The rotary steerable drilling system (RSS) is key equipment for oil and gas exploration and development in complex environments. The utilization of wireless power transmission (WPT) in RSS facilitates the efficient transmission of energy and enhances the safety and precision of drilling activities. However, accurate identification of load parameters remains the key to improving system performance. In response to this difficulty, a generalized state-space model of the system is formulated, followed by an analysis of the noise environment present in practical applications. Secondly, the improved Particle Swarm Optimization (PSO) is used for parameter identification of noise. Finally, simulation and experimental verification were conducted. The results show that in the presence of noise, the parameter identification error is within ±5%, verifying the effectiveness of the improved algorithm.