Data-Mining for Processes in Chemistry, Materials, and Engineering

Machine Learning Driven Atom-Thin Materials for Fragrance Sensing

Fragrance plays a crucial role in the daily lives. Its importance spans various sectors, from therapeutic purposes to personal care, making the understanding and accurate identification of fragrances essential. To fully harness the potential of fragrances, efficient and precise fragrance sensing and identification are necessary. However, current fragrance sensors face several limitations, particularly in detecting and differentiating complex scent profiles with high accuracy. To address these challenges, the use of atom-thin materials in fragrance sensors has emerged as a groundbreaking approach. These atom-thin sensors, characterized by their enhanced sensitivity and selectivity, offer significant improvements over traditional sensing technology. Moreover, the integration of Machine Learning (ML) into fragrance sensing has opened new opportunities in the field. ML algorithms applied to fragrance sensing facilitate advancements in four key domains: accurate fragrance identification, precise discrimination between different fragrances, improved detection thresholds for subtle scents, and prediction of fragrance properties. This comprehensive review delves into the synergistic use of atom-thin materials and ML in fragrance sensing, providing an in-depth analysis of how these technologies are revolutionizing the field, offering insights into their current applications and future potential in enhancing the understanding and utilization of fragrances.

Ensemble machine learning using hydrometeorological information to improve modeling of quality parameter of raw water supplying treatment plants

Source and raw water quality may deteriorate due to rainfall and river flow events that occur in watersheds. The effects on raw water quality are normally detected in drinking water treatment plants (DWTPs) with a time-lag after these events in the watersheds. Early warning systems (EWSs) in DWTPs require models with high accuracy in order to anticipate changes in raw water quality parameters. Ensemble machine learning (EML) techniques have recently been used for water quality modeling to improve accuracy and decrease variance in the outcomes. We used three decision-tree-based EML models (random forest [RF], gradient boosting [GB], and eXtreme Gradient Boosting [XGB]) to predict two critical parameters for DWTPs, raw water Turbidity and UV absorbance (UV254), using rainfall and river flow time series as predictors. When modeling raw water turbidity, the three EML models (rRF-Tu2=0.87, rGB-Tu2=0.80 and rXGB-Tu2=0.81) showed very good performance metrics. For raw water UV254, the three models (rRF-UV2=0.89, rGB-UV2=0.85 and rXGB-UV2=0.88) again showed very good performance metrics. Results from this study suggest that EML approaches could be used in EWSs to anticipate changes in the quality parameters of raw water and enhance decision-making in DWTPs.

Optimization of 2024-T3 Aluminum Alloy Friction Stir Welding Using Random Forest, XGBoost, and MLP Machine Learning Techniques

This study optimized friction stir welding (FSW) parameters for 1.6 mm thick 2024T3 aluminum alloy sheets. A 3 × 3 factorial design was employed to explore tool rotation speeds (1100 to 1300 rpm) and welding speeds (140 to 180 mm/min). Static tensile tests revealed the joints' maximum strength at 87% relative to the base material. Hyperparameter optimization was conducted for machine learning (ML) models, including random forest and XGBoost, and multilayer perceptron artificial neural network (MLP-ANN) models, using grid search. Welding parameter optimization and extrapolation were then carried out, with final strength predictions analyzed using response surface methodology (RSM). The ML models achieved over 98% accuracy in parameter regression, demonstrating significant effectiveness in FSW process enhancement. Experimentally validated, optimized parameters resulted in an FSW joint efficiency of 93% relative to the base material. This outcome highlights the critical role of advanced analytical techniques in improving welding quality and efficiency.

The clean energy development path and sustainable development of the ecological environment driven by big data for mining projects

Clean energy is urgently needed to realize mining projects' sustainable development (SD). This study aims to discuss the clean energy development path and the related issues of SD in the ecological environment driven by big data for mining projects. This study adopts a comprehensive research approach, including a literature review, case analysis, and model construction. Firstly, an in-depth literature review of the development status of clean energy is carried out, and the existing research results and technology applications are explored. Secondly, some typical mining projects are selected as cases to discuss the practice and effect of their clean energy application. Finally, the corresponding clean energy development path and the SD analysis model of the ecological environment are constructed based on big data technology to evaluate the feasibility and potential benefits of promoting and applying clean energy in mining projects. (1) It is observed that under different Gross Domestic Product (GDP) growth rates, the new and cumulative installed capacities of wind energy show an increasing trend. In 2022, under the low GDP growth rate, the cumulative installed capacity of global wind energy was 370.60 Gigawatt (GW), and the new installed capacity was 45 GW. With the high GDP growth rate, the cumulative and new installed capacities were 367.83 GW and 46 GW. As the economy grows, new wind energy capacity is expected to increase significantly by 2030. In 2046, 2047, and 2050, carbon dioxide (CO2) emissions reductions are projected to be 8183.35, 8539.22, and 9842.73 Million tons (Mt) (low scenario), 8750.68, 9087.16, and 10,468.75 Mt (medium scenario), and 9083.03, 9458.86, and 10,879.58 Mt (high scenario). By 2060, it is expected that CO2 emissions reduction will continue to increase. (2) The proposed clean energy development path model has achieved a good effect. Through this study, it is hoped to provide empirical support and decision-making reference for the development of mining projects in clean energy, and promote the SD of the mining industry, thus achieving a win-win situation of economic and ecological benefits. This is of great significance for protecting the ecological environment and realizing the sustainable utilization of resources.

Temporal continuous monitoring of cyanobacterial blooms in Lake Taihu at an hourly scale using machine learning

Cyanobacterial blooms in most lakes exhibit extraordinary changes in time and space. Herein, a cyanobacterial prediction model was designed for Lake Taihu based on a machine learning method. This method can generate temporally continuous (24 moments throughout the day) cyanobacterial data at a fine spatial scale of 9 km. The hourly meteorological data for 24 moments of the day were obtained from ERA5-Land data. Areal coverage of cyanobacterial blooms was derived from the hourly Geostationary Ocean Color Imager reflectance data observed only eight times a day (from ~8:00 to ~15:00, UTC+8). The cyanobacterial and meteorological data of eight moments in Lake Taihu from 2011 to 2020 were used to design the prediction model. The results were compared and validated employing nine training strategies to determine the best cyanobacterial prediction model for Lake Taihu (R = 0.42; root mean square error = 0.10). With the best-fitted model utilizing meteorological data (2011-2020), the area coverage of cyanobacterial blooms at the other 16 moments during a day were estimated. Based on this, the regional and temporal characteristics of diurnal bloom variation were evaluated at an hourly scale. The results indicated that the hourly variations in the areal coverage of cyanobacterial blooms at 24 moments of the day had similar patterns in each subregion of Lake Taihu with minor seasonal variations. The six meteorological variables adopted to construct the model had similar diurnal changes but with diverse value ranges among the seasons. Further analysis revealed that three meteorological variables (temperature, surface pressure, and evaporation) were positively related to diurnal bloom variations at an hourly scale. Overall, these results illustrate that meteorological conditions can affect the occurrence of cyanobacterial blooms at multiple time scales (e.g., hourly, daily, or monthly). The developed cyanobacterial prediction model can provide cyanobacterial data when cyanobacterial data is unavailable for the target waterbody.

A field study to estimate heavy metal concentrations in a soil-rice system: Application of graph neural networks

Accurate prediction of the concentration of heavy metals is of great significance for assessing the quality of agricultural products and reducing health risks. However, the complexity and interconnectivity of the farmland ecosystem restricts the improvement of the prediction accuracy of traditional methods. This research explored the application potential of graph neural network (GNN) technology, which can extract and learn information in large-scale networks in detail, in the field of heavy metal prediction for the first time. In this study, a heavy metal prediction model for rice, CoNet-GNN, was proposed with 17 environmental factors as input variables using the co-occurrence network and GNN. Experimental results using a dataset from a field study showed that the R² of CoNet-GNN for predicting Cd, Pb, Cr, As, and Hg had outstanding values of 0.872, 0.711, 0.683, 0.489, and 0.824, respectively. Sensitivity analysis further indicated that CoNet-GNN had good stability and robustness. Compared with random forest, gradient boosting, and multilayer perceptron, CoNet-GNN made a remarkable improvement to the prediction accuracy of all studied heavy metals. Therefore, CoNet-GNN can effectively simulate the rich relationships and laws between various factors in the soil-rice system and effectively characterize the influence diffusion path. Furthermore, it provides new ideas for heavy metal prediction based on network research methods and expands the technical scope of heavy metal evaluation.

Predictive Modeling of Compression Strength of Waste PET/SCM Blended Cementitious Grout Using Gene Expression Programming

The central aim of this study is to evaluate the effect of polyethylene terephthalate (PET) alongside two supplementary cementitious materials (SCMs)—i.e., fly ash (FA) and silica fume (SF)—on the 28-day compressive strength (CS28d) of cementitious grouts by using. For the gene expression programming (GEP) approach, a total of 156 samples were prepared in the laboratory using variable percentages of PET and SCM (0−10%, each). To achieve the best hyper parameter setting of the optimized GEP model, 10 trials were undertaken by varying the genetic parameters while observing the models’ performance in terms of statistical indices, i.e., correlation coefficient (R), root mean squared error (RMSE), mean absolute error (MAE), comparison of regression slopes, and predicted to experimental ratios (ρ). Sensitivity analysis and parametric study were performed on the best GEP model (obtained at; chromosomes = 50, head size = 9, and genes = 3) to evaluate the effect of contributing input parameters. The sensitivity analysis showed that: CS7d (30.47%) > CS1d (28.89%) > SCM (18.88%) > Flow (18.53%) > PET (3.23%). The finally selected GEP model exhibited optimal statistical indices (R = 0.977 and 0.975, RMSE = 2.423 and 2.531, MAE = 1.918 and 2.055) for training and validation datasets, respectively. The role of PET/SCM has no negative influence on the CS28d of cementitious grouts, which renders the PET a suitable alternative toward achieving sustainable and green concrete. Hence, the simple mathematical expression of GEP is efficacious, which leads to saving time and reducing labor costs of testing in civil engineering projects.

Intelligent Natural Gas and Hydrogen Pipeline Dispatching Using the Coupled Thermodynamics-Informed Neural Network and Compressor Boolean Neural Network

Natural gas pipelines have attracted increasing attention in the energy industry thanks to the current demand for green energy and the advantages of pipeline transportation. A novel deep learning method is proposed in this paper, using a coupled network structure incorporating the thermodynamics-informed neural network and the compressor Boolean neural network, to incorporate both functions of pipeline transportation safety check and energy supply predictions. The deep learning model is uniformed for the coupled network structure, and the prediction efficiency and accuracy are validated by a number of numerical tests simulating various engineering scenarios, including hydrogen gas pipelines. The trained model can provide dispatchers with suggestions about the number of phases existing during the transportation as an index showing safety, while the effects of operation temperature, pressure and compositional purity are investigated to suggest the optimized productions.

Combining Machine Learning and Computational Chemistry for Predictive Insights Into Chemical Systems

Machine learning models are poised to make a transformative impact on chemical sciences by dramatically accelerating computational algorithms and amplifying insights available from computational chemistry methods. However, achieving this requires a confluence and coaction of expertise in computer science and physical sciences. This Review is written for new and experienced researchers working at the intersection of both fields. We first provide concise tutorials of computational chemistry and machine learning methods, showing how insights involving both can be achieved. We follow with a critical review of noteworthy applications that demonstrate how computational chemistry and machine learning can be used together to provide insightful (and useful) predictions in molecular and materials modeling, retrosyntheses, catalysis, and drug design.

A recurrent neural network using historical data to predict time series indoor PM2.5 concentrations for residential buildings

Due to the severe outdoor PM2.5 pollution in China, many people have installed air-cleaning systems in homes. To make the systems run automatically and intelligently, we developed a recurrent neural network (RNN) that uses historical data to predict the future indoor PM2.5 concentration. The RNN architecture includes an autoencoder and a recurrent part. We used data measured in an apartment over the course of an entire year to train and test the RNN. The data include indoor/outdoor PM2.5 concentration, environmental parameters and time of day. By comparing three different input strategies, we found that a strategy employing historical PM2.5 and time of day as inputs performed best. With this strategy, the model can be applied to predict the relatively stable trend of indoor PM2.5 concentration in advance. When the input length is 2 h and the prediction horizon is 30 min, the median prediction error is 8.3 µg/m³ for the whole test set. For times with indoor PM2.5 concentrations between (20,50] µg/m³ and (50,100] µg/m³ , the median prediction error is 8.3 and 9.2 µg/m³ , respectively. The low prediction error between the ground-truth and predicted values shows that the RNN can predict indoor PM2.5 concentrations with satisfactory performance.

Predictive modeling of swell-strength of expansive soils using artificial intelligence approaches: ANN, ANFIS and GEP

This study presents the development of new empirical prediction models to evaluate swell pressure and unconfined compression strength of expansive soils (P_sUCS-ES) using three soft computing methods, namely artificial neural networks (ANNs), adaptive neuro fuzzy inference system (ANFIS), and gene expression programming (GEP). An extensive database comprising 168 P_s and 145 UCS records was established after a comprehensive literature search. The nine most influential and easily determined geotechnical parameters were taken as the predictor variables. The network was trained and tested, and the predictions of the proposed models were compared with the observed results. The performance of all the models was tested using mean absolute error (MAE), root squared error (RSE), root mean square error (RMSE), Nash-Sutcliffe efficiency (NSE), correlation coefficient (R), regression coefficient (R²) and relative root mean square error (RRMSE). The sensitivity analysis indicated that the increasing order of inputs importance in case of P_s followed the order: maximum dry density MDD (30.5%) > optimum moisture content OMC (28.7%) > swell percent SP (28.1%) > clay fraction CF (9.4%) > plasticity index PI (3.2%) > specific gravity G_s (0.1%), whereas, in case of UCS it followed the order: sand (44%) > PI (26.3%) > MDD (16.8%) > silt (6.8%) > CF (3%) > SP (2.9%) > G_s (0.2%) > OMC (0.03%). Parametric analysis was also performed and the resulting trends were found to be in line with findings of past literature. The comparison results reflected that GEP and ANN are efficacious and reliable techniques for estimation of P_sUCS-ES. The derived mathematical GP-based equations portray the novelty of GEP model and are comparatively simple and reliable. The R_overall values for P_sUCS-ES followed the order: ANN > GEP > ANFIS, with all values lying above the acceptable range of 0.80. Hence, all the proposed AI approaches exhibit superior performance, possess high generalization and prediction capability, and evaluate the relative importance of the input parameters in predicting the P_sUCS-ES. The GEP model outperformed the other two models in terms of closeness of training, validation and testing data set with the ideal fit (1:1) slope. Evidently the findings of this study can help researchers, designers and practitioners to readily evaluate the swell-strength characteristics of the widespread expansive soils thus curtailing their environmental vulnerabilities which leads to faster, safer and sustainable construction from the standpoint of environment friendly waste management.

Artificial Neural Networks in Modeling of Dewaterability of Sewage Sludge

Mechanical dewatering is a key process in the management of sewage sludge. However, the drainage efficiency depends on a number of factors, from the type and dose of the conditioning agent to the parameters of the drainage device. The selection of appropriate methods and parameters of conditioning and dewatering of sewage sludge is the task of laboratory work. This work can be accelerated through the use of artificial neural network (ANNs). The paper discusses the possibilities of using ANNs in predicting the dewatering efficiency of physically conditioned sludge. The input variables were only four parameters characterizing the conditioning methods and the dewatering method by centrifugation. These were the dose of the sludge skeleton builders (cement, gypsum, fly ash, and liquid glass), sonication parameters (sonication amplitude and time), and relative centrifugal force. Dewatering efficiency parameters such as sludge hydration and separation factor were output variables. Due to the nature of the research problem, two nonlinear networks were selected: a multilayer perceptron and a radial neural network. Based on the results of the prediction of artificial neural networks, it was found that these networks can be used to forecast the effectiveness of municipal sludge dewatering. The prediction error did not exceed 1.0% of the real value. ANN can therefore be useful in optimizing the dewatering process. In the case of the conducted research, it was the optimization of the sludge dewatering efficiency as a function of the type and parameters of conditioning factors. Therefore, it is possible to predict the dewatering efficiency of sludge that has not been tested in the laboratory, for example, with the use of other doses of physical conditioner. However, the condition for correct prediction and optimization was the use of a large dataset in the neural network training process.

Performance Analysis of Selected Programming Languages in the Context of Supporting Decision-Making Processes for Industry 4.0

This study analyzes the possibility of using Go (Golang) in the context of Java and Python in decision-making processes, with particular emphasis on their use in industry-specific solutions for Industry 4.0. The authors intentionally compared Go with Java and Python, which have been widely used for many years for data analysis in many areas. The research work was based on decision trees data mining algorithms, and especially on classification trees, in which the measure of entropy as a heuristics to choose an attribute was taken into account. The tests were carried out on various parameters describing calculation time, RAM usage, and CPU usage. The source data, which were the basis for the computing of the decision tree algorithm implemented using these three languages, were obtained from a commercial remote prototyping system and were related to the target customers’ choice of methods and means of the full design-creation process.

Assessing the biochemical oxygen demand using neural networks and ensemble tree approaches in South Korea

The biochemical oxygen demand (BOD), one of widely utilized variables for water quality assessment, is metric for the ecological division in rivers. Since the traditional approach to predict BOD is time-consuming and inaccurate due to inconstancies in microbial multiplicity, alternative methods have been recommended for more accurate prediction of BOD. This study investigated the capability of a novel deep learning-based model, Deep Echo State Network (Deep ESN), for predicting BOD, based on various water quality variables, at Gongreung and Gyeongan stations, South Korea. The model was compared with the Extreme Learning Machine (ELM) and two ensemble tree models comprising the Gradient Boosting Regression Tree (GBRT) and Random Forests (RF). Diverse water quality variables (i.e., BOD, potential of Hydrogen (pH), electrical conductivity (EC), dissolved oxygen (DO), water temperature (WT), chemical oxygen demand (COD), suspended solids (SS), total nitrogen (T-N), and total phosphorus (T-P)) were utilized for developing the Deep ESN, ELM, GBRT, and RF with five input combinations (i.e., Categories 1-5). These models were evaluated by root mean square error (RMSE), Nash-Sutcliffe efficiency (NSE), coefficient of determination (R²), and correlation coefficient (R). Overall evaluations suggested that the Deep ESN5 model provided the most reliable predictions of BOD among all the models at both stations.

Assessment of Supply Chain Agility to Foster Sustainability: Fuzzy-DSS for a Saudi Manufacturing Organization

Supply chain agility and sustainability is an essential element for the long-term survival and success of a manufacturing organization. Agility is an organization’s ability to respond rapidly to customers’ dynamic demands and volatile market changes. In a dynamic business environment, manufacturing firms demand agility to be evaluated to support any alarming decision. Sustainability is an aspect to sustain collaboration, value creation, and survival of firms under a dynamic competitive business scenario. Agility is a capability that drives competitiveness to foster sustainability aspects. The purpose of this article is to consider and evaluate the supply chain behavior within the context of Saudi enterprises. The efficacy and relevance of this model were explored through a case study conducted in a Saudi dairy manufacturing corporation. Owing to the complexity and a large number of calculations that are required for evaluating the agility of the supply chain, a decision support system was proposed as a tool to assess the supply chain and identifying barriers to a strategic sustainable solution for a specific organizational target. The decision support system is extensive as it contains six separate agility enablers and ninety-three agility attributes for the supply chain. The assessment was carried out using a fuzzy multi-criteria method. It combines the performance rating and importance weight of each agile supply chain-enabler-attribute. To achieve and sustain local and global success, the case organization strove to become a major local and global manufacturer to satisfy its customers, reduce its time to market, lower its total ownership costs, and boost its overall competitiveness through improving its agility across supply chain activities to foster sustainability for a manufacturing organization located in Saudi Arabia.

Special Issue on “Process Modelling and Simulation”

Collecting and highlighting novel developments that address existing as well as forthcoming challenges in the field of process modelling and simulation was the motivation for proposing this special issue on “Process Modelling and Simulation” in the journal Processes [...]