<p>The nanomaterials work together to overcome the system’s heat transfer restrictions, thermal conductivity, and surface friction when they are mixed into working fluids or adhered to the surface of system components. This improves system performance in the end. When energy saving and application sustainability are taken into account for residential and commercial AC and R applications, these factors become even more important. The existing system has nanomaterials; metal oxide nanoparticles, including copper oxide (CuO), titanium dioxide (TiO₂), and zinc oxide (ZnO), show lesser improvement in system performance. The proposed system investigates the performance improvement of automobile air conditioners by introducing nano-oxides aluminum oxide (Al₂O₃), silicon dioxide (SiO₂), and magnetite (Fe₃O₄) into a working fluid of polyalkylene glycol (PAG) lubricant and R1234yf refrigerant. The main goal is to observe these nanomaterials impact the important thermophysical characteristics of density, specific heat capacity, viscosity, and thermal conductivity. For theoretical prediction optimization, a deeper assessment method is developed employing a deep learning structure known as Improved Multi-Channel and Multi-scale Domain Supercell Thunderstorm Adversarial Neural Network (IMC-MSDSTANN). The Supercell Thunderstorm Algorithm (STA) improves the prediction power of the model. Utilizing fluids enriched by nanoparticles in conjunction with deep learning optimization technologies offers a promising path toward an intelligent, extremely effective, and environmentally friendly vehicle air conditioner. The outcomes demonstrate a 24% improvement in COP, increase in thermal conductivity, enhanced cooling efficiency, and reduced compressor power consumption in comparison with recent thermal management technologies.</p>