Nonlinear Artificial Neural Network Research Articles

GELİŞİM EŞİTSİZLİĞİ MODELİ: ULUSLARARASI TİCARETE DİNAMİK BİR YAKLAŞIM

The development disparity model focuses on the difference in the trade yields of countries by considering the nature of the trade balance in the advanced sectors. This paper employs the non-linear Artificial Neural Network technique for ten developing countries in the analysis section. According to the analysis outcomes, while China and Thailand's aggregate international trade yields have an ascending tendency, Indonesia, Paraguay, Colombia, Turkey, Argentina, and Mexico's aggregate trade yields exhibit a drastic descending trend. On the other hand, the tendency of aggregate trade yields in India and Brazil is generally negative, yet this tendency is decreasing inconsistently.

Inverse estimation of hot-wall heat flux using nonlinear artificial neural networks

Compared with cold-wall heat flux measurement, hot-wall heat flux indicates more information from dynamic heat transfer between the aerothermodynamic boundary layer and the true model’s surface; thus, hot-wall heat flux measurement has become predominant in the aerothermodynamic measurement field. In harsh aerothermodynamic in-flight or ground tests, the hot-wall heat flux must be determined from time history temperature measurements at one or more interior locations. Therefore, accounting for the temperature dependence of the thermophysical properties, hot-wall heat flux measurement essentially results in the solution of the nonlinear inverse heat conduction problem (IHCP). In this paper, a novel inverse estimation of hot-wall heat flux using nonlinear artificial neural networks (ANN) is presented. First, motivated by the hybrid method proposed by Clayton A. Pullins, David O. Hubble, and Tom E. Diller et al., [2010], a modified hybrid heat flux measurement method using two in-depth thermocouples is proposed, which avoid to directly measure surface temperature of gauge or model under harsh aerodynamic heating environments; accounting for the unknown temperature dependent thermophysical properties, a new heat flux inverse estimation model of a nonlinear ANN is proposed and identified to approximate the modified hybrid measurement method through calibration experiment. This heat flux inverse estimation method does not need to solve a first kind Volterra integral equation and to obtain the information about the thermophysical properties of heat conduction body, and the thermal inertia, locations of thermocouples. This paper presents xenon lamp calibration and arc-heated wind tunnel experiments that validate the new inverse estimation method combined with the fabricated hot-wall heat flux sensor and its probe. These experimental results show that, in general, the dynamical hot-wall heat flux estimated based on the proposed method agree well with the known calibrated values and the stagnation heat fluxes of the classic slug calorimeter.

PM2.5 concentration prediction during COVID-19 lockdown over Kolkata metropolitan city, India using MLR and ANN models

Kolkata is the third densely populated city of India and Kolkata stands in the World's 25 most polluted cities along with 10 worse polluted cities in India. The relevant study claims that due to the imposition of lockdown during COVID-19 pandemic, the atmospheric pollution level has been significantly reduced over the metropolitan city Kolkata like other cities of the world. The main objective of this study is to predict the concentration of PM2.5 using multiple linear regression (MLR) and artificial neural network (ANN) models and similarly, to compare the accuracy level of two models. The concentration of PM2.5 data has been obtained from state pollution control board, Govt. of West Bengal and daily meteorological data have been collected from the world weather website. The results show that non-linear artificial neural network model is more rational compared with multiple linear regression model due to its high precision and accuracy level (in respect to RMSE, MAE and R2). In this research artificial neural network (ANN) model exhibited higher accuracy during the training and testing phases (root mean square error (RMSE), mean absolute error (MAE) and R2 indicate 3.74, 1.14 and 0.91 respectively in training phase and 2.55, 4.32 and 0.69 in testing phase respectively). This model (ANN)) can be applied to predict the concentration of PM2.5 during the execution of urban air quality management plan.

Non-valvular atrial fibrillation recurrence after sinus rhythm restoring at different follow-up periods: phenotype-genotype high-risk groups, considering rs10465885 polymorphism in connexin-40 gene

Abstract Funding Acknowledgements Type of funding sources: None. Purpose to provide risk stratification of non-valvular atrial fibrillation (AF) recurrence after sinus rhythm restoring (SRR) in patients (pts) with AF at different follow-up periods (3 months (AFr3m), 12 months (AFr12m) and 18 months [AFr18m]), based on phenotype-genotype high-risk groups, considering rs10465885 polymorphism in connexin-40 gene (SNP-Cx40). Methods. We enrolled 186 pts (mean age (55 ± 10) years; males 123 [66,1%]) with AF (paroxysmal – 86, persistent – 72, stable – 28 pts; first onset (FO) AF – 48 pts). Clinical, laboratory and echocardiographic data were analyzed. SNP-Cx40 was genotyped by real time PCR (T – reference, C – minor allele) in 112 pts. The genotypes were distributed as follows: TT – 25,9% (n = 29); CT – 49,1% (n = 55); CC – 25,0% (n = 28). SRR was performed in 112 cases (102 pts) with non-permanent AF: 30 – pharmacological cardioversion (PCV), 62 – direct-current cardioversion (DCV), 20 cases – radiofrequency ablation (RFA). AFr3m occurred in 53 (43,4%) of 122 available cases; AFr12m – 65,5% (76/116); AFr18m – 75,2% (79/105). The Artificial Neural networks (ANNs) analysis was performed to select the AF recurrence predictors. We considered the ANN activation function value (Y) and its relation to Y cut-off value (Ycrit). In case of Y > Ycrit, the AF recurrence risk was considered as «high». Results. We built three nonlinear ANN models (multilayer perceptrons) for AFr3m (Ycrit = 0,496), AFr12m (Ycrit = 0,503) and AFr18m (Ycrit = 0,720) risk prediction. In case of SNP-Cx40 CC genotype carriage, we determined the additional increase of AFr3m risk after PCV in pts with CHA2DS2-VASc score «0» and normal (Y = 0,629) or mildly decreased (Y = 0,616) left ventricular mid-wall fractional shortening, and in the case of its moderate decrease – both after PCV (Y = 0,585) and DCV (Y = 0,627). The CC genotype was associated with AFr12m high risk in pts without heart failure (HF) and mildly increased left atrial dimension (LAD) – both after PCV (FO AF with known precise event duration (PED); Y = 0,906) or DCV (FO AF with unknown PED; Y = 0,911). Additionally, CC genotype was associated with AFr12m high risk after RFA in pts without HF and normal or mildly increased LAD (Y = 0,912), and in pts with HF B or C1 stage (according to modified AHA/ACC classification) with moderately increased LAD (Y = 0,912). The high-risk groups of AFr18m in pts with CC genotype were as follows (Y = 0,913): after RFA in pts with recurrent AF and presence of episodes lasting ≥7 days; after PCA in case of FO AF with unknown PED and index episode lasting ≥1 month; after DCV in case of FO AF with unknown PED and index episode lasting ≥12 months. Conclusion. AFr3m, AFr12m and AFr18m, besides SNP-Cx40, were non-linearly associated with SRR type, and certain clinical and echocardiographic phenotypic parameters, which could be used for AF recurrence risk stratification, with the selection of phenotype-genotype high-risk groups, considering SNP-Cx40.

COMANDO: A Next-Generation Open-Source Framework for Energy Systems Optimization

Existing open-source modeling frameworks dedicated to energy systems optimization typically utilize (mixed-integer) linear programming ((MI)LP) formulations, which lack granularity for technical system design and operation. We present COMANDO, an open-source Python package for component-oriented modeling and optimization for nonlinear design and operation of integrated energy systems. COMANDO allows to assemble system models from component models including nonlinear, dynamic and discrete characteristics. Based on a single system model, different deterministic and stochastic problem formulations can be obtained by varying objective function and underlying data, and by applying automatic or manual reformulations. The flexible open-source implementation allows for the integration of customized routines required to solve challenging problems, e.g., initialization, problem decomposition, or sequential solution strategies. We demonstrate features of COMANDO via case studies, including automated linearization, dynamic optimization, stochastic programming, and the use of nonlinear artificial neural networks (ANNs) as surrogate models in a reduced-space formulation for deterministic global optimization.

Analysis of Critical Velocity of Cold Spray Based on Machine Learning Method with Feature Selection

Cold spraying has a potential application prospect in the field of repairing and additive manufacturing. The critical velocity of the cold spray is a key factor that determines the adhesion of particles during the cold spraying process, and it only depends on the particle parameters under the same working conditions. In the present study, the relationship between particle parameters and critical velocity is investigated using a feature selection method to obtain the influence weight of different particle parameters. Based on the results of feature selection, linear and nonlinear artificial neural networks are established to predict the critical velocity, respectively. The results of the feature selection show that the mechanical parameters of the material have a higher influence weight on the critical velocity than thermal parameters. In the prediction model, the ANN (artificial neural network) method shows a good prediction, and the nonlinear ANN model achieves better generalization ability than the linear ANN model and empirical formula with 95.24% prediction accuracy on the original data set and 96.45% prediction accuracy on the new data set.

Optimal Design of Hot-Dip Galvanized DP Steels via Artificial Neural Networks and Multi-Objective Genetic Optimization

This modeling and optimization study applies a non-linear back-propagation artificial neural network, commonly denoted as BPNN, to model the most important mechanical properties such as yield strength (YS), ultimate tensile strength (UTS) and elongation at fracture (EL) during the experimental processing of hot-dip galvanized dual-phase (GDP) steels. Once the non-linear BPNN is properly trained, the most important variables of the continuous galvanizing process, including initial/first cooling rate (CR1), holding time at the galvanizing temperature of 460 °C (tg) and the final/second cooling rate (CR2), are obtained in an optimal way using an evolutionary approach. The experimental development of GDP steels in continuous processing lines with outstanding mechanical properties (550 < YS < 750 MPa, 1100 MPa < UTS and 10% < EL) is possible by using a combined hybrid approach based in BPNN and multi-objective genetic algorithm (GA). The proposed computational method is applied to the specific design of an actual manufacturing process for the first time.

Radial basis function-based node localization for unmanned aerial vehicle-assisted 5G wireless sensor networks

Localization or positioning of wireless sensor nodes is an essential task for a wide range of applications in wireless sensor networks-based fifth generation (5G) networks. Node localization using mobile aerial beacon nodes (MABNs) provides high localization accuracy and less deployment cost compared to the localization using fixed ground beacon nodes because MABN deployed in unmanned aerial vehicles sends signals to unknown nodes (UNs) through reliable air to ground (AG) channel link. The classical received signal strength indicator (RSSI)-based multilateration and optimization-based least square localization (OLSL) schemes result high localization error because of the nonlinear distortions induced in the wireless channel. So, the highly nonlinear artificial neural network (ANN) models such as multilayer perceptron (MLP) and radial basis function (RBF) are used effectively for nonlinear node localization problems. ANN-based localization techniques are also capable of localizing mobile UNs. Further, the RBF with Gaussian activation has a little edge over the MLP with sigmoid activation because the wireless channel usually modelled with a Gaussian random variable. The simulation results included in this article also justify the efficacy of the proposed RBF localization with performance gain of about (7–70) % over MLP, OLSL and RSSI localization techniques.

Do economic statistics contain information to predict stock indexes futures prices and returns? Evidence from Asian equity futures markets

This research examines the impact of local and international market factors on the pricing of stock indexes futures in East Asian countries. The purpose of this paper is to present a study of the significant factors that determine the major stock indexes futures’ prices of Hong Kong, Malaysia, Singapore, South Korea and Taiwan. This study first investigates the relationships between Hang Seng Index Futures, KLCI Futures, SiMSCI Futures, KOSPI Futures, Taiwan Exchange Index Futures and local interest rates, dividend yields, local exchange rates, overnight SP 10 years in-sample data are used for testing and developing the pricing models, and 1 year out-of-sample data is used for the purpose of verifying the predicted values of the stock indexes futures. Using simple linear regressions, local interest rates, dividend yields, exchange rates, overnight S&P500 and ATMI are found to have significant impact on these futures contracts. In this research, the next period close is predicted using simple linear regression and non-linear artificial neural network (ANN). An examination of the prediction results using nonlinear autoregressive ANN with exogenous inputs (NARX) shows significant abnormal returns above the passive threshold buy and hold market returns and also above the profits of simple linear regression (SLR). The empirical evidence of this research suggests that economic statistics contain information which can be extracted using a hybrid SLR and NARX trading model to predict futures prices with some degree of confidence for a year forward. This justifies further research and development of pricing models using fundamentally significant economic determinants to predict futures prices.

Associations Between Mental Health, Interoception, Psychological Flexibility, and Self-as-Context, as Predictors for Alexithymia: A Deep Artificial Neural Network Approach.

Background: Alexithymia is a personality trait which is characterized by an inability to identify and describe conscious emotions of oneself and others.Aim: The present study aimed to determine whether various measures of mental health, interoception, psychological flexibility, and self-as-context, predicted through linear associations alexithymia as an outcome. This also included relevant mediators and non-linear predictors identified for particular sub-groups of participants through cluster analyses of an Artificial Neural Network (ANN) output.Methodology: Two hundred and thirty participants completed an online survey which included the following questionnaires: Toronto alexithymia scale; Acceptance and Action Questionnaire 2 (AQQII); Positive and Negative Affect Scale (PANAS-SF), Depression, Anxiety, and Stress Scale 21 (DAS21); Multidimensional Assessment of Interoceptive Awareness (MAIA); and the Self-as-Context (SAC) scale. A stepwise backwards linear regression and mediation analysis were performed, as well as a cluster analysis of the non-linear ANN upper hidden layer output.Results: Higher levels of alexithymia were associated with increased psychological inflexibility, lower positive affect scores, and lower interoception for the subscales of “not distracting” and “attention regulation.” SAC mediated the relation between emotional regulation and total alexithymia. The ANNs accounted for more of the variance than the linear regressions, and were able to identify complex and varied patterns within the participant subgroupings.Conclusion: The findings were discussed within the context of developing a SAC processed-based therapeutic model for alexithymia, where it is suggested that alexithymia is a complex and multi-faceted condition, which requires a similarly complex, and process-based approach to accurately diagnose and treat this condition.

A quantitative multivariate methodology for unsupervised class identification in pistachio (Pistacia vera L.) plant leaves size

Aim of study: Genetic diversity of pistachio, can be evaluated by using different descriptors, as adopted in international certification systems. Mainly the descriptors are morphological traits as leaf, which represents an important organ for its sensibility to growth conditions during the expansion phase. This study adopted a rapid and quantitative non-hierarchic clustering classification (k-means), to extract size classes basing on the contemporary combination of different morphological traits (i.e., leaf stalk length, terminal leaf length, terminal leaf width and terminal leaf ratio) of a varietal collection composed by 21 pistachio cultivars.Area of study: Worldwide.Material and methods: The unsupervised non-hierarchic clustering technique was adopted to the entire samples of pistachio leaves from k=2 to k=15 for both four morphological variables (i.e., leaf stalk length, terminal leaf length, terminal leaf width and terminal leaf ratio) and three morphological variables (i.e., terminal leaf length, terminal leaf width and terminal leaf ratio).Main results: A classification model only on the three morphological variables (for results of statistical analysis in which the groups resulted to be more separated and different for all the variables), with k= 5 (five groups), was constructed using a non-linear artificial neural network approach. The percentages of bad prediction in both training and testing resulted equal to 0%. The “terminal leaf length” returned the higher impact (44.89%).Research highlights: The contemporary combination of different morphological leaf traits, allowed to create an automatic classification of size classes of great importance for cultivar identification and comparison.

RETRACTED ARTICLE: Development of air quality monitoring (AQM) models using different machine learning approaches

Air Quality assessment and forecasting are the essentials today and they attracted many researchers. Environmental organizations regularly monitor and predict the air contaminants to make the public awareness, provide a better environment, and suitable for human health. Physical factors like climate changes, Industrialization, Fires and Urbanization are some of the factors which directly affect and reduce the air quality. All these data are time-series and real-time data. The primary pollutant is PMx that affect the respiratory systems and cardiac activity of humans. The secondary pollutants are SO2, CO, NOx, and O3. Each has allowable range of concentration levels. In this work, meteorological elements are collected in different locations in last 5 years, with time window of 24 h and mapped to the concentration level of pollutants. The Machine Learning(ML) Methods such as Non-Linear Artificial Neural Network(ANN), Statistical Multilevel Regression, Neuro- Fuzzy and Deep Learning Long-Short-Term Memory (DL-LSTM) are used; to find the current concentration level of pollutants and will be useful for Real Time Correction (RTC) to give a feedback that can be used to reduce the contaminants in air for further days. The results are compared with the parameters such as R2, RMSE and MAPE. Using these methods, the concentration level of contaminants is predicted with the deviation of R2 in the range of 0.71–0.89. The results proved that DL-LSTM suits well when comparing to the ANN, Neuro-fuzzy and regression algorithms.

Energy Consumption Prediction in a Novel Automated Photovoltaic Design Platform

This paper describes a multi-step algorithm used to predict and typify the energy consumption profile of a prosumer, allowing the automation of the design of self-consumption photovoltaic (PV) power systems in a novel platform called PV SPREAD. The algorithm uses different methodologies to address various possible scenarios of data availability. In this paper, those scenarios are addressed using nonlinear autoregressive artificial neural networks (ANN) with external inputs (NARX) to predict energy consumption. Results reveal that the proposed algorithm successfully addresses data gaps in a hotel load profile used as a case study. The results also show the limitations of NARX when residential clients are analyzed.

Artificial neural network (ANN), M5P-tree, and regression analyses to predict the early age compression strength of concrete modified with DBC-21 and VK-98 polymers

In this study, the effect of two water-reducer polymers with smooth and rough surfaces on the workability, and the compression strength of concrete from an early age (1 day) up to 28 days of curing was investigated. The polymer contents used in this study varied from 0 to 0.25% (wt%). The initial ratio between water and cement ( $$ \frac{w}{c} $$ ) was 60%, and it slowly reduced to 0.46 by increasing the polymer contents. The compression strength of concrete was increased significantly with increasing the polymer contents by 24–95% depending on the polymer type, polymer content, $$ \frac{w}{c} $$ , and curing age. Because of a fiber net (netting) in the concrete when the polymers were added which leads to a decrease void between the particles, binding the cement particles, therefore, increased rapidly the viscosity for the fresh concrete and the compression strength of the hardened concrete. This study also aims to establish systematic multiscale models to predict the compression strength of concrete containing polymers and to be used by the construction projects with no theoretical restrictions. For that purpose, 88 concrete samples modified with two types of polymer (44 samples for each modification) has been tested, analyzed, and modeled. Linear, nonlinear regression, M5P-tree, and artificial neural network (ANN) approaches were used for the qualifications. In the modeling process, the most relevant parameters affect the strength of concrete, i.e., polymer incorporation ratio (0–0.25% of cement’s mass), water-to-cement ratio (0.46–0.6), and curing ages (1–28 days). Among the used approaches and based on the training data set, the model made based on the nonlinear regression, ANN, and M5P-tree models seem to be the most reliable models. The sensitivity investigation concludes that the curing time is the most dominating parameter for the prediction of the maximum stress (compression strength) of concrete with this dataset.

Aerodynamic State and Loads Estimation Using Bioinspired Distributed Sensing

Flying animals exploit highly nonlinear dynamics to achieve efficient and robust flight control. It appears that the distributed flow and force sensor arrays found in flying animals are instrumental in enabling this performance. Using a wind-tunnel wing model instrumented with distributed arrays of strain and pressure sensors, we characterized the relationship between the distributed sensor signals and aerodynamic and load-related variables. Estimation approaches based on nonlinear artificial neural networks (ANNs) and linear partial least squares were tested with different combinations of sensor signals. The ANN estimators were accurate and robust, giving good estimates for all variables, even in the stall region when the distributed array pressure and strain signals became unsteady. The linear estimator performed well for load estimates but was less accurate for aerodynamic variables such as angle of attack and airspeed. Future applications based on distributed sensing could include enhanced flight control systems that directly use measurements of aerodynamic states and loads, allowing for increase maneuverability and improved control of unmanned aerial vehicles with high degrees of freedom such as highly flexible or morphing wings.

ANN, M5P-tree model, and nonlinear regression approaches to predict the compression strength of cement-based mortar modified by quicklime at various water/cement ratios and curing times

This study’s goal is to establish systematic multiscale equations to estimate the compressive strength of cement mortar with a high volume of lime (L) and to be used by the construction industry with no theoretical restrictions. For that purpose, a wide tested data and the data gathered in the literature (a total of 392 tested cement mortar modified with lime) have been statically analyzed and modeled. The lime content ranged from 0 to 45% (by cement weight). Depending on literature data the w/c ranged from 0.3 to 0.74, the w/c of 0.5 was selected for this research. The compressive strength of lime-modified cement mortar for up to 28 days ranged from 3 to 75 MPa. The compression strength of the cement mortar reduced with an increasing percentage of lime. The linear and nonlinear regression, M5P-tree, and artificial neural network (ANN) technical approaches were used for the qualifications. In the modeling process, the most relevant parameters affecting the compression strength of cement mortar, i.e. lime (L) incorporation ratio (0–45% of cement’s mass), water-to-cement ratio (0.3–0.74), and curing ages (1 to 28 days). According to the statistical assessment such as R, MAE, and RMSE, the compression strength of cement mortar can be predicted very well in terms of water-to-cement ratio, lime content, and curing age using various simulation techniques. The maximum and minimum error between the actual test results and the outcome of the prediction using NLR and ANN (training dataset) were 0.01–21 MPa and 0.012–9 MPa, respectively, and ranged between 0.03 and 14 MPa and 0.02 and 6 MPa errors, respectively, in terms of tested data. The margin of error in using the nonlinear regression-based model (NLR) and ANN for the training dataset was 1.41 and 1.92, respectively, and it was 2.26 and 2.44 respectively in using the tested dataset. The outcomes of this paper suggest that the nonlinear regression-based model (NLR) and ANN are performing better than other applied models using training and testing datasets. The result of the sensitivity investigation was the curing period that is the highest dominating value for the prediction of the compressive strength of cement mortar.

The Estimation of the Critical Flashover Voltage of Insulators Using the Computational Intelligence

Background: Electric insulation is generally a vital factor in both the technical and economic feasibility of complex power and electronic systems. Several researches focus on the behavior of insulators under polluted conditions. That they are mathematical and physical models of insulators, experiments and simulation programs. Also experiments on critical flashover voltage are timeconsuming and have more limitations such as high cost and need for especial equipment’s. Objective: This paper focused on optimized predicting of critical flashover voltage of Polluted insulators based on artificial intelligence. Methods: Fuzzy logic and artificial neural networks are used in order to have the best estimation of the critical flashover. Results: In this way the correlation index (regression coefficient) improved about 2% toward previous works with same experimental data sets. Additionally, with using the properties of nonlinear artificial neural networks we can have the perfect (R=100%) prediction of the critical flashover voltage on experimental dataset. Conclusion: In this paper two methods for the estimation of critical flashover voltage of polluted insulators using fuzzy logic and neural networks was presented. the regression coefficient R achieved by the optimal parameters is 98.4% while in previous work is 96.7%. In neural network model we have regression coefficient 100% and in previous neural network model it was 99%. our test set is the same as previous works and achieved from experiments. These results show that fuzzy proposed methods are powerful and useful tools lead to a more accurate, generalized and objective estimation of the critical flashover voltage.

A tropical Atlantic dynamics analysis by combining machine learning and satellite data

The western tropical Atlantic Ocean is a very energetic and highly variable region. It is one of the main contributors to the inter-hemispheric mass and heat transports. This study aim is to give a new picture of the space and time variability of this region using statistical tools applied to five different satellite measurements (Sea Surface Temperature, Sea Surface Salinity, ocean topography, wind stress vectors). We first processed each data set by using a Self-Organizing Maps (SOM), which is an efficient clustering methodology based on non-linear artificial neural networks to compress the information embedded in the data. The SOM was then combined with a Hierarchical Ascendant Classification (HAC) to cluster the different phenomena in a small number of classes whose physical characteristics are easy to identify. Three classes were identified which allowed us to analyse the dynamics of the North Brazil Current, and the North Equatorial Countercurrent, respectively, and their links with the Inter-Tropical Convergence Zone and the Amazon and Orinoco river runoffs. The SOM + HAC analysis gave a coherent picture of the concomitant seasonal variability of the variables. Furthermore, we were able to point out the correlations existing between salinity features recently discovered and wind, temperature, and dynamic topography structures. Applying our method to the interannual signals, we showed a year to year variability which deserves further analysis.

Assess deep learning models for Egyptian exchange prediction using nonlinear artificial neural networks

Financial analysis of the stock market using the historical data is the exigent demand in business and academia. This work explores the efficiency of three deep learning (Dl) techniques, namely Bayesian regularization (BE), Levenberg–Marquardt (lM), and scaled conjugate gradient (SCG), for training nonlinear autoregressive artificial neural networks (NARX) for predicting specifically the closing price of the Egyptian Stock Exchange indices (EGX-30, EGX-30-Capped, EGX-50-EWI, EGX-70, EGX-100, and NIlE). An empirical comparison is established among the experimented prediction models considering all techniques for the time horizon of 1 day, 3 days, 5 days, 7 days, 5 days and 30 days in advance, applying on all the datasets used in this study. For performance evaluation, statistical measures such as mean squared error (MSE) and correlation R are used. From the simulation result, it can be clearly suggested that BR outperforms other models for short-term prediction especially for 3 days ahead. On the other hand, lM generates better prediction accuracy than BR- and SCG-based models for long-term prediction, especially for 7-day prediction.

ANN, M5P-tree and nonlinear regression approaches with statistical evaluations to predict the compressive strength of cement-based mortar modified with fly ash

This study aims to establish systematic multiscale models to predict the compressive strength of cement mortar containing a high volume of fly ash (FA) and to be used by the construction industry with no theoretical restrictions. For that purpose, a wide experimental data (a total of 450 tested cement mortar modified with FA) from different academic research studies have been statically analyzed and modeled. For that purpose, Linear and Nonlinear regression, M5P-tree, and Artificial Neural Network (ANN) technical approaches were used for the qualifications. In the modeling process, most relevant parameters affecting the strength of cement mortar, i.e. fly ash (class C and F) incorporation ratio (0−70% of cement's mass), water-to-binder ratio (0.235–1.2), and curing ages (1–365 days). According to the correlation coefficient (R), mean absolute error and the root mean a square error, the compressive strength of cement mortar can be well predicted in terms of w/b, fly ash, and curing time using various simulation techniques. The results of this study suggest that the Non-linear regression-based model (NLR) and ANN are performing better than other applied models using training and testing datasets. The sensitivity investigation concludes that the curing time is the most dominating parameter for the prediction of the compressive strength of cement mortar with this data set.