Health Assessment of Split Torque Transmission System Using Improved Generative Adversarial Networks
摘要
Split-torque transmission system (STTS) is a new type of helicopter main rotor transmission. It has the characteristics of light quality, fewer transmission stages, high transmission efficiency, high reliability, low noise and so on. It is of great significance to carry out health detection and assessment for helicopter STTS. In order to solve the lack of fault data and priori knowledges, this paper proposes a framework for STTS anomaly detection and health assessment based on improved generative adversarial networks (IGAN). The framework is validated with fault simulation data of a STTS.