As the on-site heating systems (such as heat pumps, renewable energy sources) are receiving more attention in terms of practical installation and academic research, the future of heating systems is shifting towards hybrid solutions. This doctoral research explores intelligent control strategies for integrating mainly district heating (DH) and heat pumps (HPs) in residential and commercial buildings. The study focuses on enhancing energy management through Model Predictive Control (MPC), a robust closed-loop optimization method, augmented with artificial intelligence (AI) and data-driven algorithms. While MPC requires accurate building models, AI integration enables adaptive learning from historical data, improving decision-making under uncertainty. A key innovation of this work is the multi-criteria optimization framework, which considers building occupant thermal comfort, environmental impact, and cost-efficiency. Despite growing interest in hybrid systems, the optimal integration of DH and HPs remains underexplored. This research aims to fill that gap by developing a trustworthy and intelligent control system validated using open datasets and real-world data. The outcome will support endusers and building managers in making informed energy decisions, contributing to sustainable and efficient urban energy systems.