Description
Spatio-Temporal Learning Using Irregular Data for Complex Dynamic Processes introduces learning, modeling, and monitoring methods for highly complex dynamic processes with irregular data. Two classes of robust modeling methods are highlighted, including low-rank characteristic of matrices and heavy-tailed characteristic of distributions. In this class, the missing data, ambient noise, and outlier problems are solved using low-rank matrix complement for monitoring model development. Secondly, the Laplace distribution is explored, which is adopted to measure the process uncertainty to develop robust monitoring models.The book not only discusses the complex models but also their real-world applications in industry.- Shows how to analyze, in great detail, the industrial operational status through spatio-temporal representation learning- Covers how to establish robust monitoring models for industrial processes with irregular data- Indicates how to adaptively update models in order to reduce frequent false alarms for dynamic processes- Explains how to take the temporal correlation into consideration to develop an adaptive monitoring model for satisfying the dynamic behaviours of industrial processes
Table of Contents
1. Background2. Low-rank characteristic and temporal correlation analytics for incipient industrial fault detection with missing data3. A robust dissimilarity distribution analytics with Laplace distribution for incipient fault detection4. Variational Bayesian Student's t-mixture model with closed-form missing value imputation for robust process monitoring of low-quality data5. Stationary subspace analysis based hierarchical model for batch process monitoring6. Recursive cointegration analytics for adaptive monitoring of nonstationary industrial processes with both static and dynamic variations7. Incremental variational Bayesian Gaussian mixture model with decremental optimization for distribution accommodation and fine-scale adaptive process monitoring8. MoniNet with concurrent analytics of temporal and spatial information for fault detection in industrial processes9. Meticulous process monitoring with multiscale convolutional feature extraction10. Summary and prospect
