Search

Supervised latent variable regression methods such as partial least squares (PLS) and dynamic PLS have found wide applications in data analytics, quality prediction, and fault monitoring in various industries. In this article, we tackle the unbalanced data problem of sparse quality measurement and abundant process data in process control systems to make use of all data samples for modeling. A novel semi-supervised dynamic latent variable regression (SemiDLVR) method is proposed for quality prediction and quality-relevant fault monitoring. The proposed SemiDLVR method integrates Laplacian manifold regularization with dynamic regularized latent variable regression (DrLVR) to form a semi-supervised framework to efficiently model unlabeled data. A unified objective that combines DrLVR and the Laplacian matrix is proposed and the solution is provided. Statistical monitoring indices are, thereafter, defined for quality-relevant fault monitoring in the semi-supervised framework. Results from experimental studies on a numerical simulation, an industrial sulfur recovery unit (SRU), and the Tennessee Eastman (TE) process benchmark are presented to demonstrate the effectiveness of the proposed method.

Industrial data are usually collinear, which can cause pure data-driven sparse learning to deselect physically relevant variables and select collinear surrogates. In this article, a novel two-step learning approach to retaining knowledge-informed variables (KIVs) is proposed to build inferential models. The first step is an improved knowledge-informed Lasso (KILasso) algorithm by removing penalty on the KIVs to produce a series of candidate subsets that guarantee the retention of the KIVs. The candidate subsets are then used to run the KILasso or ridge regression again to select the best sets of variables and estimate the final model. Two new algorithms are proposed and applied to datasets from an industrial boiler process and the Dow Chemical challenge problem. It is demonstrated that some important physically relevant variables are deselected by pure data-driven sparse methods, but they are retained using the proposed knowledge-informed methods with superior prediction performance.

Purpose: Five strategies were recommended by the American Association of Clinical Endocrinologists/American College of Endocrinology (AACE/ACE) guidelines for the treatment of postmenopausal osteoporosis (PMO) patients with a very high fracture risk. We aimed to assess their cost-effectiveness in the United States (US). Methods: A microsimulation Markov model was created to compare the cost-effectiveness of five treatment strategies, including zoledronate, denosumab, abaloparatide, teriparatide, and romosozumab in PMO patients with a recent fracture from the healthcare perspective of the US. The data used in the model were obtained from published studies or online resources. Base-case analysis, one-way deterministic sensitivity analysis (DSA) and probability sensitivity analysis (PSA) were conducted for 65-, 70-, 75-, and 80-year-old patients. Results: In base case, at 65 years, zoledronate was the cheapest strategy. The incremental cost-effectiveness ratios (ICER, which represent incremental costs per QALY gained) of denosumab, teriparatide, abaloparatide, and romosozumab against zoledronate were $13,020/QALY (quality-adjusted years), $477,331 /QALY, $176,287/QALY, and $98,953/QALY, respectively. Under a willing-to-pay (WTP, which means the highest price a consumer will pay for one unit of a good of service) threshold of $150,000/QALY, denosumab and romosozumab were cost-effective against zoledronate. The PSA results showed that denosumab was the most cost-effective option with WTP thresholds of $50,000/QALY, $100,000/QALY and $150,000/QALY. The results were similar in other age groups. The DSA results indicated that the most common parameters that have important influence on the outcome were drug persistence, incidence of adverse events, the efficacy of drugs on hip fractures and the cost of the drug. Conclusion and relevance: Among PMO patients with a very high fracture risk in the US, zoledronate is the cheapest strategy and denosumab is the most cost-effective choice among these five strategies.

AbstractThis work provides a summary of selected experimental capabilities being developed to support nonnuclear testing and demonstration of technology in support of microreactors under the U.S. Department of Energy’s (DOE’s) Microreactor Program. Major capabilities include the Single Primary Heat Extraction and Removal Emulator (SPHERE) and the Microreactor Agile Non-nuclear Experimental Test Bed (MAGNET). The SPHERE facility allows for controlled testing of the steady-state and transient heat rejection capabilities of a single heat pipe using electrical heaters that simulate nuclear heating. The facility is capable of monitoring axial temperature profiles along the heat pipe and surrounding test articles during startup, steady-state operation, and transients. Instrumentation includes noncontact infrared thermal imaging, surface thermocouples, spatially distributed fiber optic temperature and strain sensors, electrical power meters, and a water-cooled, gas-gap calorimeter for quantifying heat rejection from the heat pipe. The facility can be operated under both vacuum and inert-gas conditions. The MAGNET facility is a large-scale, 250-kW electrically heated microreactor test bed to enable nonnuclear experimental evaluation of a variety of microreactor concepts. It can be supplied to electrically heat a scaled section of a microreactor and further test the capabilities of heat rejection systems. The initial MAGNET experiments will support technology maturation and reduce uncertainty and risk associated with the design, operation, and deployment of monolithic heat pipe–based reactors. However, this test bed can broadly be applied to multiple microreactor concepts to evaluate a wide range of thermal-hydraulic and structural phenomena such as interface coupling with power conversion units and other collocated systems. MAGNET can evaluate integral thermomechanical effects during electrical heating of an array of heat pipes in a larger test article. Examples of initial testing will include thermal stresses in the monolith and the impact of debonding of a heat pipe from the core block and how that failure could impact surrounding heat pipes, i.e., understanding the potential for cascading failure. This work also discusses some modeling capabilities that can support experiment design, analysis, and interpretation, including the heat pipe code Sockeye and a comparison of thermal-structural simulations performed using ABAQUS and STAR-CCM+.

In this paper, a novel compact dynamic inner canonical correlation analysis (DiCCA) algorithm with autoregressive integrated moving average (ARIMA) inner models is proposed to better capture the latent dynamics of high dimensional time series. It can extract latent factors that capture the underlying dynamics of the data and model them using the ARIMA structure, which has fewer parameters and more flexibility than the potentially high-order AR structure used in the original DiCCA algorithm. The proposed algorithm can also handle nonstationary latent factors by explicitly modeling the unit roots. The algorithm integrates the extraction and the modeling of a latent factor in one step, resulting in a consistent inner and outer model. The algorithm is applied to an industrial dataset and shows better performance than the original DiCCA algorithm.

Statistical learning methods have been widely studied and practiced in the past for inferential modeling. In recent years, deep learning methods have been implemented for inferential sensor modeling. As a popular deep learning model, the long short-term memory (LSTM) network is capable of handling data nonlinearity and dynamics and is therefore applied for dynamic inferential modeling. In this paper, we analyze and compare LSTM with other statistical learning methods for the dynamic NOx emission prediction of a 660 MW industrial boiler. Support vector regression (SVR), partial least squares (PLS), and Least absolute shrinkage and selection operator (Lasso) with embedded dynamics are compared with LSTM for dynamic inferential modeling. The experimental results indicate that SVR, PLS, and Lasso outperform LSTM. By disabling the LSTM gates to realize a simple memory structure, the LSTM performance is signifcantly improved. The main goal of the paper is to demonstrate that a deep neural network that is effective in other domains requires close scrutiny and detailed study to show its superiority in process applications.

In this paper, a two-step sparse learning approach is proposed for variable selection and model parameter estimation with optimally tuned hyperparameters in each step. In Step one, a sparse learning algorithm is applied on all data to produce a sequence of candidate subsets of selected variables by varying the hyperparameter value. In Step two, for each subset of the selected variables from Step one, Lasso, ridge regression, elastic-net, or adaptive Lasso is employed to find the optimal hyperparameters with the best cross-validation error. Among all subsets, the one with the overall minimum cross-validation error is selected as globally optimal. The effectiveness of the proposed approach is demonstrated using an industrial NOx emission dataset and the Dow challenge dataset to predict product impurity.

Synergistic flame-retardant effect of aluminum diethyl phosphinate (AlPi) in intumescent flame retardant polypropylene (PP/IFR) system and the flame-retardant mechanism were investigated. The flame retardancy of PP/IFR/AlPi (the mass ratio of IFR to AlPi is 2 : 1) was the best, which was proved by the results of the limiting oxygen index (LOI) test, UL-94 test, and cone calorimeter test ( CCT) test. Here, the LOI value of the sample was as high as 34% and passed the V–0 rating in UL–94 test. The peak heat release rate (PHRR) decreased by 92.57%, the total heat release (THR) reduced by 90.52%. Thermogravimetric (TGA) data showed that the introduction of AlPi improved thermal stability and changed the thermal degradation behavior of PP/IFR composites. Interestingly, scanning electron microscopy (SEM), energy-dispersive X-ray spectroscopy (EDXS) and laser Raman spectroscopy (LRS) proved that PP/IFR/AlPi had formed more residual carbon, but the flame retardancy was worse than PP/IFR/AlPi. This is because when the mass ratio of IFR to AlPi is 2 : 1, the synergy between IFR and AlPi was significant, gas-phase flame retardant and condensed-phase flame retardant reached a balance and obtained the best flame retardant effect.

In this paper, we propose a stabilization strategy for lasso to use cross-validation (CV) for structure learning. It is known that cross-validation often prefers very small λ that selects an excessively large number of variables, which is also in a less stable region of λ. In this paper, we propose to reduce the heterogeneity of the model structures during the CV step. We first build a series of models using all data with a grid of λ. Then the models of all CV-folds use a revised lasso objective that penalizes deviations from the model structure using all data. Further, we propose a stable selection criterion that uses CV prediction errors jointly with a stability measure to select the most stable model with near minimum CV errors. The proposed strategy is demonstrated using data from an industrial boiler process to predict NOx emissions.

In this paper, we propose an adaptive method for dynamic predictive monitoring of industrial processes based on dynamic latent variable (DLV) models. DLV models extract dynamic latent variables with descending predictabilities and provide explicit modeling of the dynamics. By exploiting these two characteristics, the proposed method provides predictions conditional on fault-free data, such that potential faults lie in the prediction errors only. When a fault has been detected, the prediction horizon will increase in real time to avoid using faulty data for predictions. However, the prediction errors will grow as the prediction horizon increases. Based on the descending order of the predictability built in the DLVs, a DLV’s prediction will be adaptively turned off when its prediction error variance is greater than the variance of the DLV. The DLV’s prediction will be resumed when the fault data period is over. In general, the most predictive DLVs will survive the longest prediction horizon. In the limiting case when all DLV predictions are turned off, the monitoring scheme uses the data mean for prediction, which is equivalent to a static monitoring scheme. Case studies are provided to illustrate the effectiveness of the proposed method.

In this paper, dynamic predictive monitoring schemes based on dynamic latent variable models are proposed. We consider the most typical case in industrial data where dynamics usually exist in a reduced dimensional subspace. First, using dynamic latent variable models, predictions are made to reduce the dynamic latent space variability and focus on the unpredictable variabilities for process monitoring, leading to reduced control regions without reducing confidence levels. A general expression is developed to decompose the overall uncertainty into dynamic prediction errors and static variabilities. Second, monitoring based on multistep ahead prediction windows are used to generate fault-free predictions whenever a fault is detected. Third, we illustrate that oblique projections in dynamic latent variable models are required to separate static variabilities from dynamic latent variables. A detected fault is further classified as a static fault or a dynamic fault based on different monitoring indices. Case studies on a simulation data set and the Tennessee Eastman data set demonstrate the effectiveness of the predictive monitoring schemes.

Data science is emerging as a multidisciplinary field with tremendous recent development in theoretical foundations and expanded applications in both science and engineering. Engineering applications include industrial data analytics, autonomous systems, energy analytics, environmental applications, economic data modeling, image sequence modeling, and other high dimensional time-series data analytics. This paper is concerned with the integration of data science with dynamic systems, monitoring and control. The development of machine learning is reviewed in both a neural-mimic learning route and a learning control route, which deals with intrinsically uncertain dynamic systems. The paper then reviews the interaction of data with process manufacturing systems modeling and control, involving both data and first principles models with varying proportions. Problems include data reconciliation, state and disturbance estimation, system identification, process monitoring, and inferential property estimation. For time series data in process manufacturing systems, we present latent dynamic variable modeling methods to extract the principal dynamics in a low dimensional subspace of the data. The approaches effectively distill latent dynamic features from the data for easy interpretation, prediction, and visualization. Case studies are presented to illustrate how these latent dynamic analytics extract important features for process interpretation, troubleshooting, and monitoring.

Canonical correlation analysis (CCA) is widely used as a supervised learning method to extract correlations between process and quality datasets. When used to extract relations between current data and historical data, CCA can also be regarded as an auto-regressive modeling method to capture dynamics. Various dynamic CCA algorithms were developed in the literature. However, these algorithms do not consider strong dependence existing in adjacent samples, which may lead to unnecessarily large time lags and inaccurate estimation of current values from historical data. In this paper, a dynamic weighted CCA (DWCCA) algorithm is proposed to address this issue with a series of polynomial basis functions. DWCCA extracts dynamic relations by maximizing correlations between current data and a weighted representation of past data, and the weights rely only on a limited number of polynomial functions, which removes the negative effect caused by strongly collinear neighboring samples. After all the dynamics are exploited, static principal component analysis is then employed to further explore the cross-correlations in the dataset. The Tennessee Eastman process is utilized to demonstrate the effectiveness of the proposed DWCCA method in terms of prediction efficiency and collinearity handling.

A dynamic Gaussian process regression based partial least-squares (D-GPR-PLS) model is proposed to improve estimation ability and compared to the conventional nonlinear PLS. Considering the strong ability of GPR in nonlinear process modeling, this method is used to build a nonlinear regression between each pair of latent variables in the partial least-squares. In addition, augmented matrices are embedded into the D-GPR-PLS model to obtain better prediction accuracy in nonlinear dynamic processes. To evaluate the modeling performance of the proposed method, two simulated cases and a real industrial process based on wastewater treatment processes (WWTPs) are considered. The simulated cases use data from two high fidelity simulators: benchmark simulation model no. 1 and its long-term version. The second study uses data from a real biological wastewater treatment process. The results show the superiority of D-GPR-PLS in modeling performance for both data sets. More specifically, in terms of the prediction for effluent chemical oxygen demand of the real WWTP data, the value of the root-mean-square error is decreased by 31%, 16%, and 52%, respectively, in comparison with that for linear PLS, quadratic PLS, and least-squares support vector machine based PLS.

Many industrial pharmaceutical manufacturing processes are composed of multiple-step batch operations. However, uncontrolled variations often occur during operations that affect the cell growth performance. Because of the complexity of biological processes, one leading challenge in process operation is the identification of potential causes of undesirable process variabilities. In this paper, we propose a classification and diagnosis strategy to analyze cell culture manufacturing variability in bioprocesses with the objective of unveiling hidden factors affecting process yield and performance. The proposed strategy includes two parts: (i) a clustering method performed in the principal component residuals that effectively separates the low lactate batches into different clusters and (ii) a fault diagnosis method based on regularized LDA contribution analysis for exploring the leading contributors to each of the low performance classes. The proposed strategy is applied to industrial production data collected over 8 years from a batch process. The effectiveness of the proposed approach is demonstrated on this data set, allowing the classification and diagnosis of the sources of the low performance situations.

Air, water, and DINCH (Di-isononyl-cyclohexane-1,2-dicarboxylate) were selected as the cooling media to prepare polypropylene hollow fiber membranes by melt spinning. The effects of various cooling ways with different cooling rates on the row-nucleated lamellar crystallization were investigated. The crystallinity, orientation and arrangement of the crystals were examined by differential scanning calorimetry (DSC), wide-angle X-ray diffraction (WAXD), and small-angle X-ray scattering (SAXS). Membrane morphology was observed by SEM. The less uniform pores, a layer of amorphous phase, and only a small part of slit-shaped pores were observed on the surface of samples when water and DINCH were used as the heat transfer medium because the fast cooling rate froze the surface molecules and decreased the crystallinity and orientation. However, when air was used as the cooling medium, the higher crystallinity and the better orientation of the annealed hollow fibers were obtained mainly due to the slow cooling rate. Corresponding hollow fiber membranes prepared by stretching had better interconnectivity of pores and larger pure water flux. Thus this study explains why air was selected as the cooling medium to prepare polypropylene hollow fiber membranes which can have larger pure water flux. Also, it provides new insights into the structure control of polypropylene hollow fiber membranes by stretching in order to prepare membranes with high strength.

Polypropylene (PP)/poly(ethylene-co-vinyl alcohol) (EVOH) blends were prepared by melt blending using PP-g-MAH as a compatabilizer. The microstructure and rheological behavior of PP/EVOH blends were studied. The results showed the that EVOH particles were uniformly dispersed in the PP matrix and inhibited the mobility of PP chains. The isothermal crystallization kinetics and morphology development of PP/EVOH blends were compared with that of pure PP. It was found that the EVOH essentially acted as an effective nucleating agent for PP, which could significantly increase the nucleation rate because the EVOH molecules in PP/EVOH blends aggregated together to form interface. Therefore, the introduction of EVOH increased the overall rate of isothermal crystallization compared with pure PP, although the PP molecular chains would be hindered seriously if excessive amounts of EVOH were added. X-ray diffraction showed that pure PP and PP/EVOH blends exhibited an identical crystal structure, and the crystallinity of the PP phase in the blends was not significantly different from that of pure PP.

Concurrent monitoring schemes that achieve simultaneous process and quality-relevant monitoring have recently attracted much attention. In this Article, we formulate a supervised fault diagnosisframework based on canonical correlation analysis (CCA) with regularization, which includes quality-relevant and quality-irrelevant fault diagnosis. Monitoring indices based on regularized concurrent CCA models are introduced to perform quality-relevant, potentially quality-relevant, and quality-irrelevant monitoring. Additionally, contribution plots and generalized reconstruction-based contribution methods are developed, along with their implications for the diagnosis based on the various monitoring indices. Finally, the Tennessee Eastman process is used to illustrate the supervised monitoring and diagnosis of quality-relevant and quality-irrelevant disturbances, and the 15 known disturbances are classified into two categories based on whether they have an impact on product quality variables.

ABSTRACTBackground and aims: Population-based studies suggest a strong association between the presence of nonalcoholic fatty liver disease (NAFLD) and an increased risk of urolithiasis. However, the available information on the association of the severity of NAFLD with urolithiasis is limited. We hypothesised a link between the severity of NAFLD and the risk of urolithiasis.Methods: We recruited 1527 adult patients with NAFLD who completed a comprehensive health checkup. The severity of NAFLD was measured with AST to platelet ratio (APRI score). Logistic regression analysis was used to detect the association of APRI score with the risk of urolithiasis among NAFLD patients. ROC analysis was used to assess the diagnostic value of APRI score for identifying urolithiasis among NAFLD patients.Results: Multivariate analysis showed three independent risk factors for urolithiasis: obesity (OR 2.06 95%CI 1.35–3.13), APRI score (OR 1.29 95%CI 1.05–1.59), and serum uric acid (OR 1.07 95%CI 1.05–1.09), suggesting an independent association between the noninvasive staging of liver fibrosis and the risk of urolithiasis in NAFLD patients. A three-variable model (obesity, APRI score, and serum uric acid) with an AUROC of 0.73 (95% CI 0.70–0.75) was significant in identifying urolithiasis.Conclusions: The severity of NAFLD is associated with the risk of urolithiasis among NAFLD patients. Moreover, a three-variable model (obesity, APRI score, serum uric acid) could serve as a useful tool for identifying individuals at high risk for urolithiasis in these patients.

Two dimensional (2D) materials have been extensively used in applications such as energy storage, solar steam generation, power generation, and water purification. Among these 2D materials, Ti3C2TxMXene is one of the most popular candidates for energy storage applications because of its outstanding electrical and electrochemical properties. However, its applications are restricted due to the insufficient mechanical properties of pure MXene architectures caused by weak inter-sheet assembly. One potential supporting substrate used to impart mechanical integrity to MXene architectures is aramid yarn, however, one drawback of this lightweight and strong material is its lack of surface functionality, inhibiting MXene adhesion. To solve this issue, air plasma was utilized to impart hydroxyl groups to the fiber, enabling MXene to be deposited through a simple dip coating method. Upon 35 layers of MXene coating, the resulting composite fiber becomes a promising electrode material for supercapacitors demonstrating outstanding energy density (18.51 μWh cm−2) and low electrical resistance (0.4 kΩ cm−1) while retaining excellent tensile strength (3.28 GPa). Furthermore, the composite fiber can be also integrated into wearable Joule heating devices as it provides human body temperature heating with 7 V voltage applied.

A series of polypropylene hollow fiber membranes was fabricated by melt-spinning and stretching. The crystalline behavior and hard elasticity of precursor hollow fibers were studied by differential scanning calorimetry (DSC), elastic recovery and strain-stress curves. The structure and properties of membranes were investigated in detail by scanning electron microscopy (SEM), optical microscopy, pure water flux, and so on. The results showed that membranes with excellent structure and properties can be obtained at a spin-draw rate of 350?m/min and a stretching ratio of 200?%. The evolution of crystal structure was explored during the annealing and stretching processes by two-dimensional small-angle X-ray scattering (2D-SAXS). Shish-kebab structure was obtained during annealing for hollow fibers at a spin-draw rate of 350?m/min. The crystalline lamellae were destroyed and micropores were formed during the fabrication of membranes by stretching.

Modern large radio continuum surveys have high sensitivity and resolution, and can resolve previously undetected extended and diffuse emissions, which brings great challenges for the detection and morphological classification of extended sources. We present HeTu-v2, a deep learning-based source detector that uses the combined networks of Mask Region-based Convolutional Neural Networks (Mask R-CNN) and a Transformer block to achieve high-quality radio sources segmentation and classification. The sources are classified into 5 categories: Compact or point-like sources (CS), Fanaroff–Riley Type I (FRI), Fanaroff–Riley Type II (FRII), Head–Tail (HT), and Core-Jet (CJ) sources. HeTu-v2 has been trained and validated with the data from the Faint Images of the Radio Sky at Twenty-one centimeters (FIRST). We found that HeTu-v2 has a high accuracy with a mean average precision (AP@50:5:95) of 77.8%, which is 15.6 points and 11.3 points higher than that of HeTu-v1 and the original Mask R-CNN respectively. We produced a FIRST morphological catalog (FIRST-HeTu) using HeTu-v2, which contains 835,435 sources and achieves 98.6% of completeness and up to 98.5% of accuracy compared to the latest 2014 data release of the FIRST survey. HeTu-v2 could also be employed for other astronomical tasks like building sky models, associating radio components, and classifying radio galaxies.

Partial least squares and canonical correlation analysis are latent variable models suitable for quality-relevant monitoring based on process and quality data. Recently, concurrent monitoring schemes are proposed to achieve simultaneous process and quality monitoring. This paper defines and analyzes quality-relevant monitoring based on these popular latent structure modeling methods, and the associated quality-relevant monitoring statistics are defined. Additionally, contribution plots and reconstruction-based contribution diagnosis methods are developed for concurrent fault diagnosis. Multi-dimensional quality-relevant faults can be diagnosed in the same reconstruction framework. Finally, a detailed case study on Tennessee Eastman process is shown to illustrate the diagnosis of process and quality faults and the prognosis of quality-relevant faults.

A statistical study of the properties of molecular outflows is performed based on an up-to-date sample. 391 outflows were identified in published articles or preprints before February 28, 2003. The parameters of position, morphology, mass, energy, outflow dynamics and central source luminosity are presented for each outflow source. Outflow lobe polarity is known for all the sources, and 84% are found to be bipolar. The sources are divided into low mass and high mass groups according to either the available bolometric luminosity of the central source or the outflow mass. The pace of discovery of outflows over the past seven years has increased much more rapidly than in previous periods. Surveys for outflows are still continuing. The number of high-mass outflows detected (139) has considerably increased, showing that they are commonly associated with massive as well as low mass stars. Energetic mass ejection may be a common aspect of the formation of high mass as well as low mass stars. Outflow masses are correlated strongly with bolometric luminosity of the center sources, which was obtained for the first time. There are also correlations between the central source luminosity and the parameters of mechanical luminosity and the thrust or force necessary to drive the outflow. The results show that flow mass, momentum and energy depend on the nature of the central source. Despite their similarity, there are differences between the high mass and low mass outflows. Low mass outflows are more collimated than high mass outflows. On average, the mass of high mass sources can be more than two orders of magnitude larger than those of low mass outflows. The relation between flow mass and dynamical time appears to differ for the two types of outflows. Low mass sources make up 90% of outflows associated with HH objects while high mass outflows make up 61% of the sources associated with H2O masers. Sources with characteristics of collapse or infall comprise 12% of the entire outflow sample. The spatial distribution of the outflow sources in the Galaxy is presented and the local occurrence rate is compared with the stellar birth rate.

Schizophrenia is a widespread mental disease with a prevalence of about 1% in the world population. Continuous long-term treatment is required to maintain social functioning and prevent symptom relapse of schizophrenia patients. However, there are considerable individual differences in response to the antipsychotic drugs. There is a pressing need to identify more drug-response-related markers. But most pharmacogenomics of schizophrenia have typically focused on a few candidate genes in small sample size. In this study, 995 subjects were selected for discovering the drug-response-related markers. A total of 77 single-nucleotide polymorphisms of 25 genes have been investigated for four commonly used antipsychotic drugs in China: risperidone, clozapine, quetiapine, and chlorpromazine. Significant associations with treatment response for several genes, such as CYP2D6, CYP2C19, COMT, ABCB1, DRD3and HTR2Chave been verified in our study. Also, we found several new candidate genes (TNIK, RELN, NOTCH4and SLC6A2) and combinations (haplotype rs1544325–rs5993883–rs6269–rs4818 in COMT) that are associated with treatment response to the four drugs. Also, multivariate interactions analysis demonstrated the combination of rs6269 in COMTand rs3813929 in HTR2Cmay work as a predictor to improve the clinical antipsychotic response. So our study is of great significance to improve current knowledge on the pharmacogenomics of schizophrenia, thus promoting the implementation of personalized medicine in schizophrenia.

Let-7 miRNAs are involved in carcinogenesis and tumor progression through their roles in maintaining differentiation and normal development. However, there is little research focusing on the effects of let-7 on Wnt-activated self-renewal of breast cancer stem cells. By analyzing the expression levels of let-7 family members in clinical tissues, we found that higher expression levels of let-7b and let-7c were correlated with better clinical prognosis of patients with estrogen receptor (ER)a-positive breast tumor. Further, we found that only let-7c was inversely correlated with ERa expression, and there is corelationship between let-7c and Wnt signaling in clinical tissues. Aldehyde dehydrogenase (ALDH)1 sorting and mammosphere formation assays showed that let-7c inhibited the self-renewal of stem cells in ERa-positive breast cancer. Let-7c decreased ERa expression through directly binding to the 3’UTR (untranslated region), and let-7c inhibited the estrogen-induced activation of Wnt signaling. Depletion of ERa abolished let-7c functions in stem cell signatures, which further confirmed that let-7c inhibited estrogen-induced Wnt activity through decreasing ERa expression. Taken together, our findings identified a biochemical and functional link between let-7c with ERa/Wnt signaling in breast cancer stem cells.

Canonical correlation analysis (CCA) is a well-known data analysis technique that extracts multidimensional correlation structure between two groups of variables. Due to the advantages of CCA on quality prediction, CCA-based modeling and monitoring are discussed in this paper. To overcome the shortcoming of CCA that focuses on correlation but ignores variance information, a new concurrent CCA (CCCA) modeling method is proposed to completely decompose the input and output spaces into five subspaces, to retain the CCA efficiency in predicting the output while exploiting the variance structure for process monitoring using subsequent principal component decomposition in the input and output spaces, respectively. The corresponding monitoring statistics and control limits are then developed in these subspaces. The Tennessee Eastman process is used to demonstrate the effectiveness of CCCA-based monitoring methods.

Process monitoring based on Concurrent Projection to Latent Structures (CPLS) has been proposed recently as an efficient process monitoring tool to detect the input-relevant and output-relevant faults. In this paper, a novel multimode process monitoring approach is proposed to introduce external analysis into the CPLS monitoring framework. The process variables are divided into external variables, main variables and the output variables. The mode change is detected by monitoring the variation of the external variables through a mode-detection CPLS. The prediction of the main and quality variables by external variables is done by a PLS. After the influence of the external variables is removed from the main and quality variables, another process-monitoring CPLS is built to detect the main-variable-relevant faults and quality-variable-relevant faults. The proposed approach is illustrated with a simulation process.

In this paper, we formulate the demand response problem in smart grid as a bi-level game: a consumer-level noncooperative game and a one-leader-one-follower Stackelberg game between the provider-level and the consumer-level. We prove the existence of a Nash Equilibrium for the noncooperative game and a Stackelberg Equilibrium for the Stackelberg game, focus on the case with information sharing among all consumers, and design distributed algorithms for the supply side and demand side as well as the information platform. Numerical results are provided to illustrate the performance of the proposed algorithms and the effectiveness of information sharing for improving each consumer’s payoff.

Background/Objectives:: Quantification of the association between the intake of vegetables and fruits and the risk of lung cancer is controversial. Thus, we conducted a meta-analysis to assess the relationship between vegetables and fruits and lung cancer risk. Subjects/Methods:: Pertinent studies were identified by a search in PubMed and Web of Knowledge. Random-effects models were used to calculate summary relative risks (RR) and the corresponding 95% confidence intervals (CI). Publication bias was estimated using Begg’s test. Results:: Finally, 30 articles with 37 studies comprising of 20 075 lung cancer cases for vegetables intake with lung cancer risk and 31 articles with 38 studies comprising of 20 213 lung cancer cases for fruits intake with lung cancer risk were included in this meta-analysis. The combined results showed that there were significant associations between vegetables and fruits intake and lung cancer risk. The pooled RR were 0.74 (95% CI: 0.67, 0.82) for vegetables and 0.80 (95% CI: 0.74, 0.88) for fruits. Significant association was found in females on vegetables intake and lung cancer but not in males. The association was also stronger in females than males on fruits intake and lung cancer risk. No publication bias was detected. Conclusions:: Our analysis indicated that intake of vegetables and fruits may have a protective effect on lung cancer, and the associations were stronger in females. As the potential biases and confounders could not be ruled out completely in this meta-analysis, further studies are needed.

Process manufacturing serves as the pillar of the continuous manufacturing industry such as oil, gas, chemicals, nonferrous metals, iron, and steel, and thus is closely related to almost every aspect of human life. On the one hand, in order to meet several urgent but challenging demands of increasing profits, reducing materials consumption, enhancing safety, and protecting the environment, it is necessary to facilitate the development of process manufacturing with the usage of some novel and advanced techniques such as artificial intelligence (AI) and computation intelligence (CI). On the other hand, with the increasing scale of process manufacturing, another challenge is how to effectively deal with a huge amount of industrial big data in the process industry for environmental perception, modelling, optimization, decision-making, autonomous intelligent control, fault detection, and risk analysis. Therefore, it is of fundamental importance to deeply integrate AI, CI, and data sciences to achieve accurate control and optimal decision-making for process industries.

We exploit the concept of critical coupling to graphene based chip-integrated applications and numerically demonstrate that a chip-integrated nearly perfect graphene absorber at wavelengths around 1.55 μm can be obtained by graphene nearly critical coupling with a nanobeam cavity. The key points are reducing the radiation loss and transmission possibly, together with controlling the coupling rate of the cavity to the input waveguide to be equal to the absorption rate of the cavity caused by graphene. Simulation results show that the absorption of monolayer graphene with a total length of only a few microns is raised up to 97%. Our study may have potential applications in chip-integrated photodetectors.

AbstractA hot isostatic pressing (HIP) route has been developed by the Institute of Plasma Physics of the Chinese Academy of Sciences in collaboration with the Advanced Technology & Materials Co., Ltd. for bonding W/Cu tiles to Ni-electroplated CuCrZr heat sinks. During high-heat-flux testing, in the initial stage, Cu/Ni interfacial debonding was observed. Careful analyses indicated that interfacial oxidation during encapsulation for HIP processing using tungsten inert gas (TIG) welding was the main cause of the limited fatigue lifetime. Copper oxides formed during the TIG encapsulation do not decompose during HIP at 600°C. As a result, weak bonding and even some microcracks were generated, and unfortunately these microcracks could not be detected by current industrial ultrasonic probes. An oxidation-free encapsulation technique, suitable for batch processing, has been developed to achieve a thermal fatigue lifetime of more than 1000 cycles at a heat load of 5 MW/m2for the components.

We propose and numerically analyze a mid-infrared electrically controllable plasmonic waveguide directional coupler that is composed of two parallel identical straight dielectric loaded graphene plasmonic waveguide and S-shaped waveguide bends. By varying the Fermi energy level of the graphene sheet, the maximum power coupled from the input waveguide to the cross-waveguide and the corresponding coupling length could be effectively tuned. Under different Fermi energy level, this directional coupler could serve as an electrically controlled optical switch or a 3-dB splitter around the wavelength of 10.5 μm. Moreover, the size of the entire device is really in sub-wavelength scale making it very facilitative for high density integration.

Recently, a concurrent projection to latent structures (CPLS) for multivariate statistical process was proposed. It has been proved to be a better monitoring method than the traditional PLS. However, its fault diagnosis methods have not been developed yet. In this paper, we discuss a new fault diagnosis approach based on CPLS. Five monitoring indices used in CPLS are unified into two general forms. Based on these general forms, we define their complete decomposition contributions (CDC) and reconstruction-based contributions (RBC). The diagnosability of these two contribution methods is further analyzed. Finally, simulation case studies are presented to demonstrate the results.

It is very important to diagnose abnormal events in industrial processes. Based on normal operating data in a dynamic process, dynamic latent variable model provides a clear view of separating dynamic and static variations. Recent work by Li et al. (2014a) has shown an effective diagnosis in faulty variables with multidirectional reconstruction based contributions. Their further work took Granger causality analysis into accounts to explore the casual relations instead of only correlations. Although Granger causality is a widely used method for many applications, it needs time series to be stationary to calculate the causality index, which is not applicable for nonstationary fault processes. In this paper, a new causality analysis index based on dynamic time warping is proposed to determine the causal direction between pairs of faulty variables. The case study on the Tennessee Eastman process with a step fault shows the effectiveness of the proposed approach.

Many industrial manufacturing processes have multiple operation modes because of different strategy and varying feedstock. The traditional statistical process monitoring tools such as PCA and PLS cannot be applied since they assume that the process must have single mode operation region only. In this paper, all the factors that will affect the change of the mode are considered, a similarity factor including the similarity factor of PLS models and the mean shift of the external variables is introduced to measure the similarity of two sets of data. On basis of this similarity factor, a moving window is used and a mode identification approach for multimode process monitoring is proposed. The proposed approach is demonstrated on the benchmark Tennessee Eastman process.

Model maintenance is the most time-consuming and cost-intensive in industrial model predictive control. In this paper, a drill-down diagnosis algorithm for deficient models of industrial MPC via a model quality index (MQI) is proposed. The CVs with poor models can be detected first by MQI values with all controlled variables. Then, a leave-one-out algorithm is proposed to further diagnose which sub-models are deficient for the CVs with poor model performance. Thus, the effort and cost of model maintenance can be reduced. The application result to the Wood-Berry distillation column process indicates the effectiveness of the proposed assessment method.

Partial least squares(PLS) regression has been widely used to capture the relationship between inputs and outputs in static system modeling. Several dynamic PLS algorithms were proposed to capture the characteristic of dynamic systems. However, none of these algorithms provides an explicit description for dynamic inner model and outer model. In this paper, a dynamic inner PLS is proposed for dynamic system modelling. The proposed algorithm gives explicit dynamic inner model and makes inner model and outer model consistent at the same time. Several examples are given to show the effectiveness of the proposed algorithm.

The process faults of shaft furnace roasting processes, e.g. fire-emitting, flame-out, under-reduction, and over-reduction are undesirable for stable operation of the processes. The processes share multiple complexities such as multi-variate and strong correlations, which make it difficult to diagnose the faults using model-based or knowledge-based methods. In this paper, a data-driven fault diagnosis method for shaft furnace roasting processes is presented based on reconstruction and reconstruction-based contribution. The proposed method exploits historical faulty data to derive fault directions to identify ongoing faults with the help of additional explanation from contribution plots. A case study on a simulation system of shaft furnace roasting processes illustrates the effectiveness of the proposed method.

As continuous industrial processes often operate around a desirable region of profitability, the measurement series for most process variables act as stationary series. However, there are inevitably some observed time series which are nonstationary caused by unexpected disturbances. Some series grow slowly for a long time with the equipment aging, and others appear to wander around as if they have no fixed population mean. For these series, traditional dynamic PCA or other statistical modeling methods are not applicable because the statistical properties of variables are time variant. In this paper, nonstationarity test is adopted to distinguish nonstationary series from stationary series. After that, cointegration analysis is used to describe the stochastic common trends and equilibrium error, which can be used to construct monitoring indices. Case study on Tennessee Eastman process shows that the proposed nonstationary process monitoring can efficiently detect faults in the nonstationary dynamic process.

In this paper, a data-driven dynamic concurrent projection to latent structures (DCPLS) approach is proposed for quality-relevant fault diagnosis of dynamic processes. First, a novel DCPLS algorithm is proposed for dynamic modeling which captures the dynamic correlations between quality variables and process variables. Quality-specific variations, process-specific variations, and quality-process covariations of dynamic processes are monitored respectively. Secondly, a multi-block extension of DCPLS is designed to compute the contributions according to block partition of the lagged variables, in order to help localize faults. Finally, the application results on strip-thickness relevant fault diagnosis for a practical cold rolling continuous annealing process (CAP) demonstrate the effectiveness of the proposed methods.

AbstractThe first attempt at developing chopped W fiber-reinforced W (Wf/W) composites without an engineered interface or inter-phase employing hot isostatic pressing (HIP) has been made in order to study the feasibility of the powder metallurgy (PM) fabrication methodology. Micro-structures and flexural properties of sintered compacts have been examined by an optical microscope (OM), a scanning electron microscope (SEM) equipped with an electron back scattering diffraction (EBSD) instrument and three-point bending (3PB) tests. There are some chrysan-themum-like grains around each fiber in W matrices. Mechanical properties, namely strength and pseudo-plasticity, of the sintered compacts are far from satisfactory. Abnormal grain growth does not seem to have a preferential growing direction according to EBSD results. Possible causes for the abnormal grain growth and further mechanical property optimizations are hereby presented.

Dynamic principal component analysis (DPCA) has been widely used in the monitoring of dynamic multivariate processes. In traditional DPCA, the dynamic relationship between process variables are implicit and hard to interpret. To extract explicit latent factors that are dynamically correlated, a new dynamic latent variable model is proposed. The new structure can improve modeling of dynamic data and enhance the process monitoring performance. Fault detection indices are developed based on the proposed model. A case study is given to illustrate the effectiveness of the proposed new dynamic factor model.

Continuous annealing process line (CAPL) has complex process characteristics, such as strong correlation of large number of process variables, and the coexistence of multi- subsystems and multi-operation zones. Practitioners are concerned with process faults, such as strip-break fault and roll slippage fault, whose effects often confine in a specific zone. Considering the process characteristics and fault characteristics, a fault diagnosis method is proposed. Reconstruction-based variable contribution (RBVC) is extended to reconstruction-based block contribution (RBBC) for principal component analysis (PCA) method in order to find the faulty block, and RBVC is used to determine the faulty variable. Simulation results show the effectiveness of the proposed methods, and it is applied successfully to a practical CAPL.

Abstract: The reliable control problem is considered for a class of complex chaos systems with the symmetrical circulant characters. This special structure is applied to construct a reliable controller. The resulting closed-loop complex chaos systems satisfy the stability not only when all actuators are operational, but also when some actuators experience outages. An example is given to illustrate the effectiveness of the proposed methods. [Copyright &y& Elsevier]

Laser-based additive manufacturing (AM) of metallic materials can produce AM parts of high quality. The obtained density of metal parts manufactured by laser powder bed fusion (LPBF) is mostly greater than 99% and high mechanical strengths can be achieved; however, a long processing time is typically required to reach a high quality. Possibilities to shorten the LPBF processing time involve increasing either the scanning speed or the hatch distance, but both will result in greater porosity. In general, the porosity of AM parts can be reduced by subsequent post-densification of the materials by hot isostatic pressing (HIP). In this manner, not only can the LPBF process be accelerated but the fatigue strength of the parts can be increased by reliable reduction of porosity and the simultaneous homogenization of the microstructure. Nevertheless, argon porosity introduced by the LPBF process cannot be completely eliminated by HIP due to the insolubility of argon in metallic materials.

A cusp catastrophe model is presented and the necessary and sufficient conditions leading to landslides are discussed. The sliding surface is assumed to be planar and is a combination of two media: medium 1 is elastic-brittle or strain-hardening and medium 2 is strain-softening. The shear stress-strain constitutive model for the strain-softening medium is described by the Weibull’s distribution law. This paper is a generalization and extension of the paper by Qin et al. (2001b), in which the shear stress-strain constitutive model for medium 2 was described by a negative exponent distribution; a special case of the Weibull’s distribution law. It is found that the instability of the slope relies mainly on both the stiffness ratio of the media and the homogeneity index and that a new role of water is to enlarge the material homogeneity or brittleness and hence to reduce the stiffness ratio. A nonlinear dynamic model (also called a physical forecasting model), which is derived by considering the time-dependent behavior of the strain-softening medium, is used to study the time prediction of landslides. An algorithm of inversion on the nonlinear dynamic model is suggested for seeking the precursory abnormality and abstracting mechanical parameters from the observed series of a landslide. A case study of the Jimingsi landslide is analysed and its nonlinear dynamic model is established from the observation series of this landslide using the suggested model and the algorithm of inversion. It is found that the catastrophic characteristic index |D| shows a quick rise till reaching an extremely high peak value after the slope evolves into the tertiary creep, and subsequently approaches a zero value prior to instability, which can be regarded as an important precursory abnormality index. By taking into account the evolutionary characteristic of the slope being in the secondary creep, a simplified nonlinear dynamic model is proposed for studying the properties of bifurcation and chaos. It is shown that the emergence of chaos depends on the mechanical parameters of the sliding-surface media.