The status monitoring and failure detection for equipment operation have always been important means to protect equipment for its safe and reliable operation. Therefore, establishing of a self-adaptive selection and decision optimizing model based on trend prediction method can self-adaptively select trend prediction method according to actual operating status so as to improve failure prediction accuracy and expand application range of failure prediction. The failure prediction experimental device was established to verify the practical application of optimal objective function in the fault prediction. The self-adaptive selection and decision optimizing method, which realizes the failure prediction for large size rotating equipments base on vibration signal, not only can adapt failure predictions of different rotating equipments, but also can realize the real-time online prediction for rotating equipment status; moreover, it has self-adaptive judgment method for multiple vibrating trend prediction models so that the optimal prediction results has high judgment success rate. Meanwhile, it provides trend prediction method adopting multiple prediction models and provides prediction results conducted by multiple prediction models. Compared with historical actual value, it has higher judgment value of failure early warning.© (2010) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.