Abstract: Ultrasonic bird repellent is an important method to solve the problem of bird damage in power transmission equipment, but the sole mode of operation that ultrasonic bird repellent was used in the field caused problems of the adaptability of birds. This paper presented an improved parameter optimization method for ultrasonic bird repellent equipment of Q-Learning transmission line, and the reinforcement learning algorithm is applied to the parameter optimization of ultrasonic bird drive equipment of transmission lines in order to solve the problem of little historical data of birds-related faults and the adaptability of birds. In view of the shortcomings of traditional reinforcement learning algorithms in device terminal applications, which have slow convergence and long time-consuming, an improved Q-Learning algorithm based on dynamic learning rate was proposed, which adaptively optimized the weights of ultrasound in different frequency bands. The experimental results showed that the iterative convergence speed of the optimal parameters of the improved Q-Learning algorithm was greatly improved, and the success rate of bird repellent equipment after optimization was 76%, which is better than the traditional reinforcement learning algorithm mode, and can better solve the adaptability problem of birds.

    Key words: improved Q-Learning; ultrasonic bird repellent; parameter optimization; adaptability