Propagation Artificial Neural Network Model Research Articles

The suspension stability of nanoplastics in aquatic environments revealed using meta-analysis and machine learning

Nanoplastics (NPs) aggregation determines their bioavailability and risks in natural aquatic environments, which is driven by multiple environmental and polymer factors. The back propagation artificial neural network (BP-ANN) model in machine learning (R2 = 0.814) can fit the complex NPs aggregation, and the feature importance was in the order of surface charge of NPs > dissolved organic matter (DOM) > functional group of NPs > ionic strength and pH > concentration of NPs. Meta-analysis results specified low surface charge (0 ≤ |ζ| < 10 mV) of NPs, low concentration (< 1 mg/L) and low molecular weight (< 10 kg/mol) of DOM, NPs with amino groups, high ionic strength (IS > 700 mM) and acidic solution, and high concentration (≥ 20 mg/L) of NPs with smaller size (< 100 nm) contribute to NPs aggregation, which is consistent with the prediction in machine learning. Feature interaction synergistically (e.g., DOM and pH) or antagonistically (e.g., DOM and cation potential) changed NPs aggregation. Therefore, NPs were predicted to aggregate in the dry period and estuary of Poyang Lake. Research on aggregation of NPs with different particle size,shapes, and functional groups, heteroaggregation of NPs with coexisting particles and aging effects should be strengthened in the future. This study supports better assessments of the NPs fate and risks in environments.

Formation and prediction of heterocyclic amines and N-nitrosamines in smoked sausages using back propagation artificial neural network.

Heterocyclic amines (HAs) and N-nitrosamines (NAs) are formed easily during the thermal processing of food, and epidemiological studies have demonstrated that consuming HAs and NAs increases the risk of cancer. However, there are few studies on the application of back propagation artificial neural network (BP-ANN) models to simultaneously predict the content of HAs and NAs in sausages. This study aimed to investigate the effects of cooking time and temperature, smoking time and temperature, and fat-to-lean ratio on the formation of HAs and NAs in smoked sausages, and to predict their total content based on the BP-ANN model. With an increase in processing time, processing temperature and fat ratio, the content of HAs and NAs in smoked sausages increased significantly, while the content of HA precursors and nitrite residues decreased significantly. The optimal network topology of the BP-ANN model was 5-11-2, the correlation coefficient values for training, validation, testing and all datasets were 0.99228, 0.99785, 0.99520 and 0.99369, respectively, and the mean squared error value of the best validation performance was 0.11326. The bias factor and the accuracy factor were within acceptable limits, and the predicted values approximated the true values, indicating that the model has good predictive performance. The contents of HAs and NAs in smoked sausages were significantly influenced by the cooking conditions, smoking conditions and fat ratio. The BP-ANN model has high application value in predicting the contents of HAs and NAs in sausages, which provides a theoretical basis for the suppression of carcinogen formation. © 2024 Society of Chemical Industry.

Optimization of mix proportion and strength prediction of magnesium phosphate cement-based composites based on machine learning

In order to optimize the mix ratio design of magnesium phosphate cementitious composites (MPCC) to obtain excellent workability and strength, this study investigated the effects of MgO-phosphate ratio, water-binder ratio, aggregate-binder ratio, the mass ratio of coarse and fine aggregates, mineral admixture content and steel fiber content on the flowability, setting time and cubic compressive strength (CCS) of MPCC using orthogonal design of experiments with a total of 32 sets of mix ratios. The experimental results were also analyzed utilizing range and variance analysis to determine the main influencing parameters and the optimal dosage of each factor, and the optimal MPCC mix ratio was obtained. In addition, the prediction of CCS for MPCC using five machine learning models was first introduced, and the predicted results were interpreted and analyzed regarding the SHAP (Shapley Additive exPlanations) features importance. The results showed that the effect of steel fiber volume content and water-binder ratio on the flowability was significant, and M/P significantly affected the setting time of MPCC. The CCS of MPCC increased with the age of curing, and MPCC exhibited high early strength, with 1-day CCS accounting for 55%～85% of the 28-day CCS. The effect of M/P on the 1-day, 3-day and 28-day CCS was highly significant, with M/P contribution rate as high as 76.3%, 80.3% and 83.6% of the CCS at 1, 3 and 28 days, respectively. Besides, the effect of steel fiber volume content on 28-day CCS was also significant. In summary, the optimal mix ratio of MPCC was A4B1C4D1E4F4G1 (M/P = 3, w/b=0.11, a/b=1.4, c/f=3, FA content =15%, MK content= 15%, steel fiber volume content= 0%). Furthermore, random forest (RF) and particle swarm optimization of back propagation artificial neural network (PSO-BP-ANN) models have a higher prediction accuracy, and the results of SHAP feature significance further confirmed the above test results.

Prediction of biological nutrients removal in full-scale wastewater treatment plants using H2O automated machine learning and back propagation artificial neural network model: Optimization and comparison

The effective control of total nitrogen (ETN) and total phosphorus (ETP) in effluent is challenging for wastewater treatment plants (WWTPs). In this work, automated machine learning (AutoML) (mean square error = 0.4200 ∼ 3.8245, R2 = 0.5699 ∼ 0.6219) and back propagation artificial neural network (BPANN) model (mean square error = 0.0012 ∼ 6.9067, R2 = 0.4326 ∼ 0.8908) were used to predict and analyze biological nutrients removal in full-scale WWTPs. Interestingly, BPANN model presented high prediction performance and general applicability for WWTPs with different biological treatment units. However, the AutoML candidate models were more interpretable, and the results showed that electricity carbon emission dominated the prediction. Meanwhile, increasing data volume and types of WWTP hardly affected the interpretable results, demonstrating its wide applicability. This study demonstrated the validity and the specific advantages of predicting ETN and ETP using H2O AutoML and BPANN model, which provided guidance on the prediction and improvement of biological nutrients removal in WWTPs.

Terminal attitude selection method of missile attack aircraft

The appropriate terminal attitude of missile is important to improve the damage effectiveness of fragmentation warhead against aircraft targets. In this paper, three missile terminal attitude selection methods are proposed to solve the problem of terminal attitude selection in different situations according to their respective evaluation indexes. The MVE-based method uses the damage probability of each detonation point around the aircraft at a given attitude of missile to calculate the Mean Volume of Effectiveness (MVE) at the corresponding attitude, and then uses the MVE for different attitudes to select the terminal strike attitude. The detonation position-based method addresses the case where the missile detonation position can be assessed in advance. Given the effects of missile guidance errors and fuze activation position errors, Monte Carlo simulations are used to calculate the damage probability of aircraft with different strike attitudes, from which the terminal strike attitude is selected. The BP-ANN model-based method uses the constructed Back Propagation Artificial Neural Network (BP-ANN) model instead of simulation to calculate the evaluation indexes in the corresponding cases of MVE-based method and Detonation position-based method, which can improve the efficiency of attitude selection. Simulations are conducted for different scenarios to verify the feasibility and effectiveness of the proposed method.

Modeling and optimization of Graphene Oxide (GO) membranes for nanofiltration with artificial neural networks

It is important to understand the correlation between the nanostructure and membrane performances (water flux and rejection) in membrane separation technology, which is helpful to develop novel membranes. In this study, a back propagation artificial neural network model optimized with genetic algorithms was proposed to predict the nanofiltration performances (water flux and rejection) of GO membranes. The aim is to explore the feature importance of the membrane structure parameters (such as interlayer spacing, Zeta potential, water contact angle, roughness and thickness) and the operating condition (operation pressure) on the membrane properties, so as to provide clues for improving the membrane performances. The obtained results showed that genetic algorithm-back propagation artificial neural network (GABPANN) exhibited more accurate performance to predict the correlation between the parameters (including membrane structure and operation pressure) and the membrane performances according to the published experimental results. Moreover, the GABPANN results indicated that the water contact angle is the most powerful parameter determining the water flux, while the surface charge is the most powerful one determining the rejection of GO membranes. This work provided a novel strategy to efficiently optimize the nanofiltration performance of GO membranes, beneficial for better understanding and controlling the structural design of GO membranes.

Global incidence, prevalence and disease burden of silicosis: 30 years’ overview and forecasted trends

BackgroundGlobally, silicosis accounts for 90% of all pneumoconiosis cases and is a serious public health issue. It is characterized by progressive inflammation and irreversible pulmonary fibrosis. A comprehensive analysis at temporal, spatial and population levels with the most updated data from GBD 2019 is provided in this study to estimate the disease burden of silicosis from 1990 to 2019 and make predictions to 2029.MethodsWe delineated silicosis data on incidence, prevalence, and disability-adjusted life years (DALYs) as well as age-standardized rates (ASRs) across 30 years from GBD 2019. Joinpoint regression analysis was employed to detect temporal changes and estimate annual percentage change (APC) of each trend segment. Measures were stratified by time, location, age, and sociodemographic index (SDI). Back propagation artificial neural network (BP-ANN) model was applied to elaborate ASR trends from 1990 to 2019 and projections to the next 10 years.ResultsGlobally, silicosis incident, prevalent cases, and DALYs increased by 64.6%, 91.4%, and 20.8%, respectively. However, all the corresponding ASRs showed overall downward trends with an estimated average APC (AAPC) of -0.5(-0.7 to -0.3), -0.2(-0.5 to 0.0), and − 2.0(-2.2 to -1.8), respectively. Middle and high-middle SDI regions carried the heaviest disease burden. The highest disease burden of silicosis was mainly transferred to the older from 1990 to 2019. The trend of ASRs demonstrated a rapid decline between 2005 and 2019, followed by a continuous decline until 2029.ConclusionThough disease burden of silicosis has been on a decline in general from 1990 to 2019, which shows a promising prospect but cannot be ignored. We should pay more attention to implementing preventive tactics and improving the life quality of present sufferers.

Revealing and quantifying the effect of cattiite coprecipitation on the purity of K-struvite in aqueous solution

To investigate the effect of cattiite (Mg3(PO4)2·22 H2O) coprecipitate with K-struvite on the purity of K-struvite, K-struvite crystallization in aqueous solution was performed at different pHs and the initial Mg2+ concentrations. The coprecipitation was quantified by Simultaneous Thermal Analyzer, Scanning Electron Microscopy with energy-dispersive X-ray spectroscopy as well as Inductively Coupled Plasma Mass Spectrometer. Genetic algorithm-back propagation artificial neural network model (GA-BPANN) was also innovatively employed to simulate and predict K-struvite purity. The results revealed that cattiite coprecipitated with K-struvite rather than bobierrite (Mg3(PO4)2·8 H2O) in the solution. The low initial Mg2+ concentration and high pH were more conducive to K-struvite crystallization, thereby inhibiting the cattiite co-precipitation with K-struvite. Specifically, K-struvite crystals with the purity of 93.88% were achieved at pH 11 and initial Mg2+ concentration of 3 mmol/L, whereas cattiite crystals with the purity of 98.99% were observed at pH 9 and the initial Mg2+ concentration of 50 mmol/L. Notably, K-struvite and cattiite crystals have obvious differences in morphology and particle size, separately exhibiting small rod-like and large plate-like structures. Interestingly, GA-BPANN accurately simulated and predicted K-struvite purity at various pHs and initial Mg2+ concentrations. The findings herein provide a theoretical foundation for obtaining high-purity K-struvite from waste streams at industrial scale, which also provide a reference for exploring the coprecipitation of multiple minerals in waste streams.

Predicting seismic interaction effect between soil and structure group using convolutional neural network

Quantifying the influence of seismic interaction between soil and structure group (SSGI) is of great significance to seismic design, retrofit, and damage assessment of structures in densely built urban areas. To this end, this study proposes a one-dimensional convolutional neural network (1D-CNN) model to rapidly predict the influence of adjacent structures on the maximum inter-story drifts and base shears of RC frame structures. Based on an experimentally validated three-dimensional finite element method, 890 pairs of soil-single structure versus soil-structure group systems under different earthquake loadings are simulated. The dataset comprising 890 groups of input (i.e., soil and structure group parameters, ground motion acceleration) and output data (i.e., changes in maximum inter-story drift and base shear) is constructed to train the machine learning model. Subsequently, sensitivity analysis is performed to identify optimal hyperparameters for training the 1D-CNN model, whereas a back propagation artificial neural network (BP-ANN) model is established to compare the model performance. Results indicate that compared with the BP-ANN model, the 1D-CNN model has a more stable and robust architecture and features superior prediction accuracy. In particular, the developed 1D-CNN model has a mean absolute error of less than 2.3% and an absolute error of less than 5.4% for 90% of cases in the testing set. The superior performance of the 1D-CNN model makes it an effective and efficient tool to be applied to predict the seismic responses of RC frame buildings under the SSGI effect.

Changes in polyphenolic compounds and antioxidant activities of seed-used pumpkin during hydrothermal treatment

An environmentally friendly physical processing method, hydrothermal treatment (HT), was used to increase the content of specific compounds and antioxidant activities of seed-used pumpkin byproducts. The influence of hydrothermal temperature (80 °C–160 °C) and time (30–150 min) on changes in polyphenols and antioxidation was evaluated. The results revealed that the maximum free polyphenol content (140 °C for 120 min) was 3.96-fold higher than the untreated samples. Elevated temperature and long duration changed phenolic acid contents. For example, p-coumaric acid, rutin and chlorogenic acid exhibited a decreasing trend, and p-hydroxybenzoic acid, quercetin and cinnamic acid showed an increasing trend. Compared to controls, HT was significantly associated with increased antioxidant activities. To comprehensively reveal the influence of hydrothermal temperature and time on changes in polyphenolic content, back propagation artificial neural network (BP-ANN) models with accurate prediction ability were developed, and the results exhibited well-fitted and strong approximation ability (R2 > 0.95 and RMSE < 2 %) and stability.

Authentication of chaste honey adulterated with high fructose corn syrup by HS-SPME-GC-MS coupled with chemometrics

The adulteration of the honey industry is serious, especially the syrup adulteration method is difficult to detect. To develop the detection technology of chaste honey adulteration with high fructose corn syrup (HFCS), the condition of headspace solid-phase micro-extraction (HS-SPME) was optimized for extracting the volatile compounds of chaste honey. HS-SPME conditions were selected for optimization using Plackett Burman, steepest ascent and Box-Behnken. Volatile compounds from chaste honey adulterated with HFCS were analyzed by gas chromatography-mass spectrometry (GC-MS). M/Z RT pairs data from differences in chaste honey contents in HFCS-adulterated samples were analyzed by linear discriminant analysis (LDA), principal component analysis (PCA), and artificial neural network (ANN). The results indicated distinguished adulterated chaste honey at different proportions was not ideal using LDA and PCA. A back propagation (BP) artificial neural network （ANN）model was constructed based on m/z RT pair data. Correlation coefficients of training, verification, testing and comprehensive data of BP-ANN were 0.994, 0.945, 0.968 and 0.979, respectively, indicating good accuracy of the BP-ANN prediction model. The present study discusses a new strategy that determined the chaste honey contents in HFCS-adulterated samples as well as did not rely on the identification of volatile compounds.

Proteolysis modification targeting protein corona affects ultrasound-induced membrane homeostasis of saccharomyces cerevisiae: Analysis of lipid relative contributions on membrane properties.

Protein corona (PCN) adsorbed on the surface of nanoparticles has brought new research perspectives for the interaction between nanoparticles and microorganisms. In this study, the responses of saccharomyces cerevisiae' membrane lipid composition, the average length of the fatty acyl chains and the average number of unsaturation of fatty acids to ultrasound combined with nano-Fe3O4@PCN with time-limited proteolysis (nano-Fe3O4@TLP-PCN) was investigated. Lipidomic data was obtained using Ultra-high performance liquid chromatography coupled with a Q-Exactive plus mass spectrometer. The membrane potential, proton motive force assay and the membrane lipid oxidation were measured using Di-BAC4(3), DISC3(5) and C11-BODIPY581/591 as the probes. Combined with the approach of feasible virtual samples generation, the back propagation artificial neural network (BP-ANN) model was adopted to establish the mapping relationship between lipids and membrane properties. The time-limited proteolysis targeting wheat PCN-coated Fe3O4 nanoparticles resulted in regular changes of hydrodynamic diameters, ζ-potentials, and surface hydrophobicity. In addition, with the prolongation of PCN proteolysis time, disturbances of 3 S.cerevisiae membrane characteristics, and membrane lipidomic remodeling in response to ultrasound+ nano-Fe3O4@PCN were observed. The analysis of relative importance which followed revealed that ergosterol, phosphatidylserine, and phosphatidylinositol phosphate had the greatest influence on membrane potential. For membrane lipid oxidation, ceramide, phosphatidylethanolamine, and sitosterol ester contribute 16.2, 14.9, and 13.1%, respectively. The relative contributions of six lysolecithins to the dissipation of proton motive force remained limited. An adaptation mechanism of cell membrane to proteolyzed PCN, wherein lipidome remodeling could preserved functional membrane phenotypes was revealed. Furthermore, it is highlighted that the relative importances of SiE, Cer, PE and PIP in determining membrane potential, PMF dissipation and membrane lipid oxidation by establishing FVSG-BP-ANN model.

Genetic optimized Al–Mg alloy constitutive modeling and activation energy analysis

As an intelligent global optimization method, the genetic algorithm has tremendous potential for improving flow behavior modeling and analysis. Based on flow stress-true strain curves of Al–Mg AA5005 alloy under temperature 563∼773 K and strain rate 0.0003∼0.03s−1, a phenomenological model named Arrhenius-type (A-T) was established to describe the flow behavior. On this basis, the genetic optimized A-T (GA-T) model with higher precision was obtained by optimizing A-T parameters α, n, Q and lnA. To reduce the large computing power consumed by unnecessary complex topological network structure when conducting simulations by the back propagation artificial neural network (BP-ANN) model, a genetic optimized BP-ANN (GBP-ANN) model was designed through determining the initial values of weights, biases and hyper parameters. The presented GBP-ANN model inherits the advantage of the BP-ANN model’s high accuracy as well as maintaining the simplest structure. The statistical analysis demonstrates that the GBP-ANN model possesses the best flow behavior description ability among three established models. Moreover, the GBP-ANN also shows a better generalization performance than the GA-T model. Lastly, with the help of the GA-T model, the activation energy map was plotted to determine the desirable deformation condition analyze the deformation mechanism. Our work presents a combination of GBP-ANN model and genetic optimized Q analysis, thus shedding new light on high accuracy flow behavior modeling and deformation mechanism analysis.

Ultra-high performance concrete design method based on machine learning model and steel slag powder

This paper presents a design method for green ultra-high performance concrete (UHPC) through machine learning (ML) models and steel slag (SS) powder. First, different UHPC compressive strength prediction models are developed, which include the linear regression (LR) model, back propagation artificial neural network (BP-ANN) model, genetic algorithm optimized ANN (GABP-ANN) model, random forest (RF) model and support vector machine (SVM) model. The prediction results of different ML models are compared and the SVM model is identified to have higher accuracy (R2 =0.93). Then, a design model for UHPC is proposed by the modified Andreasen & Andersen (MAA) particle packing model and SVM model. The model takes into account the particle size and the interactions of the materials, resulting in UHPC with excellent mechanical properties and a low amount of cement. Finally, green UHPC with low cement dosage and different SS powder replacement rates (substituted cement) are designed. The effects of SS on the mechanical properties, microscopic properties, hydration properties, drying shrinkage properties, toxic leaching properties and environmental properties of UHPC are investigated when the cement admixture in UHPC is low. The results indicate that the mechanical properties of UHPC decrease with the increase of SS substitution rate, but the decrease is limited. The addition of SS powder can not only inhibit the dry shrinkage and reduce the heat of hydration, but also reduce the impact on the environment.

Rapid determination of the total content of oleanolic acid and ursolic acid in Chaenomelis Fructus using near-infrared spectroscopy.

Chaenomelis Fructus is a widely used traditional Chinese medicine with a long history in China. The total content of oleanolic acid (OA) and ursolic acid (UA) is taken as an important quality marker of Chaenomelis Fructus. In this study, quantitative models for the prediction total content of OA and UA in Chaenomelis Fructus were explored based on near-infrared spectroscopy (NIRS). The content of OA and UA in each sample was determined using high-performance liquid chromatography (HPLC), and the data was used as a reference. In the partial least squares (PLS) model, both leave one out cross validation (LOOCV) of the calibration set and external validation of the validation set were used to screen spectrum preprocessing methods, and finally the multiplicative scatter correction (MSC) was chosen as the optimal pretreatment method. The modeling spectrum bands and ranks were optimized using PLS regression, and the characteristic spectrum range was determined as 7,500–4,250 cm−1, with 14 optimal ranks. In the back propagation artificial neural network (BP-ANN) model, the scoring data of 14 ranks obtained from PLS regression analysis were taken as input variables, and the total content of OA and UA reference values were taken as output values. The number of hidden layer nodes of BP-ANN was screened by full-cross validation (Full-CV) of the calibration set and external validation of the validation set. The result shows that both PLS model and PLS-BP-ANN model have strong prediction ability. In order to evaluate and compare the performance and prediction ability of models, the total content of OA and UA in each sample of the test set were detected under the same HPLC conditions, the NIRS data of the test set were input, respectively, to the optimized PLS model and PLS-BP-ANN model. By comparing the root-mean-square error (RMSEP) and determination coefficient (R2) of the test set and ratio of performance to deviation (RPD), the PLS-BP-ANN model was found to have better performance with RMSEP of 0.59 mg·g−1, R2 of 95.10%, RPD of 4.53 and bias of 0.0387 mg·g−1. The results indicated that NIRS can be used for the rapid quality control of Chaenomelis Fructus.

A back propagation neural network model for accurately predicting the removal efficiency of ammonia nitrogen in wastewater treatment plants using different biological processes

Accurately predicting the water quality of treated water from a water treatment plant (WWTP) based on the obtained operating database is of great significance. However, it is difficult for common mechanistic models to work well. In this study, a back propagation artificial neural network (BPANN) model with high accuracy was developed to predict the denitrification efficiency based on a 1-year operating database. Standardized principal component analysis (PCA) methods were used to address the data, and the PCA processed data exhibited the best accuracy. In three WWTPs adopting the anaerobic/anoxic/oxic (A2O) process, the ammonia nitrogen removal efficiency of WWTPs was successfully predicted by using five variables: inlet flow rate, pH value, original ammonia nitrogen concentration, Chemical oxygen demand (COD) concentration, and total phosphorus concentration. Importantly, the obtained BPANN model can be effectively used for other widely used treatment processes, such as oxidation ditch (OD), sequencing batch reactor activated sludge process (SBR), membrane bioreactor (MBR), and cyclic activated sludge technology (CAST), by simply optimizing the training data ratios between 50/50 and 90/10. This is the first trial to set up a universal model for predicting the denitrification efficiency of WWTPs adopting common biological processes. The model could be used to choose the optimum treatment process in the new WWTP design or take action in advance to avoid the risk of excessive emissions when the already built WWTPs are subjected to sudden shocks.

Computer-assisted in vitro reconstitution of purine degradation pathway to lower the purine content in food.

With the increasing prevalence of gout and its etiological hyperuricemia, dietary control of gout based on low-purine food according to patients' eating habits is becoming a better choice compared to the existing drug treatment such as allopurinol with notorious side effects. Reconstructing the purine metabolic pathway in vitro to degrade purine substances in food into natural functional allantoin appears to be an innovative method for preparing nutritious and healthy food of low purine content. The present study reports a computer-assisted in vitro reconstruction of four purinolytic enzymes metabolizing adenosine into allantoin to reduce purine content in food for personalized dietary control of hyperuricemia and gout. Under the optimum reaction conditions of 40 °C and pH7, 0.1U of enzymes and 0.5 mmol L-1 adenosine determined by an orthogonal test design, 16 different enzyme complexes were experimentally tested. The tested enzyme composition and allantoin production values were used as input and output to build a three-layer back propagation artificial neural network (BP-ANN) model, which was further optimized by a genetic algorithm (GA). The optimum enzyme complex predicted by the GA-BP-ANN model produced 248.08±7.832 μmol L-1 allantoin, which was 19.9% higher than equimolar mixture of enzymes, and also more efficiently lowered purine contents in beer, as well as beef and yeast extracts. This is the first in vitro reconstitution of complete purine metabolic pathway by combining ANN and GA technologies, with successful application with respect to lowering the purine content in food, indicating a promising application of computer-assisted in vitro reconstitution of purinolytic pathway in low-purine food to prevent hyperuricemia and gout. © 2022 Society of Chemical Industry.

Shelf-Life Prediction and Critical Value of Quality Index of Sichuan Sauerkraut Based on Kinetic Model and Principal Component Analysis

Kinetic models and accelerated shelf-life testing were employed to estimate the shelf-life of Sichuan sauerkraut. The texture, color, total acid, microbe, near-infrared analysis, volatile components, taste, and sensory evaluation of Sichuan sauerkraut stored at 25, 35, and 45 °C were determined. Principal component analysis (PCA) and Fisher discriminant analysis (FDA) were used to analyze the e-tongue data. According to the above analysis, Sichuan sauerkraut with different storage times can be divided into three types: completely acceptable period, acceptable period, and unacceptable period. The model was found to be useful to determine the critical values of various quality indicators. Furthermore, the zero-order kinetic reaction model (R2, 0.8699–0.9895) was fitted better than the first-order kinetic reaction model. The Arrhenius model (Ea value was 47.23–72.09 kJ/mol, kref value was 1.076 × 106–9.220 × 1010 d−1) exhibited a higher fitting degree than the Eyring model. Based on the analysis of physical properties, the shelf-life of Sichuan sauerkraut was more accurately predicted by the combination of the zero-order kinetic reaction model and the Arrhenius model, while the error back propagation artificial neural network (BP-ANN) model could better predict the chemical properties. It is a better choice for dealers and consumers to judge the shelf life and edibility of food by shelf-life model.

Exploring thermal contact conductance between two contact solids by artificial neural network

Realizing accurate measurement for thermal contact conductance (TCC) is difficult and needs many additional measurements. In this paper, we design a three-layer back propagation artificial neural network (ANN) model to retrieve TCC of two contact solid specimens. The model can retrieve TCC only according to the measured temperatures and loading pressures, and the effects of loading pressure, temperature and surface roughness on TCC are considered. Besides, thermal conductivity, heat flux and the parameter of the TCC empirical fitting formula can be obtained simultaneously. The retrieved results are validated by experimental values. The results show that for two contact pairs with different surface roughness, the retrieved parameters of TCC empirical fitting formula are clear reasonable and can be explained conceptually compared with the published studies.

Flow Behavior of AA5005 Alloy at High Temperature and Low Strain Rate Based on Arrhenius-Type Equation and Back Propagation Artificial Neural Network (BP-ANN) Model.

To realize the purpose of energy saving, materials with high weight are replaced by low-weight materials with eligible mechanical properties in all kinds of fields. Therefore, conducting research works on lightweight materials under specified work conditions is extremely important and profound. To understand the relationship of aluminum alloy AA5005 among flow stress, true strain, strain rate, and deformation temperature, hot isothermal tensile tests were conducted within the strain rate range 0.0003–0.03 s−1 and temperature range 633–773 K. Based on the true stress-true strain curves obtained from the experiment, a traditional constitutive regression Arrhenius-type equation was utilized to regress flow behaviors. Meanwhile, the Arrhenius-type equation was optimized by a sixth-order polynomial function for compensating strain. Thereafter, a back propagation artificial neural network (BP-ANN) model based on supervised machine learning was also employed to regress and predict flow stress in diverse deform conditions. Ultimately, by introducing statistical analyses correlation coefficient (R2), average absolute relative error (AARE), and relative error () to the comparative study, it was found that the Arrhenius-type equation will lose accuracy in cases of high stress. Additionally, owning higher R2, lower AARE, and more concentrative value distribution, the BP-ANN model is superior in regressing and predicting than the Arrhenius-type constitutive equation.