Co-channel interference in Low Earth Orbit (LEO) satellite communication systems significantly impacts Unmanned Aerial Vehicle (UAV)-based power inspection applications. With the increasing density of LEO satellite constellations, spectrum scarcity results in severe interference due to overlapping beam coverage, degrading the communication quality of UAV-based power inspection missions. We propose a Deep Deterministic Policy Gradient-based joint power control and beamforming method to optimize interference coordination. The approach models interference coordination as a Markov Decision Process, incorporating a reward function that considers Signal to Interference plus Noise Ratio (SINR) performance, service quality requirements for diverse UAV-based power inspection tasks, and energy efficiency. Simulation results demonstrate that the proposed method significantly outperforms traditional approaches, achieving substantial improvements in three key metrics: minimum SINR, service compliance rates, and interference suppression performance while maintaining computational efficiency. The results demonstrate an effective solution for intelligent interference management in LEO satellite-supported UAV-based power inspection communications.