Slow Feature Analysis Algorithm Research Articles

Dynamic Health Monitoring of Aero-Engine Gas-Path System Based on SFA-GMM-BID

This paper proposes a dynamic health monitoring method for aero-engines by extracting more hidden information from the raw values of gas-path parameters based on slow feature analysis (SFA) and the Gaussian mixture model (GMM) to improve the capability of detecting gas-path faults of aero-engines. First, an SFA algorithm is used to process the raw values of gas-path parameters, extracting the effective features reflecting the slow variation of the gas-path state. Then, a GMM is established based on the slow features of the target aero-engine in a normal state to measure its health status. Moreover, an indicator based on the Bayesian inference distance (BID) is constructed to quantitatively characterize the performance degradation degree of the target aero-engine. Considering that the fixed threshold does not suit the time-varying characteristics of the gas-path state, a dynamic threshold based on the maximum information coefficient is designed for aero-engine health monitoring. The proposed method is verified using a set of actual operation data of a certain aero-engine. The results show that the proposed method can better reflect the degradation process of the aero-engine and identify aero-engine anomalies earlier than other aero-engine fault detection methods. In addition, the dynamic threshold can reduce the occurrence of false alarms. All these advantages give the proposed method high value in real-world applications.

Abnormal Operating Condition Identification of Industrial Processes Based on Deep Learning With Global-Local Slow Feature Analysis

Ensuring long-term safe and efficient operation of industrial processes relies on real-time identification of abnormal operating conditions. However, industrial processes frequently switch among diverse operating conditions and face harsh production environments. As a result, some extreme cases exist in historical abnormal samples can mask some slight anomalies, making them show similar process dynamics to normal operating conditions. To address this issue, this study proposes a global-local slow-feature-analysis-based convolutional neural network (GLSFA-CNN). A global slow feature analysis (SFA) model extracts coarse-scale slow features at a macroscopic level to distinguish anomalies with different process dynamics, while a local SFA algorithm extracts real-time and fine-scale slow features at a microscopic level to identify anomalies with similar process dynamics. By combining global and local slow features, anomalies with similar or different dynamics can be simultaneously identified. The one-dimensional convolutional neural network (1D-CNN) is then used to automatically extract deep features from the global-local slow features and identify abnormal operating conditions. The industrial experiment shows that the proposed method outperforms other traditional methods and achieves high anomaly identification accuracy for the industrial process with switching conditions and extreme cases.

Fault monitoring of batch process based on multi-stage optimization regularized neighborhood preserving embedding algorithm

Batch process is an important type of industrial production process, and the process mechanism is complex. It is difficult to accurately describe the dynamic changes of the production process of multi-stage time-varying batch process. In addition, the data of batch process contain not only global information but also local information. The traditional neighborhood preserving embedded algorithm is used to maintain the local geometric structure of data while ignoring the global information, and the extracted latent variables cannot fully characterize batch process. Therefore, we propose a multi-stage optimization regularized neighborhood preserving embedding (ORNPE) algorithm. First, the multiple process stages are separated by affinity propagation (AP) algorithm. Second, based on maintaining local information of neighborhood preserving embedding algorithm, slow feature analysis algorithm is used to extract dynamic time-varying global information. Then, cross-entropy is used to optimize the global information, and the extraction ability of the global information is improved. Finally, a monitoring index based on support vector data description is constructed to eliminate adverse effects of non-Gaussian data for monitoring performance. The effectiveness and advantages of the proposed algorithm based on monitoring strategy are illustrated by the penicillin fermentation process and a semiconductor industry process.

Fault detection and diagnosis of the air handling unit via combining the feature sparse representation based dynamic SFA and the LSTM network

In recent years, slow feature analysis (SFA) has been successfully employed to deal with the air handling unit (AHU) system’s time-varying dynamic properties. However, since the derived slow features are the linear combinations of all original variables, the conventional SFA based methods may suffer from poor model interpretability and result in inaccurate fault detection and diagnosis (FDD) performance. To enhance the dynamic AHU system FDD effectiveness, this paper presents a novel feature sparse representation based SFA algorithm through imposing the sparsity on the slow features, which is called the sparse three-way data based dynamic SFA (STBDSFA). We make two contributions to propose the STBDSFA approach. One contribution is to infuse the feature sparse representation technique into the constructed SFA based monitoring model by performing the sparse restriction on the loading vector, which can effectively eliminate the meaningless variables’ coupling and select the key variable responsible for the fault detection. Before building the STBDSFA algorithm, the other contribution is to construct a new three-way data based dynamic SFA (TBDSFA) monitoring model to handle the AHU system’s two-directional dynamic properties. In the established TBDSFA model, the auto-regressive moving average exogenous (ARMAE) model is first adopted to reveal the variables’ auto-correlation relationships. Then, the multiway data analysis is applied to figure out the batch-wise dynamic property in multiple batch runs. After that, the dynamic SFA model is further set up to sufficiently tackle the time-wise dynamic nature within a batch run. To effectively diagnose the pattern of detected fault, another innovative work is to develop a long short-term memory (LSTM) based fault identification approach to classify the sparse slow features of fault snapshot dataset, owning to the LSTM’s superiority of coping with the dynamic time-varying nature of the sparse slow features. By comparing with some traditional and closely related FDD methods, the case study on the ASHRAE Research Project RP-1312 experimental datasets are performed to verify the performance and effectiveness of the proposed FDD scheme for the AHU system. To be specific, in comparison with the sparse principal component analysis, sparse dynamic SFA, kernel locality preserving projection and sparse three-way data based SFA approaches, the developed STBDSFA based monitoring method not only reveals no fault detection delays to alert the eleven test faults, but also achieves the highest average fault detection rates, i.e., 99.31% for the T2 statistic and 99.77% for the SPE statistic. Contrasted with the support vector machine, temporal convolutional network, convolutional neural network and deep belief network based classification algorithms, the suggested STLSTM based fault identification approach obtains the highest average fault diagnosis rate, i.e., 95.51% for the eleven fault snapshot datasets.

Performance assessment method of dynamic process based on SFA-GPR

A dynamic process performance assessment method based on slow feature analysis and Gaussian process regression (SFA-GPR) is proposed to capture more complete feature information from industrial process data with dynamic characteristics, thus improving the accuracy of the assessment model. First, different binary number performance labels are used to categorize data setsin terms of performance grades, and the process feature subspace is extracted using the slow feature analysis algorithm, followed by the establishment of the Gaussian process regression offline model between the feature subspace and the performance labels. During the online assessment, in conjunction with the proposed performance assessment strategy, performance grades and the current process’s operating status are determined based on the performance label of sample data obtained by the assessment model. Finally, the method is used to assess the online performance of the debutanizer and cracking furnace in the ethylene plant. The findings indicate that the method has a higher level of assessment accuracy.

Slow Feature Extraction Algorithm Based on Visual Selection Consistency Continuity and Its Application

The conventional slow feature analysis (SFA) algorithm has no support of computational theory of vision for primates, nor does it have the ability to learn the global features with visual selection consistency continuity. And what is more, the algorithm is highly complex. Based on this, Slow Feature Extraction Algorithm Based on Visual selection consistency continuity and Its Application was proposed. Inspired by the visual selection consistency continuity theory for primates, this paper replaced the principal component analysis (PCA) method of the conventional SFA algorithm with the myTICA method, extracted the Gabor basis functions of natural images, initialized the basis function family; it used the feature basis expansion algorithm based on visual selection consistency continuity (the VSCC_FBEA algorithm) to replace the polynomial expansion method in the original SFA algorithm to generates the Gabor basis functions of features with long and short-term visual selectivity in the family of basis functions, which solved the drawbacks of the polynomial prediction algorithm; it also designed the Lipschitz consistency constraint, and proposed the Lipschitz-Orthogonal-Pruning-Method (LOPM algorithm) to optimize the basis function family into an over-complete family of basis functions. In addition, this paper used the feature expression method based on visual invariance theory (visual invariance theory -FEM) to establish the set of features of natural images with visual selection consistency continuity. Subsequently, it adopted three error evaluation methods and mySFA classification method to evaluate the proposed algorithm. According to the experimental results, the proposed algorithm showed good prediction performance with respect to the LSVRC2012 data set; compared with the SFA, GSFA, TICA, myICA and mySFA algorithms, the proposed algorithm is correct and feasible; when the classification threshold of the algorithm was set at 8.0, the recognition rate of the proposed algorithm reached 99.66%, and neither of the false recognition rate and the false rejection rate was higher than 0.33%. The proposed algorithm has good performance in prediction and classification, and also shows good anti-noise capacity under limited noise conditions.

Design and Application of a Slow Feature Algorithm Coupling Visual Selectivity and Multiple Long Short-Term Memory Networks

In the traditional slow feature analysis (SFA), the expansion of polynomial basis function lacks the support of visual computing theories for primates, and cannot learn the uniform, continuous long short-term features through selective visual mechanism. To solve the defects, this paper designs and implements a slow feature algorithm coupling visual selectivity and multiple long short-term memory networks (LSTMs). Inspired by the visual invariance theory of natural images, this paper replaces the principal component analysis (PCA) of traditional SFA algorithm with myTICA (TICA: topologically independent component analysis) to extract image invariant Gabor basis functions, and initialize the space and series of basis functions. In view of the ability of the LSTM to learn long and short-term features, four LSTM algorithms were constructed to separately predict the long and short-term visual selectivity features of Gabor basis functions from the basis function series, and combine the functions into a new basis function, thereby solving the defect of polynomial prediction algorithms. In addition, a Lipschitz consistency condition was designed, and used to develop an approximate orthogonal pruning technique, which optimizes the prediction basis functions, and constructs a hyper-complete space for the basis function. The performance of our algorithm was evaluated by three metrics and mySFA’s classification method. The experimental results show that our algorithm achieved a good prediction effect on INRIA Holidays dataset, and outshined SFA, graph-based SFA (SFA), TICA, and myTICA in accuracy and feasibility; when the threshold was 6, the recognition rate of our algorithm was 99.98%, and the false accept rate (FAR) and false reject rate (FRR) were both smaller than 0.02%, indicating the strong classification ability of our approach.

Generating feature spaces for linear algorithms with regularized sparse kernel slow feature analysis

Without non-linear basis functions many problems can not be solved by linear algorithms. This article proposes a method to automatically construct such basis functions with slow feature analysis (SFA). Non-linear optimization of this unsupervised learning method generates an orthogonal basis on the unknown latent space for a given time series. In contrast to methods like PCA, SFA is thus well suited for techniques that make direct use of the latent space. Real-world time series can be complex, and current SFA algorithms are either not powerful enough or tend to over-fit. We make use of the kernel trick in combination with sparsification to develop a kernelized SFA algorithm which provides a powerful function class for large data sets. Sparsity is achieved by a novel matching pursuit approach that can be applied to other tasks as well. For small data sets, however, the kernel SFA approach leads to over-fitting and numerical instabilities. To enforce a stable solution, we introduce regularization to the SFA objective. We hypothesize that our algorithm generates a feature space that resembles a Fourier basis in the unknown space of latent variables underlying a given real-world time series. We evaluate this hypothesis at the example of a vowel classification task in comparison to sparse kernel PCA. Our results show excellent classification accuracy and demonstrate the superiority of kernel SFA over kernel PCA in encoding latent variables.