systems require highly iterative simulations of DH networks, where computational speed and accuracy are crucial. This study develops a versatile dynamic hydraulic–thermal model to simulate modern DH networks with complex meshed topologies and diverse operational strategies. The model employs discrete event simulation (DES) for accurate and efficient simulations with variable temporal and spatial steps. The DES approach eliminates errors from numerical diffusion by directly calculating the water temperature using inlet temperatures and travel times. The DES model can capture water temperature profiles at any observation node over time and across spatial locations at any given moment with linear interpolation between critical temperature sampling points. The accuracy of the model in temperature prediction and computational efficiency are validated using a meshed network with multiple heating plants, which comprises 186 pipes and two dependent cycles. The average node temperature error across 80 substations is ∼1 °C. An 85-day simulation of the network requires under half a second. The thermal computational cost is linearly correlated with the number of sampling points, while hydraulic calculations only contribute to a small fraction of the total computational time. These results highlight the potential of the DES model for reliable and scalable smart DH applications, such as digital twins, fault diagnosis, operational planning, and system optimization. • Developed a versatile dynamic hydraulic-thermal district heating network model. • Lagrangian approach using the Discrete Event Simulation with variable time steps. • Validated with a meshed network with a mean residual standard deviation of 1.15 °C. • Completed 85-day dynamic simulation in 0.29 s for a network with 86 substations. • DES model demonstrates the potential for smart heating in large-scale networks.