BP Neural Network Research Articles

Evaluation of supervised machine learning regression models for CFD-based surrogate modelling in indoor airflow field reconstruction

Fast and reliable prediction of indoor airflow distribution is critical for indoor environment control. While neural networks (NN), often interchangeably referred to as Back Propagation Neural Networks (BPNNs), are popular for airflow predictions, optimizing these models is challenging due to their ”black box” nature and complex network structures. This study explores alternative robust regression models, including decision-tree-based models (e.g., XGBoost, LightGBM, Random Forest) and Support Vector Regression (SVR), for predicting indoor airflow. Two BPNN structures were initially developed to evaluate feasibility of NN models. BPNN A was trained using airflow velocities from two inlets as input neurons to directly predict the airflow velocity distribution within the domain. BPNN B was trained additionally with spatial information, including space samples and boundary wall data. Higher-dimensional training structures of BPNN B were applied to decision tree-based models and SVR to assess their capability in predicting non-linear airflow patterns. Results indicated that BPNN A achieved the highest accuracy, while the inclusion of higher-dimensional data in BPNN B led to decreased accuracy. Among all decision-tree-based models, XGBoost demonstrated the greatest potential, achieving an R2 above 99.5% and predictive errors below 10%. XGBoost also outperformed both BPNN models in speed, being 15.78 times faster than BPNN A and 252 times faster than BPNN B. The interpretability of XGBoost was further explored by analysing feature importance, which helps identify the most influential input variables while predicting the airflow velocity. This analysis is expected to offer an enhanced understanding of boundary conditions leading to optimised indoor environment strategy.

Deformation control of piezoelectric intelligent curved shell structure with multiple actuators based on neural network optimized by genetic algorithm

In aerospace engineering, it is important to achieve precise deformation control of curved shell. In this paper, the piezoelectric intelligent curved shell structure (PICSS) with multiple actuators was selected as the research object. Firstly, a finite element model of the structure was constructed, and the sample data was obtained based on simulation method for the back propagation (BP) neural network. Next, the structure of the BP neural network was determined as ‘6–13-24-3′ with the goal of obtaining the minimum mean square error. Then genetic algorithm was used to optimize the initial weights and thresholds of the BP neural network, and the input voltage of each actuator was predicted. The accuracy of voltage prediction was confirmed through collaborative simulation technology by testing the BP neural network before and after optimization. Finally, the deformation perception and control of the piezoelectric curved shell were verified through the PICSS.

Risk-stratified management of cervical high-grade squamous intraepithelial lesion based on machine learning.

The concordance rate between conization and colposcopy-directed biopsy (CDB) proven cervical high-grade squamous intraepithelial lesion (HSIL) were 64-85%. We aimed to identify the risk factors associated with pathological upgrading or downgrading after conization in patients with cervical HSIL and to provide risk-stratified management based on a machine learning predictive model. This retrospective study included patients who visited the Obstetrics and Gynecology Hospital of Fudan University from January 1 to December 31, 2019, were diagnosed with cervical HSIL by CDB, and subsequently underwent conization. A wide variety of data were collected from the medical records, including demographic data, laboratory findings, colposcopy descriptions, and pathological results. The patients were categorized into three groups according to their postconization pathological results: low-grade squamous intraepithelial lesion (LSIL) or below (downgrading group), HSIL (HSIL group), and cervical cancer (upgrading group). Univariate and multivariate analyses were performed to identify the independent risk factors for pathological changes in patients with cervical HSIL. Machine learning prediction models were established, evaluated, and subsequently verified using external testing data. In total, 1585 patients were included, of whom 65 (4.1%) were upgraded to cervical cancer after conization, 1147 (72.4%) remained having HSIL, and 373 (23.5%) were downgraded to LSIL or below. Multivariate analysis showed a 2% decrease in the incidence of pathological downgrade for each additional year of age and a 1% increase in lesion size. Patients with cytology > LSIL (odds ratio [OR] = 0.33; 95% confidence interval [CI], 0.21-0.52), human papillomavirus (HPV) infection (OR = 0.33; 95% CI, 0.14-0.81), HPV 33 infection (OR = 0.37; 95% CI, 0.18-0.78), coarse punctate vessels on colposcopy examination (OR = 0.14; 95% CI, 0.06-0.32), HSIL lesions in the endocervical canal (OR = 0.48; 95% CI, 0.30-0.76), and HSIL impression (OR = 0.02; 95% CI, 0.01-0.03) were less likely to experience pathological downgrading after conization than their counterparts. The independent risk factors for pathological upgrading to cervical cancer after conization included the following: age (OR = 1.08; 95% CI, 1.04-1.12), HPV 16 infection (OR = 4.07; 95% CI, 1.70-9.78), the presence of coarse punctate vessels during colposcopy examination (OR = 2.21; 95% CI, 1.08-4.50), atypical vessels (OR = 6.87; 95% CI, 2.81-16.83), and HSIL lesions in the endocervical canal (OR = 2.91; 95% CI, 1.46-5.77). Among the six machine learning prediction models, the back propagation (BP) neural network model demonstrated the highest and most uniform predictive performance in the downgrading, HSIL, and upgrading groups, with areas under the curve (AUCs) of 0.90, 0.84, and 0.69; sensitivities of 0.74, 0.84, and 0.42; specificities of 0.90, 0.71, and 0.95; and accuracies of 0.74, 0.84, and 0.95, respectively. In the external testing set, the BP neural network model showed a higher predictive performance than the logistic regression model, with an overall AUC of 0.91. Therefore, a web-based prediction tool was developed in this study. BP neural network prediction model has excellent predictive performance and can be used for the risk stratification of patients with CDB-diagnosed HSIL.

Research on the Optimization of the PID Control Method for an EOD Robotic Manipulator Using the PSO Algorithm for BP Neural Networks

Large-scale explosive ordnance disposal (EOD) robotic manipulators can replace manual EOD tasks, offering higher efficiency and better safety. This study focuses on the control strategies and response speeds of EOD robotic manipulators. Using Adams to establish the dynamic model of an EOD robotic manipulator and constructing a hydraulic system model in AMEsim, a co-simulation model is integrated. This study proposes a PID control strategy optimized by the particle swarm optimization (PSO) algorithm for a backpropagation (BP) neural network and simulates the system’s step response for analysis. To address the vibration issues arising during the manipulator’s motion, B-spline curves are used for trajectory optimization to reduce vibrations. The PSO algorithm optimizes the connection weight matrix of the BP neural network, solving the potential problem of local minima during the training process of the BP neural network, thereby enhancing the global search capability, learning efficiency, and network performance. Simulation results indicate that compared to traditional BP+PID control, genetic algorithm (GA)+PID control, and whale optimization algorithm (WOA)-BP+PID control, the PSO-BP+PID algorithm control rapidly tunes the PID control parameters Kp, Ki, and Kd. Under the same step function conditions, the overshoot is only 1.37%, significantly lower than other methods, and the settling time is only 14 s. After stabilization, there is almost no error, demonstrating faster response speed, higher control accuracy, and stronger robustness. This research has theoretical value and reference significance for the control methods and improvements in EOD robotic manipulators.

Research on Mental Health Assessment and Intervention Methods for College Students based on Big Data Analysis

This study introduces the novel Fuzzy-Enhanced Predictive Neural System for Student Mental Health (FEPS-MH) approach to mental health assessment and intervention for college students. FEPS-MH synergistically combines a Backpropagation (BP) Neural Network with a Deep Fuzzy-Based Neural Network (DFBNN), leveraging the strengths of both systems to handle the complexities of mental health data in the context of big data analytics. The BP Neural Network, known for its effective learning and generalization capabilities, is integrated with the DFBNN to process imprecise, uncertain, or subjective data, typical in mental health assessments. The core objective of FEPS-MH is to provide a more accurate, robust, and sensitive analysis of mental health states, incorporating the nuanced variations and uncertainties inherent in psychological data. This system is designed to analyze a vast array of data sources, including but not limited to, behavioral patterns, self-reported questionnaires, and social media interactions, to identify potential mental health issues among college students. FEPS-MH’s capabilities extend beyond mere assessment; it is also equipped to recommend personalized intervention strategies. Utilizing big data analysis, the system not only predicts potential mental health crises but also suggests tailored intervention approaches based on the unique psychological profile of each student. This study demonstrates the feasibility and effectiveness of FEPS-MH through a series of tests and validations using real-world data. The results indicate a significant improvement in both the accuracy of mental health assessments and the efficacy of suggested interventions. FEPS-MH stands as a promising tool for educational institutions, offering a data-driven, sensitive, and comprehensive approach to student mental health care. Its implementation could revolutionize the field of mental health support in college environments, making it a vital asset for proactive psychological wellness in educational settings.

Prediction of mechanical properties of manufactured sand polymer-modified mortar based on swarm intelligence algorithm

Because polymer-modified mortar (PMM) exhibits a complex and diverse composition, there is a complex nonlinear relationship between its mechanical properties and mix proportions that is challenging for conventional mechanical performance prediction methods to accurately predict. Consequently, in this study, a predictive model utilizing a backpropagation neural network (BPNN) was formulated, featuring a structure of 6–14-2. Simultaneously, three swarm intelligence algorithms were integrated—the ant colony optimization algorithm, grey wolf optimization algorithm, and bat optimization algorithm (BAT)—to collectively refine the optimization process of the BPNN prediction model. The model's input layer included cement (OPC), cellulose ether (CE), dispersible polymer powder (DPP), antifoam (AF), fly ash (FA), and tuff stone powder (SP), and the output layer consisted of the compressive and bond strengths. The model dataset comprised 520 samples (260 × 2), with 60 % (312) used for model establishment and 40 % (208) used for validation. Correlation matrix and principal component analyses were conducted on the dataset, along with a comparative analysis of the factors influencing the mechanical performance evaluation indicators. The results indicate that at 7 and 28 d, there was a positive correlation between the DPP and AF with the development of PMM mechanical properties. At 7 d, the SP and FA were negatively correlated with the compressive and flexural strength, whereas the CE was positively correlated with the bond strength. At 28 d, the OPC was negatively correlated with the compressive, bond, and flexural strengths, and positively correlated with the SP and FA. C3 represents the optimal mix proportion for the PMM, and considering the influence of all raw materials, F3 was identified as the comprehensive optimal mix proportion. The predictive performance evaluation indicators of the BAT–BPNN for the compressive and bond strengths were R2 = 0.980 and 0.942, MAE = 5.967 and 1.958, MAPE = 0.071 and 0.041, and RMSE = 4.686 and 1.594, respectively. BAT significantly improves the predictive accuracy of the BPNN model, enabling the prediction of PMM mechanical properties and providing a scientific basis for its mix proportion design.

A heavy-duty tracked vehicle model with a reduced feasible domain for motion tracking control considering dynamic characters of hybrid powertrain

Motion tracking control is an essential part of heavy-duty unmanned tracked vehicles to realize autonomous mobility, which adjusts the vehicle motion in real-time by controlling the driving torques. Achieving accurate motion tracking strongly depends on the use of a precise vehicle dynamic model. To support the motion tracking ability, a vehicle dynamic model with a reduced feasible domain is proposed for heavy-duty unmanned tracked vehicles. Firstly, a vehicle dynamic model is established based on the vehicle kinematics and dynamics. Building upon the established vehicle dynamic model, a graph neural network (GNN) is employed to reduce the feasible domain of the vehicle dynamic model, specifically focusing on the limitation of the powertrain. The powertrain components and their interconnections are abstracted into a graph, which is analyzed to study the interactions between the components using the GNN. Finally, the proposed vehicle dynamic model with a reduced feasible domain is verified through actual heavy-duty tracked vehicle experiments in various cases. Compared with the back propagation neural network, the evaluation error of the driving torque decreased by 25.86%. Results show that the proposed vehicle dynamic model with a reduced feasible domain is closer to the performance of the actual heavy-duty tracked vehicle and their powertrain. And the vehicle dynamic model accurately reduces the solution time of the motion tracking control, resulting in improved accuracy and efficiency of the motion tracking control.

Reverse design for mixture proportions of recycled brick aggregate concrete using machine learning-based meta-heuristic algorithm: A multi-objective driven study

Construction and Demolition Wastes (CDW) have a significant impact on global waste streams. Brick waste stands out as a prominent type of CDW, and numerous studies have explored its recycling for the creation of environmentally-friendly concrete. Reverse design of recycled brick aggregate concrete (RBAC) mixture proportion is presented in this paper with a focus on four key objectives, that is: compressive strength, cost, and environmental elements (i.e., energy consumption and carbon emission). Based on compiled experimental datasets of 374 samples, the back propagation neural network (BP), random forest (RF), and four meta-heuristic algorithm optimization models were constructed to achieve the desired compressive strength objective. In all machine learning (ML) methods, the compressive strength of RBAC can be predicted with high accuracy, with the SSA-BP (optimized back propagation neural network model using the sparrow search algorithm) model achieving superior results (i.e., NSE=0.91, RPD=3.2). The SSA-BP is therefore used as the objective function for compressive strength. The economic objective is primarily influenced by material costs, and the objective functions of energy consumption and carbon emission are determined by various aspects of production, transportation, and their mixing processes. In order to obtain the optimal RBAC design, the Non-Dominated Sorting Genetic Algorithm (NSGA-III) was implemented considering imperative constraints. Results indicate that cement amount and recycled brick aggregate (RBA)-to-natural aggregate proportion have a positive impact on the compressive strength. The suggested design framework allows for the creation of RBAC composite designs with varying levels of RBA substitution rates and strength targets, providing valuable guidance for tackling the CDW challenge and optimizing RBA usage.

A Novel Strategy Coupling Optimised Sampling with Heterogeneous Ensemble Machine-Learning to Predict Landslide Susceptibility

The accuracy of data-driven landslide susceptibility prediction depends heavily on the quality of non-landslide samples and the selection of machine-learning algorithms. Current methods rely on artificial prior knowledge to obtain negative samples from landslide-free regions or outside the landslide buffer zones randomly and quickly but often ignore the reliability of non-landslide samples, which will pose a serious risk of including potential landslides and lead to erroneous outcomes in training data. Furthermore, diverse machine-learning models exhibit distinct classification capabilities, and applying a single model can readily result in over-fitting of the dataset and introduce potential uncertainties in predictions. To address these problems, taking Chenxi County, a hilly and mountainous area in southern China, as an example, this research proposes a strategy-coupling optimised sampling with heterogeneous ensemble machine learning to enhance the accuracy of landslide susceptibility prediction. Initially, 21 landslide impact factors were derived from five aspects: geology, hydrology, topography, meteorology, human activities, and geographical environment. Then, these factors were screened through a correlation analysis and collinearity diagnosis. Afterwards, an optimised sampling (OS) method was utilised to select negative samples by fusing the reliability of non-landslide samples and certainty factor values on the basis of the environmental similarity and statistical model. Subsequently, the adopted non-landslide samples and historical landslides were combined to create machine-learning datasets. Finally, baseline models (support vector machine, random forest, and back propagation neural network) and the stacking ensemble model were employed to predict susceptibility. The findings indicated that the OS method, considering the reliability of non-landslide samples, achieved higher-quality negative samples than currently widely used sampling methods. The stacking ensemble machine-learning model outperformed those three baseline models. Notably, the accuracy of the hybrid OS–Stacking model is most promising, up to 97.1%. The integrated strategy significantly improves the prediction of landslide susceptibility and makes it reliable and effective for assessing regional geohazard risk.

A combination of XGBoost and neural network in LIBS spectrum processing for precise determination of critical elements in 620 iron ore samples of various origins

China is the worldwide largest importer of iron ores for supplying its iron and steel industries. The necessary inspections of iron ores not only concern total iron content but also involve minor elements, such as silicium, aluminum, magnesium, and calcium. Laser-induced breakdown spectroscopy (LIBS) promises in situ and on-site detection and analysis capabilities, which is required for efficient, reliable, and large-scale impartation inspections. Such opportunity represents at the same time, a challenge for the technique in terms of the quantitative analysis ability. A fundamental issue corresponds to the matrix effect in LIBS analysis due to a large diversity of iron ores in terms of mineralogical and chemical compositions. Demonstrated as effective to deal with the matrix effect in LIBS analyses of geological materials, the machine learning approach needs to be implemented within an optimized data processing pipeline, combining algorithms carefully chosen and tested. In this work, we developed a sequentially connected combination of an extreme gradient boosting (XGBoost) model and a back propagation neural network (BPNN) model, to effectively fit the functional relation between the concentrations and the spectra. The specificity of the approach consists in fitting the linear, low-order nonlinear, and high-order nonlinear terms in a sequential way to substantially improve the model prediction performances. The method has been developed and tested with a significant set of 620 collected iron ore samples of eight production countries and 35 different commercial brands, with a wide range of matrices in terms of mineralogical and chemical compositions. The samples underwent prior characterization using conventional laboratory analysis methods to establish the reference elemental concentrations. The results from this necessary step of laboratory investigation, with root mean square error for prediction (RMSEP) of 0.407 wt%, 0.372 wt%, 0.085 wt%, 0.131 wt%, and 0.114 wt%, respectively for total iron (TFe), SiO2, Al2O3, MgO, and CaO, show the potential for on-site inspections.

Rotational Risley prisms: Fast and high-accuracy inverse solution and application based on back propagation neural networks

In the application of rotational Risley prism systems, there is a conflict between inverse solution time and accuracy. This paper proposes an inverse solution method based on back propagation neural networks to solve this problem. The training set data for prism rotation, azimuth, and elevation angles are produced based on the beam deflection model of the actual application. After that, two independent networks are trained individually. One network is used to calculate the angular difference, while the other network calculates the synchronous rotation angle to satisfy the solutions of elevation and azimuth angles of target. The azimuth and elevation root mean square errors for the inverse solution of the spiral trajectory using the trained network were 8.04e-04 arcsec and 2.60e-3 arcsec, respectively. At a solution time of 80 µs, the pointing accuracy in the experiment was less than 1 arcsec. The experimental results demonstrate that the proposed inverse solution method can obtain high-accuracy analytical solutions with low time consumption.

A robust approach for bond strength prediction of mortar using machine learning with SHAP interpretability

Application of machine learning (ML) in predicting mortars bond strength contributes to low experimental cost and high accuracy. This study explores the performance of four ML models—Back Propagation Neural Network (BPNN), Support Vector Regression (SVR), Random Forest (RF), and eXtreme Gradient Boosting (XGBoost)—using 304 experimental data points for training and 76 for testing. Among these, the XGBoost model demonstrated the highest predictive accuracy, with R², MAE, and RMSE values of 0.95, 0.06, and 0.13 in the training set, and 0.94, 0.08, and 0.16 in the testing set, respectively. Compared with traditional models, ML approaches, particularly XGBoost, were more effective in providing reliable predictions of bond strength. SHapley Additive exPlanation (SHAP) show that substrate type, ethylene-vinyl acetate (EVA) content and water to binder ratio W/C have the most remarkable effect on bond strength, while the type of cement and admixture shows the least effect. These findings highlight the potential of ML models in optimizing mortar formulations and improving prediction reliability.

Improved black-winged kite algorithm and finite element analysis for robot parallel gripper design

This paper presents a comprehensive study on the design optimization of a robotic gripper, focusing on both the gripper modeling and the optimization of its parallel mechanism structure. This study integrates the Black-winged Kite Algorithm (BKA), Finite Element Analysis (FEA), Backpropagation Neural Network (BPNN), and response surface optimization techniques. The Good Point Set (GPS), nonlinear convergence factor, and adaptive t-distribution method improve BKA, which enhances exploration and exploitation performance, convergence speed, and solution quality. Subsequently, the parallel mechanism structure is designed to minimize the total mass, total deformation, and maximum equivalent stress. The central composite design (CCD) method was used to design the FEA experiment and establish the BKA-BPNN regression prediction model. The RMSE of this model’s training set and test set are 0.001615 and 0.0029328. A response surface optimization model is constructed to determine the best design solution. The optimized design achieves a 33.12% reduction in maximum equivalent stress, a 1.47% decrease in total mass, and a 0.16% reduction in maximum total deformation. This study provides valuable insights into the design optimization process for robotic grippers, showcasing the effectiveness of the proposed methodologies in enhancing performance while reducing mass and improving structural integrity.

Hydrogen-Bonded Organic Frameworks for Antibiotic Fluorescent Sensing Artificial Intelligence-Enhanced Anticounterfeiting.

The paramount importance of anticounterfeiting measures in safeguarding consumers from counterfeit products lies in their ability to ensure product safety and reliability. Advanced luminescent anticounterfeiting materials, particularly those responsive to multiple stimuli, afford a dynamic and multilayered security assurance. This study presents the synthesis of a novel material, Eu/Tb@GC-3, via postsynthetic modification, which exhibits notable photoluminescent properties with emission at 544 and 614 nm. The material demonstrates high selectivity and sensitivity in detecting Nitrofural and Enrofloxacin, with limits of detection at 0.0122 and 0.0280 μM, respectively. Furthermore, multistimulus responsive luminescent fibers and inks were developed, facilitating intelligent anticounterfeiting labels. The integration of these labels with back-propagation neural networks (BPNNs) significantly enhances pattern recognition and authentication capabilities, providing an efficacious strategy to combat counterfeit products and ensure consumer safety.

Optimization of thermal management performance of direct-cooled power battery based on backpropagation neural network and deep reinforcement learning

The batteries of new energy vehicles generate a large amount of heat when discharged at large multiplication rates, which can cause safety hazards if the heat is not eliminated in time. In this paper, a Swiss roll-type cooling belt is used to improve the thermal performance of lithium-ion battery packs. This paper selects six key parameters as optimization parameters, including the structure of Swiss roll-type cooling belt, coolant type, inlet coolant mass flow rate, inlet coolant gas volume fraction, battery discharge rate, and ambient temperature. Use neural network fitting and deep reinforcement learning to optimize these six parameters in order to achieve optimized cooling performance. The results show that after the cold plate has been structurally optimised and the inlet parameters have been optimised by deep reinforcement learning, the thermal performance has been significantly improved. Not only does the cold plate inlet require less mass flow, but the battery pack also performs better in terms of temperature difference.

A highly accurate and robust prediction framework for drilling rate of penetration based on machine learning ensemble algorithm

The rate of penetration (ROP) is a key indicator of drilling efficiency. Many researchers have explored the application of machine learning in ROP prediction. However, few studies have focused on the robustness of the constructed models, and developing a ROP prediction model that can achieve both high accuracy and strong robustness remains a challenge. This paper introduces a novel machine learning approach to constructing a ROP prediction model through ensemble learning algorithms. The model is based on field data from oilfields, incorporating 16 input parameters that influence ROP. First, the feasibility of the collected dataset is verified using correlation analysis. Then, ROP prediction models are developed based on various machine learning algorithms, including Decision Tree Regression (DTR), Random Forest (RF), eXtreme Gradient Boosting (XGB), Light Gradient Boosting Machine (LGBM), Support Vector Regression (SVR), and Backpropagation Neural Network (BPNN). By comparing the performance of these models under noise levels of 0%, 1.7%, and 5.1%, RF, LGBM, XGB, and SVR are selected as base learners. These base learners are then combined to construct multiple ensemble models, and the performance of the optimal ensemble model is evaluated under varying noise levels. The results show that the prediction error of the optimal model remains within 10%, and R2 is greater than 0.96. Finally, the Shapley Additive Explanations (SHAP) method is applied to perform interpretability analysis on the optimal ROP prediction model, examining the impact of different input factors on the model's predictive performance. Compared to single models and other ensemble models, the proposed ensemble model not only achieves higher accuracy but also demonstrates strong robustness and generalization capability.

Study on the response mechanisms and evolution prediction of groundwater microbial-toxicological indicators.

This study aims to investigate the response mechanisms of groundwater microbial-toxicological indicators, specifically total bacteria count (TBC) and total coliform count (TCC), to water quality indicators and environmental conditions. Using data from a water source in the western plateau of China, a predictive model focusing on TBC and TCC was developed. An orthogonal experimental design was employed to manipulate environmental conditions such as temperature, pH, and porosity, facilitating laboratory experiments. These experiments measured pH, chemical oxygen demand (COD), oxidation-reduction potential (ORP), TBC, and TCC at varying depths and environmental conditions. Principal component analysis elucidated the mechanisms by which water quality indicators and environmental conditions affect groundwater microbial-toxicological indicators. A prediction model for these indicators in plateau regions was established based on a backpropagation neural network (BP-NN), using TBC and TCC as target variables and the newly extracted principal components as influencing factors. The results demonstrate that environmental conditions and water quality indicators primarily influence the evolution of groundwater microbial-toxicological indicators by altering the ionic charge quantities, redox conditions, and temperature of the groundwater. The predictive model for groundwater microbial-toxicological indicators shows trends consistent with experimental outcomes, with an average relative error of less than 15%, meeting engineering requirements. PRACTITIONER POINTS: The values of total bacteria count (TBC) and total coliform count (TCC) under different conditions were obtained by column experiments. The influence mechanism of environmental conditions and groundwater indicators on TBC and TCC was elaborated by principal component analysis. TBC and TCC prediction models were established through the investigation of water sources in a plateau area and laboratory experiments.

Development and characterization of carboxylated copper oxide conjugated polymeric nanocomposites and correlating with computational techniques

The quality and effectiveness of composite-based components are influenced by their mechanical characteristics. Hence, a structured approach is essential to identify the key factors that contribute to achieving superior mechanical characteristics. This study investigates the utilization of an artificial neural network (ANN) to predict the mechanical characteristics, explicitly density, and hardness, of carboxyl copper oxide nanoparticle (CCONP)/DL-lactide and glycolide copolymer (PDLG) nanocomposites fabricated through microwave-assisted sintering. The research explores the effects of numerous microwave sintering conditions on the microstructure and mechanical properties of the nanocomposites. Physical characterization techniques including FTIR and TEM analysis are employed to confirm physical interactions and nanocomposite size. A back-propagation neural network with a 2–10-2 architecture was developed to predict density and hardness. The network was trained using the Levenberg-Marquardt algorithm and the Scaled Conjugate Gradient method. The predicted values are compared with experimental results, showing good agreement with correlation coefficients (R) of 0.883 for density and 0.9737 for hardness. This demonstrates the effectiveness of the ANN approach in evaluating the mechanical properties of CCONP/PDLG nanocomposites, providing a reliable tool for decision-making and potentially reducing experimental characterization costs.

Estimating water quality parameters of freshwater aquaculture ponds using UAV-based multispectral images

UAV imaging technology has become one of the means to quickly monitor water quality parameters in freshwater aquaculture ponds. The change of sunlight during a long flight affects the quality of UAV images, which will reduce the accuracy of monitoring water quality. This study aims to propose a method to correct spectral variation during UAV imaging and apply it to detect dissolved organic matter (DOM) concentration and dissolved oxygen (DO) content in freshwater aquaculture ponds. Firstly, a spectral correction method was used to transform UAV-based multispectral images. The spectral data before and after correction was extracted. Secondly, 18 spectral indices before and after correction were constructed. The optimal combination of indices was identified using correlation analysis algorithm. The estimation models of water quality parameters were then constructed and compared using the Random Forest (RF), Support Vector Regression (SVR), and BP neural network (BP) methods. The results showed that the accuracy of estimating DOM concentration using corrected spectral indices was significantly improved compared to pre-correction models, with the highest improvement of 38 % (SVR), the lowest of 23 % (BP), and an average improvement of 31 %. The RF model performed best, achieving R² = 0.81, RMSE = 3.34 mg/L, and MAE = 2.17 mg/L. For DO content estimation, the accuracy of models using corrected spectral indices was also improved significantly, with the highest improvement of 97 % (RF), the lowest of 39 % (SVR), and an average improvement rate of 67 %. The Random Forest model was again optimal, with R² = 0.69, RMSE = 1.97 mg/L, and MAE = 1.47 mg/L. This study indicates that the proposed spectral correction method helps to map the concentration of DOM and DO in freshwater aquaculture ponds with high accuracy using UAV-based multispectral images.

Co-seismic landslide susceptibility mapping for the Luding earthquake area based on heterogeneous ensemble machine learning models

ABSTRACT Considering the complexity of landslide susceptibility mapping (LSM) at the municipal-scale, it is difficult for the assumption space of a single machine learning (ML) algorithm to fully and accurately reflect the intrinsic correlation between the landslide influencing factors and the actual catastrophic events under all scenarios. This study addresses this by integrating three heterogeneous ensemble learning (HEL) algorithms(Bagging, Blending, and Stacking) with four ML models (Gradient Boosting Decision Tree (GBDT), Extreme Gradient Boosting (XGBoost), K-Nearest Neighbors (KNN), and Backpropagation-Neural Network (BPNN)), to construct 27 HEL LSM models, and to realize optimization and selection of the models and algorithms through the comparative analysis of the results of the LSM and the accuracy of the prediction of the individual models. The final results indicate that the Bagging and Blending HEL algorithms fail to enhance model prediction accuracy effectively. In contrast, the Stacking HEL algorithm significantly improves model prediction accuracy (with an accuracy increase of over 1.4% and AUC values above 0.95). The Stacking (GBDT-XGBoost) model had the highest prediction accuracy (AUC = 0.964) and thus was selected as the optimal LSM model. This study's modeling and mapping results are intended to serve as a reference for geohazard prevention and sustainability studies in related cities.