Modeling and optimization of HVAC systems using a dynamic neural network

The energy consumption of a heating, ventilating and air conditioning (HVAC) system is optimized by using a data-driven approach. Predictive models with controllable and uncontrollable input and output variables utilize the concept of a dynamic neural network. The minimization of the energy consumed while maintaining indoor room temperature at an acceptable level is accomplished with a bi-objective optimization. The model is solved with three variants of the multi-objective particle swarm optimization algorithm. The optimization model and the multi-objective algorithm have been implemented in an existing HVAC system. The test results performed in the existing environment demonstrate significant improvement of the system. Compared to the traditional control strategy, the proposed model saved up to 30% of energy.

► Energy consumption of a heating, ventilation and air conditioning (HVAC) system is optimized.
► A model is built with data mining algorithms to optimize controllable set points of the HVAC system.
► Three modified multi-objective particle swarm optimization algorithms are applied to solve the model.
► The experiments demonstrate that the optimization model has saved 30% of energy.