Modeling Of Complex Industrial Processes Research Articles

Adaptive Multi-Head Self-Attention Based Supervised VAE for Industrial Soft Sensing With Missing Data

Variational auto-encoders (VAEs) have been widely used in soft sensing due to their ability to provide a probabilistic description of the hidden space. However, VAEs are static models that do not consider process dynamics, which can limit the ability of VAEs to accurately model complex industrial processes. To tackle this problem, this paper proposes a model called adaptive multi-head self-attention based supervised VAE (AMSA-SVAE). In AMSA-SVAE, an adaptive multi-head self-attention mechanism (AMSA) is proposed based on the multi-head self-attention mechanism (MSA). AMSA can dynamically extract different attention information depending on specific tasks. By adjusting the attention weights based on the input sequence, AMSA allows for more accurate and efficient modeling of complex industrial processes. Then, AMSA is used as the encoder and decoder of SVAE for soft sensing. Furthermore, with the data generation capabilities of VAE, an adaptive multi-head self-attention based VAE (AMSA-VAE) framework is proposed to address the issue of missing data. The AMSA-VAE is used to dynamically fill in missing data, thereby extending the capabilities of AMSA-SVAE. Finally, the performance of AMSA-SVAE is verified by a set of real industrial data, and the ability of AMSA-VAE framework is demonstrated by simulating different degrees of data missing rates. By combining the dynamic modeling capabilities of AMSA-SVAE with the data generation capabilities of AMSA-VAE, the proposed approach provides a robust solution to the challenges of incomplete data in soft sensing. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> — Soft sensors are widely used to measure key parameters in industrial processes, but missing values in the data are common due to sensor failures or transmission signal interference. This poses a significant challenge for traditional soft sensors, which require complete data to accurately model. Meanwhile, the dynamic nature of industrial process data further complicates the modeling process. To solve these challenges, this paper proposes an AMSA-SVAE model for soft sensing and an AMSA-VAE framework for filling in the missing values in the data, thereby extending the capabilities of AMSA-SVAE to handle missing data. When facing a dataset with missing values, AMSA-VAE framework is first used to fill in the missing values before the filled complete data is fed into AMSA-SVAE for modeling. Finally, the proposed approaches are evaluated through two sets of experiments using a real industrial dataset, showing the excellent performance of AMSA-SVAE and AMSA-VAE framework in modeling dynamic industrial process data and addressing the missing data problem.

Modeling of complex industrial process based on active semi-supervised clustering

Since industrial processes have a wide range of operating conditions, it is difficult to build a single global model that describes a process. One solution that is widely used in control engineering practice is to combine multiple models based on collected process data. For this approach to be successful, it is important to cluster the data before the modeling. In this study, pairwise constraints and an active-learning method were incorporated into the affinity propagation algorithm, resulting in a new method called active semi-supervised affinity propagation (ASSAP) clustering. To apply ASSAP to the modeling of industrial processes, an active-learning strategy is firstly used to obtain constraints on data based on the angle of change between two data points and the probability of their belonging to the same class, and then the constraints are used to adjust the clustering process so as to improve the clustering precision. Finally, the least-squares-support-vector-machine (LS-SVM) is used to build a submodel for each cluster of data points, and then all the sub-models are integrated into a model for the whole data set.Verification of the ASSAP method was carried out on data from the UCI (University of California, Irvine) Machine Learning Repository and Olivetti dataset. In addition, ASSAP and LS-SVM are combined to be applied to the data of the combustion process of a coke oven. The result shows the effectiveness of the method of modeling of complex industrial process based on ASSAP.

A Parallel Optimization Algorithm based on FANOVA Decomposition

The analysis and modeling of complex industrial processes, like the forming of car parts, is often performed with the help of computer simulations. Optimization of such computer experiments usually relies on metamodel‐based sequential strategies. The existing sequential algorithms, however, share the limitation that they only allow a single simulation at a time. In this article, we present a very elegant way to produce a parallel optimization procedure, based on a technique from the sensitivity analysis toolbox—the functional analysis of variance graph. The proposed novel simultaneous optimization scheme is called the ParOF algorithm. It is compared with a very effective black‐box procedure—the well known efficient global optimization (EGO) algorithm, based on analytical test cases and an optimization study of a sheet forming simulation. Besides demonstrating the advantages of our parallel optimization method, the results show that it can successfully be applied to sheet metal forming for the purpose of quality improvement of the ready parts.Copyright © 2014 John Wiley &amp; Sons, Ltd.

Evolving Nonlinear Time-Series Models of the Hot Metal Silicon Content in the Blast Furnace

Neural networks are versatile tools for nonlinear modeling, but in time-series modeling of complex industrial processes the choice of relevant inputs and time lags can be a major problem. A novel method for the simultaneous detection of relevant inputs and an appropriate structure of the lower part of the networks has been developed by evolving neural networks by a genetic algorithm, where the approximation error and the number of weights are minimized simultaneously by multiobjective optimization. The networks on the Pareto front are considered possible candidate models that are evaluated on an independent test set. In order to consider the problem of drift in the variables, which may cause parsimonious models to perform poorly on the test set, the weights in the upper layer of the networks are recursively estimated by a Kalman filter. The method is illustrated on a data set from ironmaking industry, where time-series models of the hot metal silicon content in a blast furnace are evolved. The technique is demonstrated to synthesize models with a choice of inputs in agreement with findings presented in the literature and process know-how.

Process Monitoring and Modeling Using the Self-Organizing Map

The Self-Organizing Map (SOM) is a powerful neural network method for analysis and visualization of high-dimensional data. It maps nonlinear statistical dependencies between high-dimensional measurement data into simple geometric relationships on a usually two-dimensional grid. The mapping roughly preserves the most important topological and metric relationships of the original data elements and, thus, inherently clusters the data. The need for visualization and clustering occurs, for instance, in the analysis of various engineering problems. In this paper, the SOM has been applied in monitoring and modeling of complex industrial processes. Case studies, including pulp process, steel production, and paper industry are described.