Machine Learning for Process Systems Engineering, Classification, Estimation, Prediction, and Updating

A special issue of Processes (ISSN 2227-9717). This special issue belongs to the section "Process Control and Monitoring".

Deadline for manuscript submissions: closed (30 April 2024) | Viewed by 15000

Special Issue Editors


E-Mail Website
Guest Editor
CRSI LaRGES, Lebanese University, Tripoli 1300, Lebanon
Interests: system engineering; fault diagnosis; failure prognosis; modeling; simulation; AI

E-Mail Website
Guest Editor
LIS-Lab, Aix Marseille University, 13007 Marseille, France
Interests: fault diagnosis; failure prognosis; modeling; simulation; AI
Special Issues, Collections and Topics in MDPI journals

E-Mail Website
Guest Editor
Faculty of Engineering and Information Technology, City University, 1300 Tripoli, Lebanon
Interests: fault diagnosis; failure prognosis; energy; hybrid vehicle

Special Issue Information

Dear Colleagues,

Control, FDI and FTC problems are defined in several cases of application as classification or regression issues. For example, the detection and identification of faults in complex systems in the presence of multiple faults can be considered as a classification problem, and the estimation of unmeasurable state variables for the control of a system as a regression problem. Artificial intelligence tools such as Machine Learning, Deep Learning, and Reinforcement Learning are powerful methods increasingly used to provide robust solutions to these issues.

This Special Issue aims to gather and highlight works using artificial intelligence tools to solve classification, regression and model-updating problems on application cases in various fields, such as sensors, microelectronics, transport and energy.

Prof. Dr. Nazih Moubayed
Dr. Mohand Djeziri
Dr. Hiba Al-Sheikh
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Processes is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2400 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • process systems engineering
  • machine learning
  • deep learning
  • reinforcement learning
  • FDI-FTC
  • failure prognosis
  • classification
  • regression
  • estimation

Published Papers (12 papers)

Order results
Result details
Select all
Export citation of selected articles as:

Research

12 pages, 3143 KiB  
Article
New Method for Monitoring and Early Warning of Fracturing Construction
by Jiani Hu, Meilong Fu, Yang Yu and Minxuan Li
Processes 2024, 12(4), 765; https://doi.org/10.3390/pr12040765 - 10 Apr 2024
Viewed by 517
Abstract
During fracturing operations, special situations are often encountered. For example, the insufficient proppant-carrying capacity of fracturing fluid can cause quartz sand or ceramsite to settle near the wellbore and form a sand plug. Alternatively, excessive sand injection intensity can lead to severe accumulation [...] Read more.
During fracturing operations, special situations are often encountered. For example, the insufficient proppant-carrying capacity of fracturing fluid can cause quartz sand or ceramsite to settle near the wellbore and form a sand plug. Alternatively, excessive sand injection intensity can lead to severe accumulation of injected sand near the wellbore and also form a sand plug. These special situations are reflected in the fracturing operation curve as an abnormal increase in oil pressure over a short period of time. If not handled promptly, they can have unimaginable consequences. Sand plugs in fracturing operations, characterized by their speed and unpredictability, often form rapidly, within about 20 s. Conventional methods for on-site sand-plug warnings during fracturing include the oil pressure–time double logarithmic slope method and the net pressure–time double logarithmic slope method. Although these methods respond quickly, their warning results are unstable and vary significantly during actual operations. This is mainly because the fluctuations in the actual fracturing operation curve are often large, and there can be sudden pressure rises and drops even during stable periods, albeit less pronounced. To address the identification of anomalies in conventional fracturing operation monitoring and warning methods, a sand-plug warning index method has been proposed for sand-plug identification. This method combines the oil pressure–time double logarithmic slope with the oil pressure increment within 5 s, the rate of change in the oil pressure–time double logarithmic slope, and the fitted oil pressure intercept as indicators. The method has been validated using Well A in Fuling as an example. The validation results show that the dynamic analysis method can predict sand plugs while reducing warning fluctuations without affecting sensitivity. Compared to conventional methods, the warning time can be advanced by about 10 s. Full article
Show Figures

Figure 1

15 pages, 6322 KiB  
Article
Research on the Calculation Method for the Wellbore Temperature of Hot Nitrogen Gas Circulation Wax Removal and Plug Removal in Offshore Oil Fields
by Weigang Du, Yongsheng An, Liyong Guan, Chengchen Xiong, Runshi Huo and Bing Tang
Processes 2023, 11(11), 3196; https://doi.org/10.3390/pr11113196 - 9 Nov 2023
Viewed by 794
Abstract
The problem of wax deposition widely exists in offshore oil fields; waxing of the oil tubing will lead to a reduction in the cross-sectional area of the flow and, in serious cases, the flow path will be blocked, causing the well to stop [...] Read more.
The problem of wax deposition widely exists in offshore oil fields; waxing of the oil tubing will lead to a reduction in the cross-sectional area of the flow and, in serious cases, the flow path will be blocked, causing the well to stop production. In order to cope with this problem, a thermal dynamic wax removal method has emerged in recent years that utilizes hot nitrogen gas circulation between the oil tube and annulus to raise the temperature of the oil tube to achieve the purpose of wax removal and plug removal and is quick and easy to operate. Unlike conventional wellbore temperature calculation methods, the wellbore temperature field under hot nitrogen circulation conditions is influenced both by the reservoir temperature gradient and the hot nitrogen injection temperature, injection pressure, and injection rate. In this paper, a temperature calculation model for a wellbore considering both annulus injection temperature and tubing temperature and their interactions is modeled, which can consider the effects of different hot nitrogen injection temperatures, injection rates, and injection pressures. The model is used to calculate the temperature distribution for different injection parameters in order to ensure that the tubing temperature is higher than the wax precipitation temperature and that the annulus temperature is not higher than the maximum temperature resistance of the rubber in the packer. The study provides a design method for wax removal and plug removal with hot nitrogen gas circulation. Full article
Show Figures

Figure 1

17 pages, 6147 KiB  
Article
Optimizing Task Completion Time in Disaster-Affected Regions with the WMDDPG-GSA Algorithm for UAV-Assisted MEC Systems
by Tianhao Ma, Yulu Yang, Han Xu and Tiecheng Song
Processes 2023, 11(10), 3000; https://doi.org/10.3390/pr11103000 - 18 Oct 2023
Cited by 1 | Viewed by 853
Abstract
In this paper, we investigate a UAV-assisted Mobile Edge Computing (MEC) system that is deployed with multiple UAVs to provide timely data processing services to disaster-stricken areas. In our model, since base stations are unavailable in disaster-affected areas, we solely employ UAVs as [...] Read more.
In this paper, we investigate a UAV-assisted Mobile Edge Computing (MEC) system that is deployed with multiple UAVs to provide timely data processing services to disaster-stricken areas. In our model, since base stations are unavailable in disaster-affected areas, we solely employ UAVs as MEC servers, as well as enable real-time data transmission during UAV flights by estimating and compensating for the Doppler Frequency Shift (DFS). Subsequently, an optimization problem is formulated to jointly optimize the trajectories and offloading strategies of multiple UAVs to minimize the task completion time. We enhance the performance of the Multi-Agent Deep Deterministic Policy Gradient (MADDPG) algorithm by using a weighted strategy algorithm, and we thus propose the Weighted-Strategy-Based Multi-Agent Deep Deterministic Policy Gradient (WMDDPG) algorithm for optimizing UAV trajectories. We employ the Greedy-Based Simulated Annealing (GSA) algorithm to overcome the limitations of the greedy algorithm and to obtain the best offloading strategy. The results demonstrate the effectiveness of the proposed WMDDPG-GSA algorithm, as it outperforms benchmark algorithms. Full article
Show Figures

Figure 1

16 pages, 2851 KiB  
Article
Mathematical Modeling of Prediction of Horizontal Wells with Gravel Pack Combined with ICD in Bottom-Water Reservoirs
by Shili Qin, Ning Zhang, Bobo Cao, Yongsheng An and Runshi Huo
Processes 2023, 11(9), 2777; https://doi.org/10.3390/pr11092777 - 17 Sep 2023
Viewed by 589
Abstract
During the development of horizontal wells in bottom-water reservoirs, the strong heterogeneity of reservoir permeability leads to premature bottom-water breakthroughs at locations with high permeability in the horizontal wellbore, and the water content rises rapidly, which seriously affects production. To cope with this [...] Read more.
During the development of horizontal wells in bottom-water reservoirs, the strong heterogeneity of reservoir permeability leads to premature bottom-water breakthroughs at locations with high permeability in the horizontal wellbore, and the water content rises rapidly, which seriously affects production. To cope with this problem, a new technology has emerged in recent years that utilizes gravel filling to block the flow in the annulus between the horizontal well and the borehole and utilizes the Inflow Control Device (ICD) completion tool to carry out segmental water control in horizontal wells. Unlike conventional horizontal well ICD completions that use packers for segmentation, gravel packs combined with ICD completions break the original segmentation routine and increase the complexity of the production dynamic simulation. In this paper, the flow in different spatial dimensions, such as reservoirs, gravel-packed layers, ICD completion sections, and horizontal wellbores, is modeled separately. Furthermore, the annular pressures at different locations are used as the solution variable for the coupled solution, which realizes the prediction of oil production, water production, and the water content of gravel packs combined with ICD completion of horizontal wells. The model is used to calculate the effects of different crude oil viscosities, different reservoir permeabilities, different permeabilities of gravel-packed layers, and different development stages on the water control effects of gravel packs combined with ICD completions and conventional ICD completions under field conditions. Full article
Show Figures

Figure 1

21 pages, 5300 KiB  
Article
A Novel Cellular Network Traffic Prediction Algorithm Based on Graph Convolution Neural Networks and Long Short-Term Memory through Extraction of Spatial-Temporal Characteristics
by Geng Chen, Yishan Guo, Qingtian Zeng and Yudong Zhang
Processes 2023, 11(8), 2257; https://doi.org/10.3390/pr11082257 - 26 Jul 2023
Cited by 1 | Viewed by 1455
Abstract
In recent years, cellular communication systems have continued to develop in the direction of intelligence. The demand for cellular networks is increasing as they meet the public’s pursuit of a better life. Accurate prediction of cellular network traffic can help operators avoid wasting [...] Read more.
In recent years, cellular communication systems have continued to develop in the direction of intelligence. The demand for cellular networks is increasing as they meet the public’s pursuit of a better life. Accurate prediction of cellular network traffic can help operators avoid wasting resources and improve management efficiency. Traditional prediction methods can no longer perfectly cope with the highly complex spatiotemporal relationships of the current cellular networks, and prediction methods based on deep learning are constantly growing. In this paper, a spatial-temporal parallel prediction model based on graph convolution combined with long and short-term memory networks (STP-GLN) is proposed to effectively capture spatial-temporal characteristics and to obtain accurate prediction results. STP-GLN is mainly composed of a spatial module and temporal module. Among them, the spatial module designs dynamic graph data based on the principle of spatial distance and spatial correlation. It uses a graph convolutional neural network to learn the spatial characteristics of cellular network graph data. The temporal module uses three time series based on the principle of temporal proximity and temporal periodicity. It uses three long and short-term memory networks to learn the temporal characteristics of three time series of cellular network data. Finally, the results learned from the two modules are fused with different weights to obtain the final prediction results. The mean absolute error (MAE), root mean square error (RMSE), and R-squared (R2) are used as the performance evaluation metrics of the model in this paper. The experimental results show that STP-GLN can more effectively capture the spatiotemporal characteristics of cellular network data; compared with the most advanced model in the comparison model on the real cellular traffic dataset in one cell, the RMSE can be improved about 81.7%, the MAE is improved about 82.7%, and the R2 is improved about 2.2%. Full article
Show Figures

Figure 1

11 pages, 2348 KiB  
Article
Synthetic Minority Oversampling Enhanced FEM for Tool Wear Condition Monitoring
by Yuqing Zhou, Canyang Ye, Deqiang Huang, Bihui Peng, Bintao Sun and Huan Zhang
Processes 2023, 11(6), 1785; https://doi.org/10.3390/pr11061785 - 12 Jun 2023
Viewed by 993
Abstract
Recent advances in artificial intelligence (AI) technology have led to increasing interest in the development of AI-based tool wear condition monitoring methods, heavily relying on large training samples. However, the high cost of tool wear experiment and the uncertainty of tool wear change [...] Read more.
Recent advances in artificial intelligence (AI) technology have led to increasing interest in the development of AI-based tool wear condition monitoring methods, heavily relying on large training samples. However, the high cost of tool wear experiment and the uncertainty of tool wear change in the machining process lead to the problems of sample missing and insufficiency in the model training stage, which seriously affects the identification accuracy of many AI models. In this paper, a novel identification method based on finite-element modeling (FEM) and the synthetic minority oversampling technique (SMOTE) is proposed to overcome the problem of sample missing and sample insufficiency. Firstly, a few tool wear monitoring experiments are carried out to obtain experimental samples with low cost. Then, a FEM model based on the Johnson–Cook constitutive model was established and verified according to the experimental samples. Based on the verified FEM model, the simulated missing sample in the experiments can be supplemented to compose a complete training set. Finally, the SMOTE is employed to expand the sample size to construct a perfect training set to train the SVM classification model. End milling tool wear monitoring experiments demonstrate that the proposed FEM-SMOTE method can obtain 98.7% identification accuracy, which is 30% higher than that based on experimental samples. The proposed method provides an effective approach for tool wear condition monitoring with low experimental cost. Full article
Show Figures

Figure 1

21 pages, 2703 KiB  
Article
Batch Process Modeling with Few-Shot Learning
by Shaowu Gu, Junghui Chen and Lei Xie
Processes 2023, 11(5), 1481; https://doi.org/10.3390/pr11051481 - 12 May 2023
Cited by 1 | Viewed by 968
Abstract
Batch processes in the biopharmaceutical and chemical manufacturing industries often develop new products to meet changing market demands. When the dynamic models of these new products are trained, dynamic modeling with limited data for each product can lead to inaccurate results. One solution [...] Read more.
Batch processes in the biopharmaceutical and chemical manufacturing industries often develop new products to meet changing market demands. When the dynamic models of these new products are trained, dynamic modeling with limited data for each product can lead to inaccurate results. One solution is to extract useful knowledge from past historical production data that can be applied to the product of a new grade. In this way, the model can be built quickly without having to wait for additional modeling data. In this study, a subspace identification combined common feature learning scheme is proposed to quickly learn a model of a new grade. The proposed modified state-space model contains common and special parameter matrices. Past batch data can be used to train common parameter matrices. Then, the parameters can be directly transferred into a new SID model for a new grade of the product. The new SID model can be quickly well trained even though there is a limited batch of data. The effectiveness of the proposed algorithm is demonstrated in a numerical example and a case of an industrial penicillin process. In these cases, the proposed common feature extraction for the SID learning framework can achieve higher performance in the multi-input and multi-output batch process regression problem. Full article
Show Figures

Figure 1

38 pages, 7657 KiB  
Article
Framework for the Implementation of Smart Manufacturing Systems: A Case in Point
by Muhammad Hammad, Md Shamimul Islam, Mohammad Asif Salam, Ali Turab Jafry, Inayat Ali and Wasim Ahmed Khan
Processes 2023, 11(5), 1436; https://doi.org/10.3390/pr11051436 - 9 May 2023
Cited by 1 | Viewed by 2332
Abstract
Smart manufacturing has become a vital technique for increasing productivity and efficiency. Firms are following a smart manufacturing implementation system to compete in the market. Therefore, it is mandatory to find the crucial factors that enable the implementation of intelligent manufacturing in enterprises. [...] Read more.
Smart manufacturing has become a vital technique for increasing productivity and efficiency. Firms are following a smart manufacturing implementation system to compete in the market. Therefore, it is mandatory to find the crucial factors that enable the implementation of intelligent manufacturing in enterprises. This study proposes the framework for a new model factory based on the three-dimensional model that extends the product lifecycle layer. It also analyzes the significant attributes and interdependence relationships of causes and effects through the fuzzy DEMATEL approach for the selected small and medium enterprises discussed as a case study. The results show that the factors in Region 1 are significant attributes that need to be focused on for the development and establishment of small and medium enterprises under consideration. These attributes include design documentation (A11), intelligently management of small and medium enterprises (A3), visualization and monitoring of logistics and production (A6), flow of information, energy, and materials (A12), management platform and data acquisition for equipment (A7), and visualization of quality and process (A5). The sensitivity analysis is also performed to check the results’ validity, reliability, and robustness. This study aids any manufacturing firm in analyzing the critical attributes that contribute to implementing smart manufacturing. Full article
Show Figures

Figure 1

17 pages, 4604 KiB  
Article
Artificial Intelligent Power Forecasting for Wind Farm Based on Multi-Source Data Fusion
by Qingtian Wang, Yunjing Wang, Kegong Zhang, Yaxin Liu, Weiwei Qiang and Qiuzi Han Wen
Processes 2023, 11(5), 1429; https://doi.org/10.3390/pr11051429 - 8 May 2023
Cited by 3 | Viewed by 1605
Abstract
Wind power forecasting is a typical high-dimensional and multi-step time series prediction problem. Data-driven prediction methods using machine learning show advantages over traditional physical or statistical methods, especially for wind farms with complex meteorological conditions. Thus, effective use of different data sources and [...] Read more.
Wind power forecasting is a typical high-dimensional and multi-step time series prediction problem. Data-driven prediction methods using machine learning show advantages over traditional physical or statistical methods, especially for wind farms with complex meteorological conditions. Thus, effective use of different data sources and data types will help improve power forecasting accuracy. In this paper, a multi-source data fusion method is proposed, which integrates the static information of the wind turbine with observational and forecasting meteorological information together to further improve the power forecasting accuracy. Firstly, the characteristics of each time step are re-characterized by using the self-attention mechanism to integrate the global information of multi-source data, and the Res-CNN network is used to fuse multi-source data to improve the prediction ability of input variables. Secondly, static variable encoding and feature selection are carried out, and the time-varying variables are combined with static variables for collaborative feature selection, so as to effectively eliminate redundant information. A forecasting model based on the Encoder–Decoder framework is constructed with LSTM as the basic unit, and the Add&Norm mechanism is introduced to further enhance the input variable information. In addition, the self-attention mechanism is used to integrate the global time information of the decoded results, and the Time Distributed mechanism is used to carry out multi-step prediction. Our training and testing data are obtained from an operating wind farm in northwestern China. Results show that the proposed method outperforms a classic AI forecasting method such as that using the Seq2Seq+attention model in terms of prediction accuracy, thus providing an effective solution for multi-step forecasting of wind power in wind farms. Full article
Show Figures

Figure 1

22 pages, 2412 KiB  
Article
Stability Analysis and Network Topology Optimization of Multi-Agent Systems for Material Transport in Industrial Parks
by Jichao Fei, Linfeng Zhang, Xiaorong Zhu and Yiming Wu
Processes 2023, 11(4), 1276; https://doi.org/10.3390/pr11041276 - 20 Apr 2023
Viewed by 1332
Abstract
Multiple Automated Guided Vehicles promise to be an effective solution for executing tasks such as material transportation and inspection in industrial parks. In particular, system stability is the key to maintaining connectivity among multiple agents. In this paper, we build the stability model [...] Read more.
Multiple Automated Guided Vehicles promise to be an effective solution for executing tasks such as material transportation and inspection in industrial parks. In particular, system stability is the key to maintaining connectivity among multiple agents. In this paper, we build the stability model of multi-agent system (MAS) under the background of swarm robots transporting materials in an industrial park, and propose a network topology optimization method to improve the stability of MAS. In concrete, we first analyze the effect of channel environment on network topology, and the communication delay is analyzed. Then considering the communication delay and artificial potential field, we establish the stability analysis model of MAS and obtain the stability condition of the MAS by using Lyapunov correlation theorem. Finally, we formulate the network topology optimization problem of MAS by maximizing the second smallest eigenvalue of the Laplacian and get the optimal solutions. Analysis and simulation results show the effectiveness of the proposed model and algorithm. Full article
Show Figures

Figure 1

19 pages, 4974 KiB  
Article
GraphSAGE-Based Multi-Path Reliable Routing Algorithm for Wireless Mesh Networks
by Pan Lu, Chuanfang Jing and Xiaorong Zhu
Processes 2023, 11(4), 1255; https://doi.org/10.3390/pr11041255 - 19 Apr 2023
Cited by 3 | Viewed by 1567
Abstract
Wireless mesh networks (WMN) promise to be an effective way to solve the “last mile” access problem on the Internet of Things (IoT) and the key to next-generation wireless networks. The current routing algorithms of WMN are difficult to adapt to complex environments [...] Read more.
Wireless mesh networks (WMN) promise to be an effective way to solve the “last mile” access problem on the Internet of Things (IoT) and the key to next-generation wireless networks. The current routing algorithms of WMN are difficult to adapt to complex environments and guarantee the reliable transmission of services. Therefore, this paper proposes a reliable routing algorithm that combines the improved breadth-first search and a graph neural network, namely GraphSAGE. The algorithm consists of two parts: (1) A multi-path routing algorithm based on the improved breadth-first search. This algorithm can continuously iterate link information based on network topology and output all shortest paths. (2) A GraphSAGE-based performance optimization algorithm. This algorithm creates a method to generate network labels for supervised training of GraphSAGE. Then, the network labels and GraphSAGE are used to learn graph features to obtain the value of network performance for each shortest path. Finally, the path with the best network performance is selected for data transmission. Simulation results show that in the face of complex environments, the proposed algorithm can effectively alleviate network congestion, improve throughput, and reduce end-to-end delay and packet loss rate compared with the traditional shortest-path routing algorithm and the Equal-Cost Multi-Path routing (ECMP). Full article
Show Figures

Figure 1

27 pages, 4305 KiB  
Article
Clustering Approach for the Efficient Solution of Multiscale Stochastic Programming Problems: Application to Energy Hub Design and Operation under Uncertainty
by Mohammed Alkatheri, Falah Alhameli, Alberto Betancourt-Torcat, Ali Almansoori and Ali Elkamel
Processes 2023, 11(4), 1046; https://doi.org/10.3390/pr11041046 - 30 Mar 2023
Cited by 1 | Viewed by 1102
Abstract
The management of the supply chain for enterprise-wide operations generally consists of strategic, tactical, and operational decision stages dependent on one another and affecting various time scales. Their integration usually leads to multiscale models that are computationally intractable. The design and operation of [...] Read more.
The management of the supply chain for enterprise-wide operations generally consists of strategic, tactical, and operational decision stages dependent on one another and affecting various time scales. Their integration usually leads to multiscale models that are computationally intractable. The design and operation of energy hubs faces similar challenges. Renewable energies are challenging to model due to the high level of intermittency and uncertainty. The multiscale (i.e., planning and scheduling) energy hub systems that incorporate renewable energy resources become more challenging to model due to an integration of the multiscale and high level of intermittency associated with renewable energy. In this work, a mixed-integer programming (MILP) superstructure is proposed for clustering shape-based time series data featuring multiple attributes using a multi-objective optimization approach. Additionally, a data-driven statistical method is used to represent the intermittent behavior of uncertain renewable energy data. According to these methods, the design and operation of an energy hub with hydrogen storage was reformulated following a two-stage stochastic modeling technique. The main outcomes of this study are formulating a stochastic energy hub optimization model which comprehensively considers the design and operation planning, energy storage system, and uncertainties of DRERs, and proposing an efficient size reduction approach for large-sized multiple attributes demand data. The case study results show that normal clustering is closer to the optimal case (full scale model) compared with sequence clustering. In addition, there is an improvement in the objective function value using the stochastic approach instead of the deterministic. The present clustering algorithm features many unique characteristics that gives it advantages over other clustering approach and the straightforward statistical approach used to represent intermittent energy, and it can be easily incorporated into various distributed energy systems. Full article
Show Figures

Figure 1

Back to TopTop