For the purpose of addressing non-stationary, poor repeatability, abundant information in the speed-up and speed-down process of a rotor-bearing system, combining with volterra series (VS) and neighborhood rough sets (NRS), a new hybrid intelligent diagnosis method is proposed. The VS is a type of nonparametric model of a nonlinear system, it can model a wide range of nonlinear systems, and can get the volterra kernel that includes the related characteristics of the system through identification. The NRS extracts useful information based only on the data itself, and is used for redundant attributes reduction to make the selected features more objective. In this paper, speed signal and vibration peak-peak value were selected as input and output signals, identified volterra kernels were applied as fault features first, then the NRS was applied for feature selection, and finally support vector machine(SVM) was used as a classifier to recognize faults of the speed-up and speed-down process. The experiment results demonstrate the proposed model not only identifies the fault type, but also identifies the fault severity.

To efficiently find hidden clusters in datasets with complex distributed data，inspired by complementary strategies, a hybrid genetic clustering algorithm was developed, which is on the basis of the geodesic distance metric, and combined with the Fuzzy C-Means clustering (FCM) algorithm. First, instead of using Euclidean distance，the new approach employs geodesic distance based dissimilarity metric during all fitness evaluation. And then, with the help of FCM clustering, some sub-clusters with spherical distribution are partitioned effectively. Next, a genetic algorithm based clustering using geodesic distance metric, named GCGD, is adopted to cluster the clustering centers obtained from FCM clustering. Finally, the final results are acquired based on above two clustering results. Experimental results on eight benchmark datasets clustering questions show the effectiveness of the algorithm as a clustering technique. Compared with conventional GCGD, the hybrid clustering can decrease the computational time obviously, while retaining high clustering correct ratio.

In this paper, an angular contact ball bearing analysis model has been established based on Hertz contact theory. Solving nonlinear equations by improved genetic algorithm (IGA) and verified the effectiveness of the method through numerical experiment. The results proved that the adoption of IGA to solve the analysis model of angular contact ball bearing, can effectively resolve the problem of the Newton-Raphson method such as no convergence and need initial value, the calculation results is satisfied.