WAVELET-BASED DIAGNOSTICS FOR HELICOPTER MAIN TRANSMISSION

Recently developed technologies are beginning to serve an important application role in the domain of flexible/scalable, next generation, condition-based maintenance tools. Development of such a new generation of reliable, intelligent, diagnostic tools will have an important impact on a wide variety of vibrating platforms concerned with issues of safety, readiness, and excessive wear/deterioration. This paper presents the application of wavelet-based techniques coupled with neural networks to the development of fault detection and identification systems. The family of wavelet basis functions are specifically engineered to allow for real-time implementation. The method is applied to the complex problem of detecting and identifying a small set of faults from vibration measurements of a helicopter main transmission.