Search

Quality-relevant process monitoring provides important guarantees for the safety of industrial operation, which is based on the assumption that data collection is complete and low-order autocorrelated. However, real industrial processes always exhibit complex characteristics such as multi-rate sampling and high-order dynamic, which pose great challenges for process monitoring. To this end, a multi-rate high-order dynamic twin-latent-variable probabilistic (MHDTVP) model is presented in this paper to extract data correlations among multi-rate measurements from quality-relevant and irrelevant perspectives. Moreover, to reveal the dynamics in the multi-rate sampling process, an autoregressive twin-latent-variable structure is designed to extract both quality-relevant and quality-irrelevant high-order dynamic features. In the MHDTVP model, parameters are trained through an efficient expectation maximization (EM) iteration framework. Finally, the performance conclusions of MHDTVP are validated with the Tennessee Eastman process (TEP) and Thermal Power Plant (TPP). The experimental results demonstrate that the proposed model exhibits superior monitoring efficiency for multi-rate dynamic processes compared to similar approaches. [Display omitted] • A novel twin-latent-variable probabilistic model is proposed for the purpose of addressing quality-relevant and quality-irrelevant dynamic process monitoring. • To handle measurements with multiple sampling rates, a multi-rate strategy is designed based on the proposed model. • To reveal high-order autocorrelation, the model breaks the bottleneck of traditional first-order dynamic modeling in multi-rate processes. [ABSTRACT FROM AUTHOR]

This study established a numerical model for soil-structure interaction (SSI) to examine the effects of the spatial incidence angle of SV waves and soil nonlinearity, utilizing viscoelastic artificial boundaries (VAB) and equivalent nodal force (ENF) method. Both the foundation's and superstructure's torsion and rocking responses were then analyzed. The findings indicate that subjected to spatially oblique incident SV waves, the rectangular foundation primarily has the rocking response while the torsional response is negligible. Furthermore, the maximum torsional and rocking angles about the x-axis at each frame floor are significantly enlarged by comparison with the perpendicular incident case. Moreover, the soil nonlinearity could increase the foundation's rocking angle and enlarge the maximum torsion and rocking responses of the structure's floors. Consequently, structural seismic damage assessment requires considering both the soil nonlinearity and incident seismic wave angles. • The nonlinear behavior of 3D soil under spatial oblique incident SV waves is simulated using the revised Davidenkov model. • The torsion and rocking response of foundations and frameworks under spatial oblique incident SV waves is investigated. • The incident angle of SV waves and soil nonlinearity influence the dynamic responses of the superstructure and foundation. [ABSTRACT FROM AUTHOR]

Mud layer height of thickener is the key quality index of thickening process which is difficult to achieve real-time detection with existing methods in reality. While the need of developing a soft sensor model which can be used for real-time detection of mud layer height, we proposed an end-to-end mud layer height prediction method with attention mechanism-based convolutional neural network (CNN). The dynamic features are firstly extracted from the image samples based on CNN, and then two types of attention mechanism are embedded sequentially to contribute to more precise prediction results. Compared with the traditional spatial attention mechanism, the regional spatial attention mechanism we proposed selectively divides the spatial feature map into regions, while regions containing important features are assigned larger weights. Adding the channel and regional spatial attention mechanism in CNN not only effectively improve both the precision and calculation speed, but also affect the dimension of the output feature map, so as to avoid the loss of channel or spatial attention information of the feature map. To verify the validity of the proposed method, different attention mechanisms are embedded in the CNN, and the corresponding experiments are carried out on the dataset of the thickener mud layer. The experimental results demonstrate the feasibility and effectiveness of the mud layer height prediction method. • A novel convolution neural network model combined with attention mechanism is designed for mud layer height of the thickener prediction. • A regional spatial attention mechanism is proposed which divides the feature map into regions in the spatial dimension according to the distribution of features, and the weights of each region are constructed to increase the weights of the regions containing the key information. • Applying the proposed method of DAM with region divided to the thickening process improves the network features extraction capability, and the prediction accuracy. [ABSTRACT FROM AUTHOR]

The process operating performance assessment is critical for the smelting process of electro-fused magnesium furnaces to improve quality of the magnesia product and pursue optimal comprehensive economic benefit. This paper proposes a new method of multi-source heterogeneous information deep feature fusion (MSHIDFF) to achieve higher accuracy operating performance assessment in the electro-fused magnesium smelting process. Firstly, we utilize convolutional neural network, bidirectional long short-term memory network and stacked auto-encoder to extract deep features from raw image, sound and current of different performance grades. Furthermore, those multi-source deep features are fused and the softmax regression with attention mechanism is employed to train a neural network classifier for the fused deep features of different performance grades. The simulation results show that the proposed MSHIDFF method obtains the superior assessment accuracy. • CNN, Bi-LSTM and SAE are used to extract the features of image, sound and current. • The depth features of different information are fused to form comprehensive features. • Softmax with AM is trained to assess the comprehensive features of different grades. • The proposed MSHIDFF method is applied to the EFMF and obtains the superior accuracy. [ABSTRACT FROM AUTHOR]

• Health state assessment and remaining useful life prognostic of hydrocyclone • Augmenting degradation sequences by fusing mechanism modeling and TimeGAN learning • A unified framework for constructing and clustering degradation health indicators • Quantifying prediction uncertainty of remaining useful life of equipment Deploying health state (HS) assessment and remaining useful life (RUL) prognostic on the industrial equipment can effectively avoid shutdown and reduce productivity losses. This paper proposes a temporal self-clustering (TSC) and time-series generative adversarial network (TimeGAN) based uncertainty-aware health prognostic algorithm for the hydrocyclone, a critical machine in the mineral processing. To label the HS of degradation data, an autoencoder based TSC algorithm is proposed. It integrates the extraction and clustering of health indicators into a unified framework, so that the correlation information between feature extraction and clustering can be captured for cluster analysis. A Benders decomposition based optimization approach is designed to synthetically optimize network weights and clustering centers, thereby extracting and labeling data features more reasonably. For quantifying prediction uncertainty, an uncertainty-aware prognostic algorithm is proposed to predict the RUL distribution, and the uncertainty can be quantified as the standard deviation (Std) of the distribution. Furthermore, a mechanism and TimeGAN fusion temporal sequence generation algorithm is proposed to augment the degradation process sequences of the equipment. Verification experiments using the actual data show that the proposed algorithm can effectively evaluate the HS of the equipment and provide a sensible and convincing RUL prediction. [ABSTRACT FROM AUTHOR]

In the industrial flotation process, the operational adjustment is still manual, which mainly relies on the operator's observation of the flotation froth. Due to limited experience or operation lag, technical indexes such as concentrate grade are difficult to control within the qualified range. In the era of big data, case-based reasoning (CBR) and Bayesian network (BN) are two advanced technologies that can realize intelligent operational adjustment. Although CBR is highly reliable, it is rough and has poor generalization performance. Besides, BN is challenging in responding to multi-working conditions and strong nonlinearities. Inspired by the advantages of the integrated models, a two-step meticulous operational adjustment approach for the flotation process combining CBR and BN is proposed in this article. A case library is constructed in the offline stage, consisting of cases whose technical index has been improved by an operational adjustment in history. After introducing a new case, the rough operational adjustment solution is first determined by CBR. Based on this, a new incremental database is constructed and used for online training of the BN model. After receiving the evidence of the new case's problem attribute, the precise operational adjustment can be determined by BN reasoning. The final case solution to be performed is the sum of the rough and precise operational adjustment received in the two steps. Experiments in a real-world copper flotation process verify the performance and merit of the proposed hybrid approach. The results show that intelligent operational adjustment can significantly improve the copper concentrate grade index. • A hybrid approach integrating case-based reasoning and Bayesian network is proposed. • Proposed approach is applied to the operational adjustment of industrial flotation process. • Proposed integrated approach owns better performance compared with single CBR and BN. [ABSTRACT FROM AUTHOR]

Industrial processes often show typical nonstationarity, resulting from frequent variation of working condition. It is rather challenging to assess the operating performance of nonstationary processes, since the process property is time-variant and the standard of performance levels may differ from working conditions. Besides the nonstationarity, nonlinearity widely exists in industrial processes, making it hard to analyze process characteristics. To solve the above problems, a three-dimensional (3-D) online modeling and assessment method of operating performance is proposed in the present work. Considering that the inner drive of nonstationarity is often the fluctuation of working condition, 3-D space is established, including sample, variable, and working condition axes. Thus, the process can be modeled and assessed online along direction of the working condition indicator to handle both nonstationarity and nonlinearity. Within the 3-D space, the selection of online training data and establishment of assessing model are solved simultaneously through multi-objective optimization, to make the training data suitable for performance assessment and preserve high assessment accuracy. Bayesian inference based criterion is then defined to determine the performance level. Finally, the validation on a three-phase flow facility and a coal mill of a power plant show feasibility and superiority of the proposed method on performance assessment for nonstationary industrial processes with nonlinearity. • Performance assessment is solved for general nonlinearly nonstationary processes. • 3-D online modeling and assessment method is proposed. • Determination of training samples and model are conducted simultaneously. • Solution of optimization involves GA and eigenvalue decomposition. [ABSTRACT FROM AUTHOR]

Research on energy-saving retrofits of the existing industrial systems can effectively improve the systems' energy-saving level and increase company revenue. The energy-saving bottleneck diagnosis for systems is the foundation of the energy-saving retrofit design. The diagnosis also aims to yield a more targeted energy-saving retrofit, thus ensuring the energy-saving effect and investment income of the transformation project. Therefore, this paper proposes a new method to diagnose the energy-saving bottleneck of the existing industrial systems. First, the energy-saving bottleneck diagnosis is decomposed into two parts: the diagnosis of energy-saving bottleneck units and the diagnosis of energy-saving bottleneck factors. Then, the energy density decomposition method is improved and a multi-mode energy efficiency density decomposition method is proposed. Next, an energy-saving bottleneck diagnosis method for industrial systems based on multi-mode energy efficiency density decomposition and an orthogonal test method is proposed. The proposed method can analyze the energy-saving potential of existing systems, accurately locate energy-saving bottlenecks, and avoid the dependence on a large amount of expert experience or complete production data in the existing methods. Finally, the proposed method is applied to an actual circulating cooling water system and the system energy-saving bottleneck diagnosis is realized, thus verifying the practicability of the method. • Energy-saving bottlenecks in industrial systems are analyzed and decomposed. • A method for energy-saving bottleneck diagnosis from two directions is proposed. • An improved energy efficiency density decomposition method is proposed. • The diagnosis of energy-saving bottlenecks in industrial systems is achieved. • The verification is performed for an industrial circulating cooling water system. [ABSTRACT FROM AUTHOR]

Assessment of operating performance evaluates optimal degree for processes under normal conditions. Maintaining in good operating performance indicates high benefit of the production. To assess the performance for a newly built process with a small number of historical data, a cross-domain feature transfer learning (CDFTL) method is proposed in this research. The assumption that the data from the source and target domains should be independent and identically distributed is no longer necessary. The common and specific features are both utilized to guarantee the assessment precision and model generalization ability. Finally, the proposed method is applied to a copper flotation process. [ABSTRACT FROM AUTHOR]

Abstract To adapt to the change in the environment, the model needs to own the ability to update to ensure the performance of the decision. In this paper, a new updating strategy is proposed for the safe control Bayesian network (BN) model under the abnormity in the thickening process of gold hydrometallurgy. First of all, the abnormality in the thickening process of gold hydrometallurgy is analyzed deeply. The new safe control BN model is established for three main abnormities. Furthermore, the general framework of BN model updating strategy based on the new expert knowledge and the new dataset is proposed, which mainly includes the parameter updating learning and the structure updating learning. For the structure updating learning, the useful information in the established model is reserved, and only the partial structure which does not adapt to the change in the environment is changed by searching for the unconformable nodes as the target nodes. Finally, the proposed method is applied to update the safe control BN model in the thickening process of gold hydrometallurgy. The simulation results demonstrate that it is effective and owns the better performances to update the established model as the change of the dosage of flocculants. [ABSTRACT FROM AUTHOR]

This study proposed a decision system that can output the flotation backbone flowchart using the natural properties of copper ore. The proposed decision system includes three decision tasks: product scheme, flotation scheme and grinding scheme. Each decision task is a multi-label classification problem. To improve the classification effect of each sub-label, extreme gradient boosting (XGBoost) is used as a subclassifier, because of its ability to deal with small and high-dimensional samples. To selectively utilize the relations between the sub-labels in the same task, a modified classifier chain (MCC) was proposed. To specifically use the effect of a front-end task on a back-end task, the decision system connects the MCC-XGBoost corresponding to the three tasks in series. Accordingly, the outputs of a front-end task becomes the candidate features of a back-end task. To improve the recall rates of minority classes, the classification thresholds are customized using the Yoden index. Finally, the high performance of the decision system was demonstrated by hypothesis testing. [ABSTRACT FROM AUTHOR]

This paper presents a control framework to identify the abnormal condition and make safe control scheme by fusing the related multi-source information based on the analysis of the abnormal exhausting condition in the electro-fused magnesia smelting process. For the abnormal condition identification, the case-based reasoning is used to infer the degree of the abnormal condition. When the similarity of the matched case doesn't exceed the set threshold, Bayesian network is used to infer the degree. For the safe control scheme, taking advantage of the idea of the remaining lifetime predicting, and the adjustment of the control variables is obtained. The application results show that the proposed method is effective and it owns better performance than the manual and the existing research results. [ABSTRACT FROM AUTHOR]

In his research, a modified Dynamic Causal Diagram (DCD) based process operating performance optimality assessment and non-optimal cause identification method is proposed to handle the online assessment of the complicated chemical processes, where certainties and uncertainties coexist. In addition, the multi-source information usage problem of the traditional DCD is addressed. In the online application, a newly developed scheme is developed, which distinguishes stable performance grade and grades switch transition. For the non-optimal performances, a contribution based root cause identification method is developed, on the framework of the constructed DCD. Finally, the operating performance of a gold hydrometallurgy process is assessed by both the traditional DCD and the modified DCD. The superiority on accuracy and robustness of the modified DCD is demonstrated, and the responsible events for non-optimality is successfully isolated by the proposed contribution based non-optimal cause identification technique. [ABSTRACT FROM AUTHOR]

This paper develops a safe control scheme under the abnormity based on Bayesian network (BN) for the thickening process of gold hydrometallurgy. By analyzing the causes and corresponding solutions of the abnormity, the operator experience of removing the abnormity is transformed to BN which is used to make safe decisions automatically. The proposed BN-based approach can provide real-time decision support when the abnormity happens. The BN combines the expert knowledge with quantitative data analysis for the abnormity which can make decisions to remove the abnormity. The BN is established off-line and used to infer on-line. After receiving abnormal phenomena as evidences, the posterior probabilities of the decision variables with different grades can be obtained by BN reasoning, which provide real-time safety analysis. The posterior probabilities can be updated using the abnormal latest information. If the abnormity is not removed, once again the BN analysis is invoked based on the newly established situation and the appropriate set of decisions are obtained accordingly. The application results show that the proposed approach can make the safe operation control decisions to remove the abnormity in the thickening process effectively. [ABSTRACT FROM AUTHOR]

To pursue optimal comprehensive economic benefit, the assessment on the operational performance of the process becomes more and more critical for industrial processes. In this study, a novel assessment method regarding the operating optimality as well as the cause identification strategy are proposed for nonlinear industrial processes based on nonlinear optimality related variation information (NORVI). The proposed method is committed to extracting the NORVI from each performance grade and using it to establish the assessment model. Since the process variation information unrelated to operating optimality is abandoned in evaluation, the accuracy of the assessment results is improved. Additionally, owing to only the process data used in developing the assessment model, the time-consuming data alignment work is avoided, which improves the application efficiency of the proposed algorithm. Furthermore, based on the similarities between the NORVI of the test data and those of the modeling data of each performance grade, the actual operational performance of the process can be evaluated in real time. For the nonoptimal performance grade, the idea of variable selection is used to develop the cause identification strategy. Finally, the efficiency of the proposed method is illustrated with a case of gold hydrometallurgical process. [ABSTRACT FROM AUTHOR]