Abstract:In order to overcome the low robustness and weak generalization in existing deep autoencoder (AE) for soft sensor modeling, a novel feature-disentangled autoencoder (FDAE) integrating residual network (Resnet) (FDAE-Resnet) is proposed in this paper. Different from the traditional deep AE that only can learn entangled features, the FDAE can obtain disentangled multi-source features including trend features, periodic features and spatial features by a new trend-periodic long short-term memory (TPLSTM) and a novel dynamic self-attention convolutional neural network (DSACNN). The trend and periodic signals decomposed from input variables are fed into the TPLSTM to learn trend and periodic features in time and frequency domain, respectively. Then, the DSACNN is utilized to capture dynamic spatial features in spatial domain by adding a new attention mechanism. Moreover, disentangled multi-source features are obtained by concatenating trend features, periodic features and spatial features together. Finally, the Resnet is utilized to build the soft sensor model by establishing the relationship between disentangled multi-sources features and outputs. To illustrate the effectiveness and superiority of the proposed method, the FDAE-Resnet is applied in the actual polypropylene process industry for melt index modeling. The experiment results show that compared with other state-of-the-art methods, the FDAE-Resnet can reduce the root mean square error by 26.2% and the mean absolute percentage error by 38.2% on average in the changed working conditions, respectively.

Supervised Neighborhood Preserving Embedding for Feature Extraction and Its Application for Soft Sensor Modeling

Nonlinear Feature Extraction for Soft Sensor Modeling Based on Weighted Probabilistic PCA

Nonlocal Structure Constrained Neighborhood Preserving Embedding Model and Its Application for Fault Detection

Neighbourhood Sensitive Preserving Embedding for Pattern Classification.

Virtual Sensors of Nonlinear Industrial Processes Based on Neighborhood Preserving Regression Model

Feature Extraction Via Multi-View Non-Negative Matrix Factorization with Local Graph Regularization.

Dynamic Features Incorporated Locally Weighted Deep Learning Model for Soft Sensor Development.

Soft-Sensor Development Through Deep Learning with Spatial and Temporal Feature Extraction of Complex Processes

Neighborhood Preserving Embedding Based on Temporal Extension and Its Application in Fault Detection

Time Neighborhood Preserving Embedding Model And Its Application For Fault Detection

Deep Nonlinear Dynamic Feature Extraction for Quality Prediction Based on Spatiotemporal Neighborhood Preserving SAE

Robust Neighborhood Embedding for Unsupervised Feature Selection.

Weighted Nonlinear Dynamic System for Deep Extraction of Nonlinear Dynamic Latent Variables and Industrial Application

An Efficient Composite Deep Latent Feature Learning Framework With Layerwise Random Mapping for Complicated Industrial Soft Sensor Modeling

Unsupervised feature extraction by low-rank and sparsity preserving embedding

Nonlinear Probabilistic Latent Variable Regression Models for Soft Sensor Application: from Shallow to Deep Structure

Probabilistic Sequential Network for Deep Learning of Complex Process Data and Soft Sensor Application

Feature Extraction Of Hyperspectral Imagery Based On Deep Nmf

Robust non-negative supervised low-rank discriminant embedding (NSLRDE) for feature extraction

Novel Feature-Disentangled Autoencoder integrating Residual Network for Industrial Soft Sensor

Gated Stacked Target-Related Autoencoder: A Novel Deep Feature Extraction and Layerwise Ensemble Method for Industrial Soft Sensor Application