Applying Neuro-Fuzzy Model in Indoor Comfort Microclimate Control

This paper proposes a concept for managing climate control systems (HVAC systems) in large commercial buildings based on predictive models. The proposed control concept reduces the consumption of gas used for room heating while maintaining the acceptable range of the required temperature. The authors propose fuzzy and neuro-fuzzy models to ensure comfort temperature and humidity in the indoor environment, and thus minimize energy consumption. The management strategy is formed using predictive models. The developed management strategy is applied to the climate control system through a hardware and software complex consisting of a client connected to the system and a server that forecasts changes in indoor temperature, gas consumption and forms a management strategy. The tests that implemented the proposed concept were performed in a commercial building. The efficiency of the proposed concept is shown in comparison with the control algorithm built into the HVAC system.