This study developed a methodology to estimate the thermal consumption in the electric vehicle. It makes use of a database collected from 3 Hyundai Kona EVs via OBD-II device from Jeju, South Korea for the past seven months. The data consists of basic information of vehicle trip including time, battery, motor and auxiliary components, collected in 1s interval. This study first examined the Kona climate control subsystem, the operational principle of the vehicle's heating, ventilation and cooling systems (HVAC), and battery thermal management systems (BTMS). Later, a lumped-capacitance model was developed for EV thermal systems. Lastly, we adopted a physics-guided neural network (PGNN) to enhance and verify our model accuracy. The proposed method has a simple and fast computation speed that allows real-time climate control energy consumption tracking and the purposed model is able to reduce the EV power consumption.