Real-world Industrial Processes Research Articles

Investigation of a Process to Eco-Design Led Lighting Products to Enhance the Adoption of Eco-Design Methods and Tools by Industry

To date, many studies have been carried out to develop new approaches and methods to eco-design products. However, these have not been implemented and adopted by industry as much as they should. A better understanding of real-world industrial eco-design and development processes, and the eco-design tools applied during these, could inform the development of more effective and applicable eco-design methods and tools, for generic as well as for specific product categories (e.g., LED lighting products). This paper addresses this issue by describing and examining a real-world process followed to design and develop a LED lighting product by a lighting manufacturer, via case study research. The case study involved direct participatory observation to gather the data and provided new insights about the stages of the design and development process, as well as the tools applied, which were examined and discussed to inform the improvement of existing methods and tools, or the development of better new methods and tools.

Process monitoring of batch process based on overcomplete broad learning network

Process monitoring is a common strategy for monitoring the industrial production operation state. It can detect abnormal conditions in industrial processes and provide certain guidance for production. Many classical data-driven process monitoring approaches neglect the non-Gaussian and nonlinearity of the data. For solving the above problems, this paper designs an overcomplete broad learning system (OBLS) with incremental learning ability. The method combines multiple fault data into one data matrix. Then the overcomplete approach is employed to capture the non-Gaussian information from the original data to obtain a mixed matrix. Next, the weights of the OBLS network are trained according to the extracted feature matrix containing non-Gaussian information and its corresponding fault label. Meanwhile, the nonlinearity of the data is addressed by the OBLS network. Finally, the incremental learning capabilities of the OBLS network enable it to be updated quickly when new fault samples are added to the training set without entire retraining process. The experimental results in numerical examples, penicillin fermentation simulation platform and real-world industrial process demonstrate the superiority and feasibility of the OBLS model.

Input selection methods for data-driven Soft sensors design: Application to an industrial process

Soft Sensors (SSs) are inferential models which are widely used in industry. They are generally built through data-driven approaches that exploit industry historical databases. Selection of input variables is one of the most critical issues in SSs design. This paper aims at highlighting difficulties arising from the implementation of data-driven input selection methods when solving real-world case studies. A procedure is, therefore, proposed for input selection, based on both data-driven and expert-driven input selection methods. The procedure allows designing SSs with good prediction accuracy and a low number of inputs.The design of an SS for a real-world industrial process is used. The results reported show that the selection methods proposed in literature do not give consistent results when applied to the considered case study. The key role for plant expert knowledge emerges, outlining the opportunity of judicious use of automatic data-driven procedures.

Autoencoder-based nonlinear Bayesian locally weighted regression for soft sensor development

The framework of locally weighted learning (LWL) has established itself as a popular tool for developing nonlinear soft sensors in process industries. For LWL-based soft sensors, the key factor for achieving high performance is to construct accurate localized models. To this end, in this paper a nonlinear local model training algorithm called nonlinear Bayesian weighted regression (NBWR) is proposed. In the NBWR, the nonlinear features of process data are first extracted by the autoencoder; then, given a query sample a local dataset is selected on the feature space and a fully Bayesian regression model with differentiated sample weights is developed. The benefits of this approach, which include better consistency of correlation, stronger abilities to deal with process nonlinearities and uncertainties, overfitting and numerical issues, lead to superior performance. The NBWR is used for developing a soft sensor under the LWL framework, and a real-world industrial process is used to evaluate the performance of the NBWR-based soft sensor. The experimental results demonstrate that the proposed method outperforms several benchmarking soft sensing approaches.

Bayesian Nonlinear Gaussian Mixture Regression and its Application to Virtual Sensing for Multimode Industrial Processes

Virtual sensors have established themselves as effective tools in process industries for online estimating variables that are crucial but difficult to measure. However, multimode industrial processes render developing high-accuracy virtual sensors quite challenging. The main difficulties lie in that, in multimode processes, the distributions of process data are strongly non-Gaussian and the mathematical relationships between the explanatory and primary variables are highly nonlinear. Even within one operating mode, the primary variables could depend on the explanatory variables in nonlinear ways. In order to address these issues, this article proposes a virtual sensing approach named Bayesian nonlinear Gaussian mixture regression (BNGMR) with the aid of single-hidden layer feedforward neural networks (SLFNs). In the BNGMR, a fully Bayesian model structure that absorbs the merits of SLFNs and the mixture models is designed. In addition, we develop a training algorithm for the BNGMR to realize predictive virtual sensor development based on variational inference. Extensive assessments of the performance of the BNGMR are carried out using both artificial example and real-world industrial processes. The experiments have demonstrated the predictive advantage of the BNGMR over several benchmark methods and also have provided practitioners with good illustrations. Note to Practitioners —Although the case studies on two real-world industrial processes have verified the effectiveness and feasibility of the Bayesian nonlinear Gaussian mixture regression (BNGMR) for industrial applications, it is still worth emphasizing that before training the BNGMR, some important works on data preprocessing should be done, which include: 1) careful selection of explanatory variables that are closely related to primary variables; 2) time alignment matching between primary and explanatory variables; and 3) dealing with samples with missing values (for example, removing them) and elimination of outliers. We suggest to complete 1) and 2) based on process mechanisms and to pay particular attention to removing the outlier, as the presence of outliers could skew the estimations of Gaussian components and result in significant performance deterioration.

Bayesian Just-in-Time Learning and Its Application to Industrial Soft Sensing

Just-in-time learning (JITL), which can deal with both process nonlinearities and time-varying characteristics, has become a widely used tool for industrial soft sensing. High performance of JITL lies in selecting an accurate relevant sample set and developing a good base learner, which, however, still have some issues unresolved. In this article, a Bayesian JITL (BJTIL) is established to improve the performance of a JITL-based soft sensor in terms of relevant sample selection and base learner construction. The BJITL has dual implications. First, a semi-supervised relevant sample selection strategy with a mixture of Mahalanobis distances based on the fully Bayesian Dirichlet process mixture model is proposed, such that both labeled and unlabeled samples can be exploited, and complicated non-Gaussian distributions (such as those with multi-peaks or severe asymmetry) can be accounted for. Second, a weighted fully Bayesian Gaussian regression model with randomized mean and covariance is proposed as the base learner training algorithm so as to deal with the overfitting and numerical issues. Two real-world industrial processes are employed to evaluate the performance of the BJITL when it is applied to soft sensor development. The results demonstrate that the BJITL can achieve higher predictive accuracy in contrast with some state-of-the-art schemes for relevant sample selection and base learner construction. In addition, it is shown that the BJTIL can provide better predictive uncertainties.

Semisupervised Bayesian Gaussian Mixture Models for Non-Gaussian Soft Sensor.

Soft sensors have been widely accepted for online estimating key quality-related variables in industrial processes. The Gaussian mixture models (GMM) is one of the most popular soft sensing methods for the non-Gaussian industrial processes. However, in industrial applications, the quantity of samples with known labels is usually quite limited because of the technical limitations or economical reasons. Traditional GMM-based soft sensor models solely depending on labeled samples may easily suffer from singular covariances, overfitting, and difficulties in model selection, which results in the performance deterioration. To tackle these issues, we propose a semisupervised Bayesian GMM (S2BGMM). In the S2BGMM, we first propose a semisupervised fully Bayesian model, which enables learning from both the labeled and unlabeled datasets for remedying the deficiency of infrequent labeled samples. Subsequently, a general framework of weighted variational inference is developed to train the S2BGMM, such that the rate of learning from unlabeled samples can be controlled by penalizing the unlabeled dataset. Case studies are carried out to evaluate the performance of the S2BGMM through a numerical example and two real-world industrial processes, which demonstrate the effectiveness and reliability of the proposed approach.

Distributed Dynamic Process Monitoring Based on Minimal Redundancy Maximal Relevance Variable Selection and Bayesian Inference

The conventional distributed modeling strategy generally includes all the process variables in large-scale process monitoring, thus submerging the local fault information. Meanwhile, fault diagnosis issues in the aforementioned process are also worth studying. To make up the deficiencies of the general distributed method, this brief presents an advantageously distributed fault monitoring and diagnosis scheme for large-scale dynamic processes. First, the optimal variable subblock is established by the minimal redundancy maximal relevance (mRMR) algorithm for each measured variable individually. Second, the corresponding monitoring model is set up in every subblock, respectively. Finally, all the distributed multiblocks’ results are combined to form an integrated decision through the Bayesian inference. The proposed scheme considers not only the interpretation of the correlation of variables but also the redundancy between them in the block division, which better characterizes the dynamic relationships among variables, thereby facilitating the monitoring performance significantly. After that, a new contribution plot method based on subblock analysis is applied for fault isolation. Both a real-world industrial process and the Tennessee Eastman benchmark process validate that the novel monitoring strategy is more efficient than the existing state-of-the-art approaches.

Nonlinear industrial soft sensor development based on semi-supervised probabilistic mixture of extreme learning machines

Soft sensors play an important role in process industries for monitoring and control of key quality variables, and calibration of analyzers. Owing to the merits of fast learning speed and good generalization performance, extreme learning machines (ELMs) have been widely accepted to develop soft sensor models for nonlinear industrial processes. However, there still exist some challenges in developing high-accuracy ELM-based soft sensors. Specifically, ELMs with shallow networks seem to have inadequate representation capabilities for complex nonlinearities, while ELMs with deep networks have difficulties in determining the number of hidden layers and hidden nodes for each layer which readily results in overfitting. In addition, in soft sensor applications, labeled samples are usually limited due to technical or economical reasons, which adds obstacles to model training. To deal with these issues, we propose a semi-supervised probabilistic mixture of ELMs (referred to as the ‘S2PMELMs’). In the S2PMELMs, localized ELMs are trained and combined, which are completed in a unified probabilistic way such that process nonlinearities and uncertainties can be accommodated. Moreover, based on the variational Bayes expectation–maximization algorithm, we develop a training algorithm for the S2PMELMs, where unlabeled samples are able to be exploited and the regularization parameter for each ELM can be adaptively determined. The performance of the S2PMELMs is evaluated through two real-world industrial processes, and the results demonstrate the advantages of the proposed method in contrast with several state-of-the-art relevant soft sensing approaches.

Parallel Computing and SGD-Based DPMM For Soft Sensor Development With Large-Scale Semisupervised Data

Soft sensors based on Gaussian mixture models (GMM) have been widely used in industrial process systems for modeling the nonlinearity, non-Gaussianity, and uncertainties. However, there are still some challenging issues in developing high-accuracy GMM-based soft sensors. First, labeled samples are usually scarce due to technical or economical limitations, causing traditional supervised GMM-based soft sensing methods fail to provide satisfactory performance. Second, tremendous amounts of unlabeled samples are gathered, nevertheless, how to fully exploit those unlabeled samples in terms of improving both the predictive accuracy and computational efficiency remains unresolved. In this paper, in order to deal with these issues, two computationally efficient soft sensing methods, namely the parallel computing-based semisupervised Dirichlet process mixture models (P–S $^2$ DPMM) and stochastic gradient descent-based S $^2$ DPMM (SGD–S $^2$ DPMM), are proposed. The S $^2$ DPMM is first developed to mine information contained in both labeled and unlabeled samples for predictive accuracy enhancement, and subsequently is extended to the P–S $^2$ DPMM and SGD–S $^2$ DPMM to handle large-scale process data with sufficient and limited computing resources, respectively. Two case studies are carried out on real-world industrial processes, and the results obtained demonstrate the effectiveness of the proposed methods.

Transition metal cation-exchanged SSZ-13 zeolites for CO2 capture and separation from N2

CO2 capture from post-combustion flue gas mixture (CO2/N2:15/85) is challenging and requires adsorbents with high capacity and high selectivity toward CO2. Our work validated the potential of transition metal cation-exchanged SSZ-13 zeolites for efficient CO2 capture, as evaluated by unary static isothermal adsorption and binary dynamic column breakthrough experiments as well as predicted performance in pressure/vacuum swing adsorption (P/VSA) process. Among the investigated transition metals (Co(II), Ni(II), Zn(II), Fe(III), Cu(II), Ag(I), La(III), and Ce(III)) exchanged SSZ-13, Co(II)/SSZ-13 and Ni(II)/SSZ-13 showed the highest CO2 uptake (4.49 and 4.45 mmol/g, respectively) and superior selectivity of CO2 over N2 (52.55 and 42.61, respectively) at 273 K and 1 atm. We attribute such outstanding separation performance to the Pi backdonation exclusively formed between CO2 and transition metal cation sites. This demonstrates a new approach of developing adsorbents for CO2 capture in the real-world industrial processes.

Semi-supervised mixture of latent factor analysis models with application to online key variable estimation

Data-driven virtual sensors have extensive applications in industrial processes for online estimating those important but difficult-to-measure variables. In the virtual sensor application, labeled samples could be very infrequent due to technical or economical limitations, thereby virtual sensors developed upon insufficient labeled samples may not be well trained, which leads to poor estimation performance. In addition, industrial processes are inherently stochastic and the vast majority of them present nonlinear and non-Gaussian characteristics. To cope with these issues, this paper proposes a semi-supervised mixture of latent factor analysis models (S2MLFA). In the S2MLFA, the insufficiency of labeled samples is remedied by exploiting both labeled and unlabeled data sets, while the nonlinear and non-Gaussian characteristics are handled by the mixture model structure. Moreover, the process uncertainties are modeled by the probabilistic model formulation. An efficient expectation–maximization-based learning algorithm is developed for training the S2MLFA, and a modified Akaike information criterion is presented for model selection. The S2MLFA is investigated by a numerical example and two real-world industrial processes, through which the effectiveness and feasibility of the proposed schemes are verified.

Rheology of cubic particles suspended in a Newtonian fluid.

Many real-world industrial processes involve non-spherical particles suspended in a fluid medium. Knowledge of the flow behavior of these suspensions is essential for optimizing their transport properties and designing processing equipment. In the present work, we explore and report on the rheology of concentrated suspensions of cubic-shaped colloidal particles under steady and dynamic shear flow. These suspensions exhibit a rich non-Newtonian rheology that includes shear thickening and normal stress differences at high shear stresses. Scalings are proposed to connect the material properties of these suspensions of cubic particle to those measured for suspensions of spherical particles. Negative first normal stress differences indicate that lubrication hydrodynamic forces dominate the stress in the shear-thickened state. Accounting for the increased lubrication hydrodynamic interactions between the flat surfaces of the cubic particles allows for a quantitative comparison of the deviatoric stress in the shear-thickened state to that of spherical particles. New semi-empirical models for the viscosity and normal stress difference coefficients are presented for the shear-thickened state. The results of this study indicate that cubic particles offer new and unique opportunities to formulate colloidal dispersions for field-responsive materials.

Optimal control of grinding mill circuit using model predictive static programming: A new nonlinear MPC paradigm

The recently developed reference-command tracking version of model predictive static programming (MPSP) is successfully applied to a single-stage closed grinding mill circuit. MPSP is an innovative optimal control technique that combines the philosophies of model predictive control (MPC) and approximate dynamic programming. The performance of the proposed MPSP control technique, which can be viewed as a ‘new paradigm’ under the nonlinear MPC philosophy, is compared to the performance of a standard nonlinear MPC technique applied to the same plant for the same conditions. Results show that the MPSP control technique is more than capable of tracking the desired set-point in the presence of model-plant mismatch, disturbances and measurement noise. The performance of MPSP and nonlinear MPC compare very well, with definite advantages offered by MPSP. The computational speed of MPSP is increased through a sequence of innovations such as the conversion of the dynamic optimization problem to a low-dimensional static optimization problem, the recursive computation of sensitivity matrices and using a closed form expression to update the control. To alleviate the burden on the optimization procedure in standard MPC, the control horizon is normally restricted. However, in the MPSP technique the control horizon is extended to the prediction horizon with a minor increase in the computational time. Furthermore, the MPSP technique generally takes only a couple of iterations to converge, even when input constraints are applied. Therefore, MPSP can be regarded as a potential candidate for online applications of the nonlinear MPC philosophy to real-world industrial process plants.

A new verification technique for large processes based on identification of relevant tasks

Verification recently has become a challenging topic for business process languages. Verification techniques like model checking allow to ensure that a process complies with domain-specific requirements, prior to the execution. To execute full-state verification techniques like model checking, the state space of the process needs to be constructed. This tends to increase exponentially with the size of the process schema, or it can even be infinite. We address this issue by means of requirements-specific reduction techniques, i.e., reducing the size of the state space without changing the result of the verification. We present an approach that, for a given requirement the system must fulfill, identifies the tasks relevant for the verification. Our approach then uses these relevant tasks for a reduction that confines the process to regions of interest for the verification. To evaluate our new technique, we use real-world industrial processes and requirements. Mainly because these processes make heavy use of parallelization, full-state-search verification algorithms are not able to verify them. With our reduction in turn, even complex processes with many parallel branches can be verified in less than 10s.

Scheduling just-in-time part supply for mixed-model assembly lines

With increasing cost competition and product variety, providing an efficient just-in-time (JIT) supply has become one of the greatest challenges in the use of mixed-model assembly line production systems. In the present paper, therefore, we propose a new approach for scheduling JIT part supply from a central storage center. Usually, materials are stored in boxes that are allotted to the consumptive stations of the line by a forklift. For such a real-world problem, a new model, a complexity proof as well as different exact and heuristic solution procedures are provided. Furthermore, a direct comparison with a simple two-bin kanban system is provided. Such a system is currently applied in the real-world industrial process that motivates our research. It becomes obvious that this policy is considerably outperformed according to the resulting inventory- and α-service levels. Moreover, at the interface between logistics and assembly operations, strategic management implications are obtained. Specifically, based on the new approach, it is the first time a statistical analysis is being made as to whether widespread Level Scheduling policies, which are well-known from the Toyota Production System, indeed facilitate material supply. Note that in the literature it is frequently claimed that this causality exists.

LRProb control chart based on logistic regression for monitoring mean shifts of auto-correlated manufacturing processes

With the growth of automation in manufacturing, process quality characteristics are being measured at higher rates and data is more likely to be auto-correlated. Traditional statistical process control (SPC) techniques of control charting are not applicable in many process industries because process parameters are highly auto-correlated. Several attempts such as some time series based control charts have been made in the previous years to extend traditional SPC techniques. However, these extensions pose some serious limitations for monitoring the process mean shifts. These charts require that a suitable model has been identified for the time series of process observations before residuals can be obtained. In this paper, a logistic regression (LR)-based process monitoring model is proposed for enhancing the monitoring of processes. It is capable of providing a comprehensible and quantitative assessment value for the current process state, which is achieved by the event occurrence probability calculation of LR. Based on these probability values over the time series, a novel chart: LRProb chart, is developed for monitoring and visualising process changes. The aim of this research is to analyse the performance of the LRProb chart under the assumption that only a small number of predictable abnormal patterns are available. To such aim, the performance of the LRProb chart is evaluated on two real-world industrial cases and simulated processes. Given the simplicity, visualisation and quantification of the proposed LRProb chart, this approach is proved from the experiments to be a feasible alternative for quality monitoring in the case of auto-correlated process data.

Modelling Gas Processing Unit: An Inter-disciplinary Approach Based on Petri Nets

A modelling approach that will facilitate an in-depth understanding of the interactions of the different phenomena, human interactions and environmental factors constituting �real world� industrial processes is presented. An important industrial system such as Gas Processing Unit (GPU) have inter-related internal process activities coexisting with external events and requires a real time inter-disciplinary approach to model them. This modeling framework is based on identifying as modules, the part of processes that have interactions and can be considered active participants in overall behaviour. The selected initial set of modules are structured as Petri net models and made to interact iteratively to provide process states of the system. The modeling goal is accomplished by identifying the evolution of the process states as a means of effective representation of the �actual running�� of the industrial process. The paper discusses the function and the implementation of the modelling method as applicable to the industrial case of GPU.

Plant Models for Use with Computer-Aided Control System Analysis and Design Packages

A portfolio of plant models based upon real-world industrial plants and processes has been developed for use in control system analysis and design. The seven nonlinear models in this portfolio are Written In two widely-used simulatIon languages, ACSL and TSIM 2. A set of linear models has been extracted from the control literature and put in the form of Matlab and Ecstasy files. The nonlinear models are being formed into a model library that will be accessible over an academic data network in the United Kingdom

Nonlinear process control with neural nets

There exists a need to develop better mathematical models for physical processes, but it is also essential to understand that it is impossible to describe real-world industrial processes by exact mathematical models. Pattern recognition based methodology implemented in the architecture of artificial neural network can be used to model knowledge intensive feedback control systems. The procedure for development and practical design of neural-net based control systems is described and demonstrated by the example of a nonlinear hydraulic system. The results obtained in computer simulations and experiments are presented to illustrate the new approach.