Optimization of Hydronic Heating System in a Commercial Building: Application of Predictive Control with Limited Data

The optimization of building equipment control is one of the key levers for improving energy efficiency. This article presents an application of predictive control to a commercial building heated by a hydronic system, comparing its performance to a traditional control strategy based on a conventional heating curve. The approach is developed and validated using TRNSYS18 modeling, which allows for comparison of the control methods under the same weather boundary conditions. The goal of the proposed strategy is to balance energy consumption and indoor thermal comfort. One challenge of the approach is to model building behavior with a limited number of commonly measured variables, without the need for additional sensors, demonstrating its practical feasibility for real-world application. The strategy aims to determine the optimal hourly control sequence of the secondary heating circuit’s water setpoint temperature, so it’s not the boiler supply water temperature that is optimized, but rather the temperature of the water that feeds the radiators. Limited data availability is a major constraint for accurately capturing the system dynamics. A black box approach, combining two Machine Learning models, is developed: an artificial neural network to predict indoor temperature and a support vector machine to predict gas consumption. After integrating weather forecasts, occupancy scenarios, and comfort requirements, a genetic algorithm determines the optimal hourly temperature setpoints to control the heating system. This work demonstrates the possibility of creating sufficiently accurate models for this type of application using limited data. It offers a simplified and efficient optimization approach to heat control in such buildings. The case study results show energy savings up to 30% compared to a traditional control method.