Neural Network Ensemble Model Research Articles

Synergistic ternary deep eutectic solvents: An archetype for sustainable and eco-conscious Li and Co recovery from spent batteries

The development of sustainable materials for energy is a critical challenge in the pursuit of a circular economy. This study presents an innovative approach for recycling lithium-ion batteries (LIBs) using a novel, sustainable deep eutectic solvent (DES) composed of glycine, ascorbic acid, and water. This natural, chloride-free ternary DES leverages the biocompatible and eco-friendly properties of its constituents, achieving exceptional leaching efficiency of 99.1 % for lithium and 97.9 % for cobalt under optimized conditions. A multifaceted methodology integrating advanced factorial design, kinetic modeling, and machine learning was employed to refine the process parameters and enhance leaching efficiency. The extraction kinetics were predominantly governed by endothermic surface reactions, characterized by activation energies of 81.5 kJ/mol for Co and 110.9 kJ/mol for Li. An ensemble neural network model accurately predicted the extraction with an R2 > 0.95, demonstrating the robustness of the computational approach. The DES exhibited excellent recyclability over five cycles, thereby emphasizing its potential for practical industrial applications in the field of sustainable materials management. This approach facilitated the selective recovery of high-purity lithium oxalate and cobalt oxalate, underscoring the efficiency and environmental benefits of the process. DFT calculations provided insights into the extraction mechanism, revealing the crucial role of hydrogen bonding and synergistic effects of the DES components. This pioneering study not only advances LIB recycling but also establishes a comprehensive, quantitatively validated framework, paving the way for future innovations in sustainable materials management and supporting global efforts towards a circular economy in energy storage.

A 20-year (1998–2017) global sea surface dimethyl sulfide gridded dataset with daily resolution

Abstract. The oceanic emission of dimethyl sulfide (DMS) plays a vital role in the Earth's climate system and constitutes a substantial source of uncertainty when evaluating aerosol radiative forcing. Currently, the widely used monthly climatology of sea surface DMS concentration falls short of meeting the requirement for accurately simulating DMS-derived aerosols with chemical transport models. Hence, there is an urgent need for a high-resolution, multi-year global sea surface DMS dataset. Here we develop an artificial neural network ensemble model that uses nine environmental factors as input features and captures the variability of the DMS concentration across different oceanic regions well. Subsequently, a global sea surface DMS concentration and flux dataset (1° × 1°) with daily resolution spanning from 1998 to 2017 is established. According to this dataset, the global annual average concentration was ∼ 1.71 nM, and the annual total emissions were ∼ 17.2 Tg S yr−1, with ∼ 60 % originating from the Southern Hemisphere. While overall seasonal variations are consistent with previous DMS climatologies, notable differences exist in regional-scale spatial distributions. The new dataset enables further investigations into daily and decadal variations. Throughout the period 1998–2017, the global annual average concentration exhibited a slight decrease, while total emissions showed no significant trend. The DMS flux from our dataset showed a stronger correlation with the observed atmospheric methanesulfonic acid concentration compared to those from previous monthly climatologies. Therefore, it can serve as an improved emission inventory of oceanic DMS and has the potential to enhance the simulation of DMS-derived aerosols and associated radiative effects. The new DMS gridded products are available at https://doi.org/10.5281/zenodo.11879900 (Zhou et al., 2024).

Investigating the rheological characteristics of alkali-activated concrete using contemporary artificial intelligence approaches

Abstract Using artificial intelligence-based tools, this research aims to establish a direct correlation between the alkali-activated concrete (AAC) mix design factors and their performances. More specifically, the machine learning system was fed new property data obtained from AAC mixes used in laboratory experiments. The rheological parameters (yield stress [static/dynamic] and plastic viscosity) of AAC were predicted using the multilayer perceptron neural network (MLPNN) and bagging ensemble (BE) models. In addition, the R 2 values, k-fold analyses, statistical checks, and the dissimilarity between the experimental and predicted compressive strength were employed to assess the performance of the created models. Also, the SHapley additive exPlanation (SHAP) approach was used for examining the relevance of influencing parameters. The BE approach was found to be significantly accurate in all prediction models, with R 2 greater than 0.90, and MLPNN models were found to be moderately precise, with R 2 slightly below 0.90. However, the error assessment through statistical checks and k-fold analysis also validated the higher precision of BE models over the MLPNN models. Building models that can calculate rheological properties of AAC for different values of input parameters could save a lot of time and money compared to doing the tests in a laboratory. In order to ascertain the required amounts of raw materials of AAC, investigators, as well as businesses, may find the SHAP study helpful.

Can eXplainable AI Offer a New Perspective for Groundwater Recharge Estimation?—Global‐Scale Modeling Using Neural Network

AbstractDue to the difficulties in estimating groundwater recharge and cross‐boundary nature of many aquifers, estimating groundwater recharge at large scale has been called upon. Process‐based models as well as data‐driven models have been established to meet this need. Meanwhile, with the advent of explainable artificial intelligence (XAI) methods, data‐driven machine learning models can take advantage of enhanced explainability while keeping the strength of high flexibility. In this study, an ensemble neural network model was built to check the suitability of the model to predict groundwater recharge and the possibility to gain new insights from large data set. Recent large inputs of groundwater recharge data and additional input for the Arabian Peninsula collated in this study were fed to the model with multiple predictors related to climatology considering seasonality, soil and plant characteristics, topography, and hydrogeology. The model showed higher performance (adjusted R2: 0.702, RMSE: 193.35 mm yr−1) than a recent global process‐based model in predicting groundwater recharge. Using XAI methods as individual conditional expectations and Shapley Additive Explanation interaction values, the model behavior was analyzed and possible linear and non‐linear relationships between the predictors and the groundwater recharge rate were found. Long‐term averaged precipitation and enhanced vegetation index showed non‐linear relationships with groundwater recharge rate, while slope, compound topographic index, and water table depth showed low importance to the model results. Most model behaviors followed the domain knowledge, while multi‐correlation between predictors and data skewness hindered the model from learning.

Prediction of multi-stage froth flotation efficiency of complex lead–zinc sulfide ore using an integrated ensemble neural network–random forest model

Despite the long history of flotation in the mineral processing industry, its prediction and understanding remains a great challenge owing to its many variables acting in a complex manner. In this study, we introduced machine learning (ML) models to predict the grade and yield of a multi-stage flotation process of a complex lead–zinc sulfide ore. Over 100 batch flotation tests were conducted in a stepwise manner to characterize different rougher-cleaner-scavenger configurations. Performing a pre-flotation of talc prior to sulfide flotation remarkably improved the grade of lead concentrate. The experimental data were divided into four subsets for the ML models: rougher without pre-flotation, rougher with pre-flotation, cleaner, and cleaner-scavenger datasets. Different ML models were evaluated to determine whether they could predict the lead grade, zinc grade, and yield related to key flotation parameters, including particle size, reagent dosage, and pulp pH. The integrated ensemble neural network and random forest model yielded the best prediction results with R2 values of 0.924, 0.902, 0.973, and 0.894 for the rougher without pre-flotation, rougher with pre-flotation, cleaner, and cleaner-scavenger subsets, respectively. The developed ML model, with the connection of subset models, effectively predicted the flotation outcome of the rougher-cleaner-scavenger circuit, demonstrating a better prediction performance than previous methods. This indicates that the developed ML model can potentially predict flotation process performance and evaluate the efficiency of newly designed multi-stage froth flotation processes.

Bootstrap sampling style ensemble neural network for inverse design of optical nanoantennas

The process of machine learning-assisted nanophotonicsinverse design has been plagued by the problem of non-uniqueness for a long time, which is a problem worth studying. In this paper, we present a novel methodology for the design of bowtie optical nanoantennas (BONAs) by employing a Bootstrap Sampling Style Ensemble Neural Network (BSENN) model. Our approach combines a bagging algorithm with a tandem neural network to address the non-uniqueness challenge inherent in the inverse design process of BONAs. By splitting the data, training in batches, and integrating the results, our BSENN model is able to provide reliable predictions and offer a solution to the non-uniqueness problem. The main objective of our work is to explore diverse BONAs design structures that yield identical spectral responses, thereby providing a broader range of alternatives for the design of optical nanoantennas. Through the utilization of the BSENN model, we aim to enhance the design process and offer increased flexibility and versatility in the field of optical nanoantenna design.

An ensemble neural network model for predicting the energy utility in individual houses

With the rapid increase in the world's human population and the advancement of technology, energy utilization has substantially increased. To sustain a stable stream of energy, it is vital to anticipate power utilization beforehand. This issue makes predicting short-term electrical time series use difficult. In this research, the user proposed a neural network with a deep learning prototype that performs energy prediction by combining. Long Short-Term Memory (LSTM) and Feed Forward Neural Networks (FFNN). In order to enhance the performance, the ensemble FFNN-LSTM network and the Stationary Wavelet Transform (SWT) method are combined to create a hybrid deep learning algorithm. The SWT's reduction of instability and expansion of data dimensionality may improve the FFNN-LSTM forecasting accuracy. The suggested technique's predicting effectiveness is even further improved by the ensemble Long short-term neural network. On the basis of a real-world household energy usage database gathered by the "UK-DALE" project, validation studies were carried out. Researchers determined that the LSTM-FFNN based SWT produced a better result in terms of accuracy and energy utilization than other existing results after contrasting the experimental findings to the baseline, which is the conventional LSTM and Moving Average (MA) based upon the Root Mean Squared Error (RMSE) score.

(Invited) Electrochemical Interfaces in Energy Storage: Theory Meets Experiment

Understanding the dynamic processes at solid-liquid interfaces in electrochemical devices like batteries is key to developing more efficient and durable technologies for the green transition. Fundamental and performance-limiting interfacial processes like the formation of the Solid-Electrolyte Interphase (SEI) [1,2] and dendritic growth [3] span numerous time- and length scales. Despite decades of research, the fundamental understanding of structure-property relations remains elusive. Ab initio molecular dynamics (AIMD) generally provides sufficient accuracy to describe chemical reactions and the making and breaking of chemical bonds at these interfaces [4]. Still, the cost is prohibitively high to reach sufficiently long time- and length scales to ensure proper statistical sampling [5]. Machine learning (ML) potentials offer a potential solution to this challenge. Still, training ML-based potentials capable of handling activated processes in organic or aqueous electrolytes remains a fundamental challenge since the potential must capture both intra- and intermolecular interactions in the electrolyte and during chemical reactions at the interface [5]. Here, we present new approaches using phase field models [3], graph neural networks [6] and new transition state training sets [7] for chemical reaction networks, and machine/deep learning models to predict the spatio-temporal evolution of electrochemical interphases [2]. We also discuss the development of methods like symbolic regression to learn the laws of electrolyte transport [8], uncertainty quantification training and evaluating neural network ensemble models [9] using models trained on multi-sourced and multi-fidelity data from multiscale computer simulations, operando characterization, high-throughput synthesis, and testing, to provide, e.g., uncertainty-aware and explainable ML for early prediction of battery degradation trajectories (Figure 1) [10]. Figure 1

ENN: Hierarchical Image Classification Ensemble Neural Network for Large-Scale Automated Detection of Potential Design Infringement

This paper presents a two-stage hierarchical neural network using image classification and object detection algorithms as key building blocks for a system that automatically detects a potential design right infringement. This neural network is trained to return the Top-N original design right records that highly resemble the input image of a counterfeit. This work proposes an ensemble neural network (ENN), an artificial neural network model that aims to deal with a large amount of counterfeit data and design right records that are frequently added and deleted. First, we performed image classification and objection detection learning per design right using acclaimed existing models with high accuracy. The distributed models form the backbone of the ENN and yield intermediate results aggregated at a master neural network. This master neural network is a deep residual network paired with a fully connected network. This ensemble layer is trained to determine the sub-models that return the best result for a given input image of a product. In the final stage, the ENN model multiplies the inferred similarity coefficients to the weighted input vectors produced by the individual sub-models to assess the similarity between the test input image and the existing product design rights to see any sign of violation. Given 84 design rights and the sample product images taken meticulously under various conditions, our ENN model achieved average Top-1 and Top-3 accuracies of 98.409% and 99.460%, respectively. Upon introducing new design rights data, a partial update of the inference model was performed an order of magnitude faster than the single model. The ENN maintained a high level of accuracy as it was scaled out to handle more design rights. Therefore, the ENN model is expected to offer practical help to the inspectors in the field, such as customs at the border that deal with a swarm of products.

Deep Learning-Based Positioning Error Fault Diagnosis of Wire Bonding Equipment and an Empirical Study for IC Packaging

Little research has been done for artificial intelligence applications of semiconductor backend. This study aims to develop a deep learning based fault diagnosis framework as prognostics and health management (PHM) solutions with a comprehensive analytics process. A linear encoder sensor is employed to measure position, while data preparation is conducted to convert position information into a signal. Then, signal processing is used to extract the features, in which high pass filter is applied to normalize the signal and emphasize the features. High-dimensional features including time domain statistical features, time-frequency domain features obtained by continuous wavelet transform (CWT), and frequency domain features converted by fast Fourier transform (FFT) are extracted to prepare the dataset. An ensemble neural network model that integrates deep neural network (DNN) and convolutional neural network (CNN) is employed to recognize the pattern and diagnose the positioning errors. The accuracy and false alarm rate are considered to support maintenance decisions. An empirical study was conducted in a world leading IC assembly and testing company for validation. The results have shown that the proposed approach can effectively detect inappropriate installation of the bond head in wire bonding equipment for predictive maintenance and cost reduction.

Forest fire detection in aerial vehicle videos using a deep ensemble neural network model

PurposeThe purpose of this paper is to present a deep ensemble neural network model for the detection of forest fires in aerial vehicle videos.Design/methodology/approachPresented deep ensemble models include four convolutional neural networks (CNNs): a faster region-based CNN (Faster R-CNN), a simple one-stage object detector (RetinaNet) and two different versions of the you only look once (Yolo) models. The presented method generates its output by fusing the outputs of these different deep learning (DL) models.FindingsThe presented fusing approach significantly improves the detection accuracy of fire incidents in the input data.Research limitations/implicationsThe computational complexity of the proposed method which is based on combining four different DL models is relatively higher than that of using each of these models individually. On the other hand, however, the performance of the proposed approach is considerably higher than that of any of the four DL models.Practical implicationsThe simulation results show that using an ensemble model is quite useful for the precise detection of forest fires in real time through aerial vehicle videos or images.Social implicationsBy this method, forest fires can be detected more efficiently and precisely. Because forests are crucial breathing resources of the earth and a shelter for many living creatures, the social impact of the method can be considered to be very high.Originality/valueThis study fuses the outputs of different DL models into an ensemble model. Hence, the ensemble model provides more potent and beneficial results than any of the single models.

Neural network models for situation similarity assessment in hybrid-CBR

The case-based reasoning method has a high potential for solving tasks of intelligence decision-support. To implement it, it is necessary to solve the problem of comparing situations and selecting the one that is most similar to the current situation in the knowledge base. The problem arises in the case of heterogeneous objects and situations with many different types of parameters and their possible uncertainty. In this paper, an approach based on machine (deep) learning is investigated for this task. It is proposed to carry out the process of selecting situations and solutions from the knowledge base in two stages: recognition of the states of the elements of a complex object and the relationships between them, then the formation of a representation of the situation in the state space and its use for comparing situations using neural networks. An ensemble neural network model based on a multi-layer network is proposed. It successfully simulates the cognitive functions of a human (expert), correctly selects similar situations and ranks them according to the similarity parameter. Proposed neural network models provide the implementation of a hybrid-CBR approach for decision-making on complex objects.

통합적인 인공 신경망 모델을 이용한 발틱운임지수 예측

통합적인 인공 신경망 모델을 이용한 발틱운임지수 예측

A Cost Sensitive SVM and Neural Network Ensemble Model for Breast Cancer Classification

A Cost Sensitive SVM and Neural Network Ensemble Model for Breast Cancer Classification

Residual stress prediction of arc welded austenitic pipes with artificial neural network ensemble using experimental data

The prediction of weld-induced residual stress assists the structural integrity assessment of welded structures. In this study, residual stress measurements of girth welded austenitic stainless-steel pipes were used to develop two Artificial Neural Network (ANN) ensemble models to predict through-thickness residual stress profiles in Weld Centre Line (WCL). One model was developed for axial and the other for hoop residual stress prediction. The inputs of the models were the pipe radius to thickness ratio, the thickness, the heat input (weld arc electrical energy per unit run length [kJ/mm]) and the normalised through-thickness position. The hyperparameters were tuned, and the models were trained with various initial weight vectors, creating an ensemble of ANNs. The models’ performance was assessed by a test set and by sensitivity studies which revealed the models’ output trends.

Development of Deep Convolutional Neural Network Ensemble Models for 36-Month ENSO Forecasts

Development of Deep Convolutional Neural Network Ensemble Models for 36-Month ENSO Forecasts

A deep learning based two-layer predictor to identify enhancers and their strength.

The enhancer is a DNA sequence that can increase the activity of promoters and thus speed up the frequency of gene transcription. The enhancer plays an essential role in activating gene expression. Currently, gene sequencing technology has been developed for 30years from the first generation to the third generation, and a variety of biological sequence data have increased significantly every year. Due to the importance of enhancer functions, it is very expensive to identify enhancers through biochemical experiments. Therefore, we need to study new methods for the identification and classification of enhancers. Based on the K-mer principle this study proposed a feature extraction method that others have not used in convolutional neural networks. Then, we combined it with one-hot encoding to build an efficient one-dimensional convolutional neural network ensemble model for predicting enhancers and their strengths. Finally, we used five commonly used classification problem evaluation indicators to compare with the models proposed by other researchers. The model proposed in this paper has a better performance by using the same independent test dataset as other models.

A knowledge transfer enhanced ensemble approach to predict the shear capacity of reinforced concrete deep beams without stirrups

AbstractThis paper proposes a novel learning algorithm, the transfer ensemble neural network (TENN) model, to increase the performance of shear capacity predictions on small datasets, illuminating the usefulness of advanced machine learning techniques in general. By incorporating ensemble learning and transfer learning, the TENN model is designed to control the high variability inherent in machine learning models trained on small amounts of data. The novel TENN model is validated to predict the shear capacity of deep reinforced concrete (RC) beams without stirrups across varying data availability levels. Knowledge acquired through pretraining a model on slender RC beams is utilized for training a model to better predict the shear capacity of deep RC beams without stirrups. To evaluate the performance of the TENN model, three baseline models are developed and examined across multiple data availability levels. The novel TENN model outperforms the baseline models, particularly when trained on a very limited dataset. Furthermore, the proposed algorithm achieves a higher accuracy than the currently accepted design standards in accurately predicting deep RC beams' shear capacity and demonstrates the capabilities of the TENN model to extrapolate in other domains where large‐scale or physical testing is cost‐prohibitive.

Evolutionary design of machine-learning-predicted bulk metallic glasses.

The size of composition space means even coarse grid-based searches for interesting alloys are infeasible unless heavily constrained, which requires prior knowledge and reduces the possibility of making novel discoveries. Genetic algorithms provide a practical alternative to brute-force searching, by rapidly homing in on fruitful regions and discarding others. Here, we apply the genetic operators of competition, recombination, and mutation to a population of trial alloy compositions, with the goal of evolving towards candidates with excellent glass-forming ability, as predicted by an ensemble neural-network model. Optimization focuses on the maximum casting diameter of a fully glassy rod, D max, the width of the supercooled region, ΔT x, and the price-per-kilogramme, to identify commercially viable novel glass-formers. The genetic algorithm is also applied with specific constraints, to identify novel aluminium-based and copper-zirconium-based glass-forming alloys, and to optimize existing zirconium-based alloys.

Deep Learning Prediction Model for Heart Disease for Elderly Patients

The detection of heart disease is a problematic task in medical research. This diagnosis utilizes a thorough analysis of the clinical tests from the patient’s medical history. The massive advances in deep learning models pursue the development of intelligent computerized systems that aid medical professionals to detect the disease type with the internet of things support. Therefore, in this paper, we propose a deep learning model for elderly patients to aid and enhance the diagnosis of heart disease. The proposed model utilizes a deeper neural architecture with multiple perceptron layers with regularization learning techniques. The model performance is verified with a full and minimum set of features. Fewer features enhance the processing time of the classification process while the accuracy is compromised. The performance of classifiers with less features has been analyzed with experimental results. The proposed system is built on the Internet of Things Platform for medical data for the classification process which aids medical professionals to detect heart diseases through cloud platforms. The results accuracy is matched to classical learning models such as Convolutional Neural Network (CNN), Deep CNN, and neural ensemble models. The analysis of the proposed diagnostic system can determine the heart disease risks efficiently. Experimental results demonstrate that flexible modeling and tuning of the hyperparameters can attain an accuracy of up to 97.11%.