<p>This paper presents an analysis of battery consumption in a quadcopter under different control strategies. While classical controllers such as Proportional–Integral–Derivative (PID) provide robust and stable performance, they are not inherently designed for energy optimization. On the other hand, learning-based approaches such as reinforcement learning (RL) offer adaptability but may suffer from instability. Iterative learning control (ILC) can improve performance in repetitive tasks, yet its applicability is limited in dynamic environments. Mentioned controllers are tested under wind shear and Dryden turbulence conditions. In addition to simulations, a theoretical relationship between thrust generation and power consumption is incorporated to provide an energy-based perspective. The effects of various factors such as flight speed, payload and control strategy on battery consumption and state of charge (SOC) are investigated. Motor currents and state of charge are monitored to assess energy efficiency. The results show that payload mass has the most significant impact on energy consumption. To mitigate this effect, a parachute-based payload release mechanism is also considered. Among the tested controllers, PID and ILC achieve lower energy consumption and tracking error compared to the RL-based approach. PID consumes 52 units of energy, while IL and RL consume 51.8 and 59 respectively. The findings highlight the importance of integrating theoretical energy models into controller design to improve flight efficiency and operational duration.</p>