Predicting optimal energy usage in buildings using artificial neural networks based on heuristic algorithms

This thesis investigates the networks of artificial neural networks (ANNs) that are purposed to address energy-related issues. The expansion of the energy system is investigated in the cooling and air conditioning of heating, ventilation and air conditioning (HVAC) energy usage in smart buildings. The research details how minimizing heat storage (HL) and cooling load (CL) in the building can save energy. The study uses the OBL-HGS integrated region (OBL-HGS) with the Multi-Layer Perceptron (MLP) model to increase the energy utilization. It is seen that the OBL-HGS software overcomes the limitations of the HGS software and performs exceptionally well in complex view scenarios. The research estimates the HL and CL values in buildings using the UCI Energy Efficiency dataset and shows that the OBL-HGS-MLP model achieves superior accuracy and generalization compared to the traditional HGS and conventional methods. The results show that combining OBL-HGS applications with MLP provides a new strategy demonstration for optimizing energy distribution and new perspectives on energy management and sustainability. The thesis allows to design energy-efficient smart buildings and to estimate energy load in HVAC systems. It is claimed that this flexible, regular model can help building designers to achieve desired changes in HVAC systems and emphasizes the importance of using appropriate cooling control equipment in building design for the moderate climate control constraint of global energy crises.