In hydraulic power generation units, the vertical turbo-generator system plays a major role in energy conversion. During its operation, various faults such as shaft misalignment, uneven bearing clearances, and rotor–stator rub affect the operation. Therefore, vibration condition monitoring is essential. Present work focuses on the rotor dynamic analysis by considering the coupled parallel misalignment, the bearing wear, and unbalances. The numerical study is performed by finite element analysis of the rotor system using Timoshenko beam elements. Initially, the free vibration analysis is conducted and validated with 3-D solution and further, the unbalance response is obtained with nonlinear bearing forces. Further, the effects of parallel misalignment and oil film bearing wear on the dynamic response are illustrated. A set of vibration spectra correspond to the above faults at different operating speeds are obtained and a training data consisting of the spectral characteristics for each fault signature is prepared. Finally, radial basis function neural network model is used to classify the faults in the system using measured frequency response signal.