Tensor dynamic neighborhood preserving embedding algorithm for fault diagnosis of batch process

Intelligence modeling interpretable for incipient fault diagnosis of batch process represents a serious challenge. For the multivariate, nonlinear, and high-dimensionality characteristics of process data, existing fault diagnosis solutions are easily concealed by noise while neglect the low amplitude and noise interference of the incipient faults. In this paper, an intelligent incipient fault diagnosis model based on deep time series feature extraction network is proposed, which integrates denoising autoencoder (DAE), layer normalization and dropout layer added LSTM (LD-LSTM), and stacked autoencoder (SAE) to obtain the efficient features. DAE is applied to restore the data and eliminate noise interference. LD-LSTM is utilized to extract time series features, in which the layer normalization and the dropout layer are added between the layers of the LSTM to solve the problems of slow model convergence and over-fitting. SAE performs the feature extraction to obtain deep time series features and learns more efficient expressions. The softmax function is applied to achieve the incipient fault diagnosis based on the deep time series features. A case study on a fed-batch penicillin fermentation process suggests that the proposed method can obtain an accuracy rate of 98.97%, which has increased an average of 20% than the traditional shallow machine learning. The accuracy of fault recognition and comparative experiments demonstrate the effectiveness and feasibility of this model, and provide a solution for the application of intelligent fault diagnosis in batch industries.

This paper investigates fault diagnosis in batch processes and presents a comparative study of feature extraction and classification techniques applied to a specific biotechnological case study: the fermentation process model by Birol et al. (Comput Chem Eng 26:1553-1565, 2002), which is a benchmark for advanced batch processes monitoring, diagnosis and control. Fault diagnosis is achieved using four approaches on four different process scenarios based on the different levels of noise so as to evaluate their effects on the performance. Each approach combines a feature extraction method, either multi-way principal component analysis (MPCA) or multi-way independent component analysis (MICA), with a classification method, either artificial neural network (ANN) or support vector machines (SVM). The performance obtained by the different approaches is assessed and discussed for a set of simulated faults under different scenarios. One of the faults (a loss in mixing power) could not be detected due to the minimal effect of mixing on the simulated data. The remaining faults could be easily diagnosed and the subsequent discussion provides practical insight into the selection and use of the available techniques to specific applications. Irrespective of the classification algorithm, MPCA renders better results than MICA, hence the diagnosis performance proves to be more sensitive to the selection of the feature extraction technique.

Batch process data contain between‐run and within‐run sources of variation, which complicates data analysis and process monitoring. Multilevel methods, such as multilevel simultaneous component analysis, greatly enhance the interpretation of large sets of batch process data by separating the between‐run and within‐run variation. Using a multilevel approach in batch process monitoring greatly reduces the occurrence of false alarms that are caused by trivial variations and process changes that were made on purpose. This also reduces the need for model updating, which often is a large effort and causes downtime of the monitoring system. Furthermore, relevant phenomena that are masked by larger but uninteresting sources of variation when standard batch process monitoring methods are applied can often be detected if the multilevel approach is used. Chemical batch process data are used to illustrate the methods. Copyright © 2012 John Wiley & Sons, Ltd.

There has been a lot of research activity in the area of batch process analysis and monitoring for abnormal situation detection since the pioneer work of Nomikos and MacGregor []. However, some of the key ideas and the thought process that led to those first papers have been forgotten. Batch process data are dynamic data. The whole philosophy of looking at batch process data with latent variables was developed because batch process variables are both autocorrelated and cross‐correlated. Statistical process control by definition checks deviations from a nominal behavior (a target). Therefore for statistical process control of batch processes we should look at deviations of process variable trajectories from their nominal trajectories and from their nominal auto/cross‐correlations. An added advantage to modeling the deviations from the target trajectory is that a non‐linear problem is converted to a linear one that it is easy to tackle with linear latent variable methods such as principal component analysis (PCA) and partial least squares (PLS). This paper first takes a critical look at the true nature of batch process data. The general case where variables are not present during the entire duration of the batch is addressed. It is then illustrated how proper centering (by taking the deviations from the target trajectory) can retain valuable information on auto‐ and cross‐correlation of the process variables. This auto‐ and cross‐correlation is only modeled with a certain types of models. Topics such as scaling and trajectory alignment are revisited and issues arising when using the indicator variable approach are addressed. The development of control charts for multiblock, multiway PCA/PLS is discussed. Practical issues related to applications in industry are addressed. Then some of the methods that have appeared in the literature are examined as to their assumptions, their advantages and disadvantages and their range of applicability. Finally the nature of transition data (start‐ups, grade transitions) is discussed and issues related to aligning, centering and scaling such types of data are presented. Copyright © 2003 John Wiley & Sons, Ltd.

Fault Diagnosis for Batch Processes by Improved Multi-model Fisher Discriminant Analysis

[1] A recent article by Rosenlof and Reid [2008] (hereinafter referred to as RR08) examines the relationship in the tropics between long-term cooling in the stratosphere and warming of the sea surface, with special emphasis in the western Pacific. In the stratosphere, radiosonde observations seem to show that most of the cooling took place since the mid-1990s, apparently coincident to a considerable extent with an increase in the rate of warming of sea surface temperatures (SSTs). A link was proposed by RR08 whereby the increase in SSTs leads to enhanced convection, which is responsible, dynamically, for cooling in the stratosphere. [2] The stratospheric analyses of RR08 rely considerably on tropical radiosonde observations, particularly those from several island stations in the western tropical Pacific. There is extensive evidence that the long-term variation in worldwide radiosonde temperatures is affected critically by inhomogeneities introduced through changes in instruments and measurement practices [e.g., Gaffen, 1994; Gaffen et al., 2000; Lanzante et al . , 2003a, 2003b; Climate Change Science Program (CCSP), 2006], a fact which is acknowledged by RR08. However, they minimize the importance of any such inhomogeneities in the six stations located in the ‘‘warm pool’’ region which they employ [RR08, paragraph 10]: ‘‘However, for our analysis, and in particular for changes noted post-1990, these discontinuities are not sufficient to preclude the use of these stations.’’ The purpose of this comment is to demonstrate that in fact such discontinuities are present, post-1990, and critically affect the evaluation of stratospheric temperatures made by RR08. [3] In their Figures 2, 3, and 5, RR08 display temperature time series for several levels in the lower stratosphere for the four stations located in the heart of the western Pacific warm pool. They note that while most of the earlier part of the records show little long-term trend, a pronounced cooling is seen for about the last 10 years of the record ( 1995–2005). Furthermore, they note an abrupt rise in temperature in early 1999 that lasts for about 2 years. As further evidence to bolster their findings, they note [RR08, paragraph 12] the similarity of the time series from the different stations: ‘‘. . . the overall similarity of the time series in Figure 2 shows that the assumption of homogeneity over the warm pool region is reasonable.’’ Unfortunately, there is considerable evidence to dispute the veracity of these claims. [4] The apparent substantial cooling during the final 10 years of the record can be explained largely by a change in radiosonde instrument type, specifically the transition from Viz to Vaisala, hereafter referred to as the VVT. Abrupt cooling upon introduction of Vaisala radiosondes was noted by Parker et al. [1997] for stations in Australia and New Zealand in the mid to late 1980s, and for U.S. controlled tropical Pacific stations beginning in 1995 [Stendel et al., 2000]. More generally, there has been a systematic worldwide change to Vaisala radiosondes since the 1980s with an expected artificial cooling, especially in the stratosphere [Lanzante et al., 2003a; Christy et al., 2003; CCSP, 2006] (see also Integrated Global Radiosonde Archive (IGRA) station history metadata available at http:// www1.ncdc.noaa.gov/pub/data/igra/igra-metadata.txt). [5] The IGRA metadata indicate that the VVT occurs in November or December 1995 for five of the six stations shown in Figure 1 of RR08 (Lihue, Truk, Koror, Yap, and Pago Pago). The literature cited above suggests that the major part of the drop in stratospheric temperature after 1995 (as seen quite clearly in Figure 5 of RR08) is almost certainly artificial. To further illustrate this point, Figure 1a shows several time series of stratospheric temperatures at Koror. The thick black curve is based on the IGRA data [Durre et al., 2006], which were used by RR08. Note that the IGRA data, although of very high quality (i.e., they have been carefully checked for consistency in time and in the vertical, random errors and outliers, as well as duplicate soundings, etc.), have not been adjusted to account for artificial inhomogeneities. A number of teams have produced adjusted radiosonde temperature time series, each using a different homogenization method. Also shown in Figure 1a are corresponding adjusted time series (HadAT2) [Thorne et al., 2005], RATPAC [Free et al., 2005], IUK [Sherwood et al., 2008], RAOBCORE 1.4 [Haimberger et al., 2008], and RICH [Haimberger et al., 2008] produced by these teams. In comparing the IGRA with the various adjusted time series in Figure 1a it is seen that the adjusted versions all show much less long-term cooling, as indicated by the accompanying trend lines. A closer look at Figure 1a reveals that most of the difference is attributable to the transition across 1995. The adjusted time series shows a JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 114, D12104, doi:10.1029/2008JD010542, 2009 Click Here for Full Article

Batch process data can be arranged in a three-way matrix (batch × variable × time). This paper provides a critical discussion of various aspects of the treatment of these multiway data. First, several methods that have been proposed for decomposing three-way data matrices are discussed in the context of batch process data analysis and monitoring. These methods are multiway principal component analysis (MPCA)—also called Tucker1—parallel factor analysis (PARAFAC) and Tucker3. Secondly, different ways of unfolding, mean centering and scaling the three-way matrix are compared and discussed with respect to their effects on the analysis of batch data. Finally, the role of the time variable in batch process data is considered and methods suggested to predict the per cent completion of batch runs with unequal duration are discussed. Copyright © 1999 John Wiley & Sons, Ltd.

Batch process monitoring is of great importance to ensure the stable operation during the process running. However, traditional deep learning methods have certain limitations when dealing with complex data structures and dynamic features that are prominent in industrial batch processes. This paper proposes a GraphSAGE-LSTM-based deep canonical correlation analysis (GSA-LSTM-CCA) to enhance the performance of feature learning for batch process monitoring. First, the k-nearest neighbor method is used to construct a graph for the variable-unfolding batch process data as a supplementary structural information for the original spatiotemporal correlations. Then, GraphSAGE and LSTM are used to extract the structural features and time-varying features of data, respectively. Finally, the learned features are weighted and fused to construct a residual generator, and CCA is utilized to carry out process monitoring. The effectiveness and superiority of the proposed method are verified in an actual injection molding process.

The evaporation duct is a particular type of atmospheric structure that always appears on the open ocean. Predicting the evaporation duct height (EDH) accurately and in a timely manner is of great significance for the practical application of marine wireless communication equipment. Understanding the characteristics of EDH time series is an essential prerequisite for establishing an appropriate prediction model. Moreover, the sampling timescales of EDH data may influence the dynamic characteristics of the EDH time series as well. In this study, EDH time series datasets at three timescales, hourly, daily, and monthly, were constructed as the case study. Statistical methods, namely the augmented Dickey–Fuller test and Ljung–Box test, were adopted to verify the stationary and white noise characteristics of the EDH time series. Then, rescaled range analysis was applied to calculate the Hurst exponent to study the fractal characteristics of the EDH time series. An extensive analysis and discussion of the chaotic dynamics of the EDH time series are provided. From the perspective of nonlinear dynamics, the phase space was constructed from the time delay τ and embedding dimension m, which were calculated from the mutual information method and the Grassberger–Procaccia algorithm, respectively. The maximum Lyapunov exponent was also calculated by the small data volume method to explore the existence of chaos in the EDH time series. According to our analysis, the EDH time series are stationary and have a non-white noise characteristic. The Hurst exponents for all three timescales were greater than 0.5, indicating the predictability of the EDH time series. The phase space diagrams exhibited strange attractors in a well-defined region for all the timescales, suggesting that the evolution of the EDH time series can possibly be explained by deterministic chaos. All of the maximum Lyapunov exponents were positive, confirming the chaos in the EDH time series. Further, stronger chaotic characteristics were found for the finer-resolution time series than the coarser-resolution time series. This study provides a new perspective for scholars to understand the fluctuation principles of the evaporation duct at different timescales. The findings from this study also lay a theoretical and scientific foundation for the future application of chaotic prediction methods in the research on the evaporation duct.

Understanding the dynamics of air traffic flow is important to achieving advanced air traffic management. This work explores the dynamic evolution and fluctuation characteristics of multistate air traffic time series from a complex network perspective, which is essential for understanding the nature of an air traffic system. With the application of the fundamental diagram (FD), we discover that the relative velocity, flight distance and trajectory similarity are the three key variables for interpreting the arrival traffic flow states of the Xiamen Gaoqi International Airport. According to these three variables, time series are classified into four traffic states based on the K-means algorithm: free flow (FF), transitional flow (TF), slightly congested flow (SCF) and heavily congested flow (HCF). The extracted time series in different states are converted into complex networks using the visibility graph method. We analyze and compare the statistical features of the networks in the four states in terms of indexes, such as the degree distribution and network structure. The results indicate that the complex network characteristics can be used to distinguish air traffic states from the original traffic flow. Our work may be helpful for scholars and engineers to better understand the intrinsic nature of air traffic and for the development of intelligent assistant decision-making systems for air traffic management.

Singular spectrum analysis and Fisher–Shannon analysis of spring flow time series: An application to Anjar Spring, Lebanon

As tensors provide a natural representation for the higher-order relations, tensor factorization techniques such as Tucker decomposition and CANDECOMP/PARAFAC decomposition have been applied to many fields. Tucker decomposition has strong capacity of expression, but the time complexity is unpractical for the large-scale real problems. On the other hand, CANDECOMP/PARAFAC decomposition is linear in the feature dimensionality, but the assumption is so strong that it abandons some important information. Besides, both of TD and CP decompose a tensor into several factor matrices. However, the factor matrices are not natural for the representation of the higher-order relations. To overcome these problems, we propose a near linear tensor factorization approach, which decompose a tensor into factor tensors in order to model the higher-order relations, without loss of important information. In addition, to reduce the time complexity and the number of the parameters, we decompose each slice of the factor tensors into two smaller matrices. We conduct experiments on both synthetic datasets and real datasets. The experimental results on the synthetic datasets validate that our model has strong capacity of expression. The results on the real datasets show that our approach outperforms the state-of-the-art tensor factorization methods.

Abstract. A long AE index time series is used as a crucial magnetospheric quantity in order to study the underlying dynainics. For this purpose we utilize methods of nonlinear and chaotic analysis of time series. Two basic components of this analysis are the reconstruction of the experimental tiine series state space trajectory of the underlying process and the statistical testing of an null hypothesis. The null hypothesis against which the experimental time series are tested is that the observed AE index signal is generated by a linear stochastic signal possibly perturbed by a static nonlinear distortion. As dis ' ' ating statistics we use geometrical characteristics of the reconstructed state space (Part I, which is the work of this paper) and dynamical characteristics (Part II, which is the work a separate paper), and "nonlinear" surrogate data, generated by two different techniques which can mimic the original (AE index) signal. lie null hypothesis is tested for geometrical characteristics which are the dimension of the reconstructed trajectory and some new geometrical parameters introduced in this work for the efficient discrimination between the nonlinear stochastic surrogate data and the AE index. Finally, the estimated geometric characteristics of the magnetospheric AE index present new evidence about the nonlinear and low dimensional character of the underlying magnetospheric dynamics for the AE index.

The sintering temperature (ST) is a critical index for condition monitoring and process control of coal-fired equipment and is widely used in the production of cement, aluminum, electricity, steel, and chemicals. The accurate prediction of the ST is important for control systems to anticipate tragedies. In this article, we propose a deep learning model for forecasting the ST using automatic spatiotemporal feature extraction from multivariate thermal time series. A hybrid deep neural network named deep convolutional neural network and gated recurrent unit network (DCGNet) is designed to extract multivariate coupling and nonlinear dynamic characteristics for forecasting the ST. DCGNet uses convolutional neural networks and gated recurrent unit (GRU) to extract the local spatial-temporal dependence patterns among the multivariates, and another parallel GRU using the historical ST data as input is incorporated to more accurately capture the dynamic characteristics of ST time series. Based on the real-world data, application results show that the proposed approach has high forecasting accuracy and robustness, thus having broad application prospects in industrial processes.

A growing number of web users around the world have started to post their opinions on social media platforms and offer them for share. Building a highly scalable evolution prediction model by means of evolution trend volatility plays a significant role in the operations of enterprise marketing, public opinion supervision, personalized recommendation, and so forth. However, the historical patterns cannot cover the systematical time-series dynamic and volatility features in the prediction problems of a social network. This paper aims to investigate the popularity prediction problem from a time-series perspective utilizing dynamic linear models. First, the stationary and nonstationary time series of Weibo hot events are detected and transformed into time-dependent variables. Second, a systematic general popularity prediction model N- SEP 2 M is proposed to recognize and predict the nonstationary event propagation of a hot event on the Weibo social network. Third, the explanatory compensation variable social intensity (SI) is introduced to optimize the model N- SEP 2 M. Experiments on three Weibo hot events with different subject classifications show that our prediction approach is effective for the propagation of hot events with burst traffic.