A new approach to health condition identification of rolling bearing using hierarchical dispersion entropy and improved Laplacian score

Since bearing fault signal under complex running status is usually manifested as the characteristics of nonlinear and non-stationary, which implies it is difficult to extract accurate affluent features and achieve effective fault identification via conventional signal processing tools. In this article, a hybrid intelligent fault identification scheme, the combination of hierarchical dispersion entropy and improved Laplacian score, is proposed to address this problem, which is mainly composed of three procedures. First, the particle swarm optimization–based optimized hierarchical dispersion entropy is adopted to excavate multilevel fault symptoms from low-frequency and high-frequency components, which can both solve the shortcoming of missing of high-frequency feature information existing in the recently presented multiscale dispersion entropy and artificial parameter selection issue of hierarchical dispersion entropy. Second, an improved feature selection strategy based on improved Laplacian score is proposed to select the sensitive features to establish a low-dimensional feature data set by incorporating the weight coefficient into Laplacian score. Finally, the established low-dimensional feature data set is fed to a Softmax classifier to automatically identify different health conditions of rolling bearing. The efficacy of the proposed method is validated by two experimental cases. Results show that our approach is highly effective in recognizing different fault categories and severities of rolling bearing. Meanwhile, our approach exhibits higher accuracy and better identification performance than some similar entropy-based hybrid approaches and other identification methods reported in this article.

Laplacian score and genetic algorithm based automatic feature selection for Markov State Models in adaptive sampling based molecular dynamics

Adaptive sampling molecular dynamics based on Markov State Models use short parallel MD simulations to accelerate simulations, and are proven to identify hidden conformers. The accuracy of the predictions provided by it depends on the features extracted from the simulated data that is used to construct it. The identification of the most important features in the trajectories of the simulated system has a considerable effect on the results. Methods In this study, we use a combination of Laplacian scoring and genetic algorithms to obtain an optimized feature subset for the construction of the MSM. The approach is validated on simulations of three protein folding complexes, and two protein ligand binding complexes. Results Our experiments show that this approach produces better results when the number of samples is significantly lesser than the number of features extracted. We also observed that this method mitigates over fitting that occurs due to high dimensionality of large biosystems with shorter simulation times.