Owing to the importance of condition maintenance, it is urgently required to predict condition in order to avoid unexpected failure. This article presents a new comprehensive prognostic approach for condition prediction based on wavelet packet transform and least squares support vector machine (LS-SVM). Comparision with traditional LS-SVM is also done to show its advantages. Simulation and experiment have been conducted to test the method. In the experiment, vibration data that were collected from the equipment is used to predict condition.