Input Variables Of Model Research Articles

Improving Reliability in Reconstruction of Landsat EVI Seasonal Trajectory over Cloud-Prone, Fragmented, and Mosaic Agricultural Landscapes

Although the Landsat 30 m Enhanced Vegetation Index (EVI) products are important input variables in land surface models, recurring Landsat 5/7 EVI time series over cloud-prone, fragmented, and mosaic agricultural landscapes is still a great challenge. In this study, we put forward a simple, but effective “Light and Temperature-Driven Growth model and Double Logistic function fusion algorithm” (LTDG_DL). The empirical basis of the LTDG_DL algorithm was traced from the de Wit crop growth simulation model and the commonly observed nonlinear correlation between the EVI and the Leaf Area Index (LAI). It assimilates the ground daily solar radiation and air temperature to generate seasonal profiles of the empirical LAI and EVI and conducts the within-season calibration of the empirical EVI by adjusting crop growth using cloud-free Landsat EVI observations. The initial date of seedling emergence (DOYini) and the accumulated Growing Degree Days for completion of the vegetative and Flowering stage (FGDDs) were variables to which the algorithm’s accuracy was most sensitive. The variable-constrained optimization of the LTDG_DL algorithm was performed by loading the seedling emergence calendar of local prevailing crops and establishing an FGDD lookup table with an exhaustive sampling without replication method. Compared to temporal interpolation functions and Landsat–MODIS spatiotemporal fusion algorithms, the LTDG_DL algorithm had superior performance in the predictions of the EVI increment slope at the vegetative growth stage, the timing of the peak EVI, and the protection of key Landsat EVI observations over cloud-contaminated and complex landscape agricultural systems. Finally, the advantages and limitations of the LTDG_DL algorithm are discussed.

Wind Turbine Gearbox Condition Monitoring Using Hybrid Attentions and Spatio-Temporal BiConvLSTM Network

Gearbox fault deterioration can significantly impact the safety, reliability, and efficiency of wind turbines, resulting in substantial economic losses for wind farms. However, current condition monitoring methods face challenges in effectively mining the hidden spatio-temporal features within SCADA data and establishing reasonable weight allocations for model input variables. To tackle these issues, we proposed a novel condition monitoring method for wind turbine gearboxes called HBCE, which integrated a feature-time hybrid attention mechanism (HA), the bidirectional convolutional long short-term memory networks (BiConvLSTM), and an improved exponentially weighted moving-average (iEWMA). Specifically, utilizing historical health SCADA data acquired through the modified Thompson tau data-cleaning algorithm, a normal behavior model (HA-BiConvLSTM) of gearbox was constructed to effectively extract the spatio-temporal features and learn normal behavior patterns. An iEWMA-based outlier detection approach was employed to set dynamic adaptive thresholds, and real-time monitor the prediction residuals of HA-BiConvLSTM to identify the early faults of gearbox. The proposed HBCE method was validated through actual gearbox faults and compared with conventional spatio-temporal models (i.e., CNN-LSTM and CNN&LSTM). The results illustrated that the constructed HA-BiConvLSTM model achieved superior prediction precision in terms of RMSE, MAE, MAPE, and R2, and the proposed method HBCE can effectively and reliably identify early anomalies of a wind turbine gearbox in advance.

Adaptive Data‐Driven Modeling Strategy Based on Feature Selection for an Industrial Natural Gas Sweetening Process

AbstractAs the core process of natural gas purification plant, natural gas sweetening directly affects the production efficiency and product quality of the purification plant. However, process modeling based on sulfur content prediction presents challenges in adaptability and accuracy. To tackle this, a machine learning‐based modeling approach is proposed that integrates an adaptive immune genetic algorithm with random forest (RF) to intelligently select process features as input variables for RF modeling. The industrial result indicates that the proposed method is able to remove interfering variables and adaptively achieve optimal model precision for different scenarios. It offers a novel research instrument for product quality monitoring in natural gas sweetening plants.

Deformation prediction model of concrete face rockfill dams based on an improved random forest model

The unique structure and complex deformation characteristics of concrete face rockfill dams (CFRDs) create safety monitoring challenges. This study developed an improved random forest (IRF) model for dam health monitoring modeling by replacing the decision tree in the random forest (RF) model with a novel M5' model tree algorithm. The factors affecting dam deformation were preliminarily selected using the statistical model, and the grey relational degree theory was utilized to reduce the dimensions of model input variables. Finally, a deformation prediction model of CFRDs was established using the IRF model. The ten-fold cross-validation method was used to quantitatively analyze the parameters affecting the IRF algorithm. The performance of the established model was verified using data from three specific measurement points on the Jishixia dam and compared with other dam deformation prediction models. At point ES-10, the performance evaluation indices of the IRF model were superior to those of the M5' model tree and RF models and the classical support vector regression (SVR) and back propagation (BP) neural network models, indicating the satisfactory performance of the IRF model. The IRF model also outperformed the SVR and BP models in settlement prediction at points ES2-8 and ES4-10, demonstrating its strong anti-interference and generalization capabilities. This study has developed a novel method for forecasting and analyzing dam settlements with practical significance. Moreover, the established IRF model can also provide guidance for modeling health monitoring of other structures.

An Approach for the Estimation of Concentrations of Soluble Compounds in E. coli Bioprocesses.

Accurate estimations of the concentrations of soluble compounds are crucial for optimizing bioprocesses involving Escherichia coli (E. coli). This study proposes a hybrid model structure that leverages off-gas analysis data and physiological parameters, including the average biomass age and specific growth rate, to estimate soluble compounds such as acetate and glutamate in fed-batch cultivations We used a hybrid recurrent neural network to establish the relationships between these parameters. To enhance the precision of the estimates, the model incorporates ensemble averaging and information gain. Ensemble averaging combines varying model inputs, leading to more robust representations of the underlying dynamics in E. coli bioprocesses. Our hybrid model estimates acetates with 1% and 8% system precision using data from the first site and the second site at GSK plc, respectively. Using the data from the second site, the precision of the approach for other solutes was as fallows: isoleucine -8%, lactate and glutamate -9%, and a 13% error for glutamine., These results, demonstrate its practical potential.

Dynamic In Silico Model of Type 2 Diabetes Treated With Metformin Combined With Exercise: A Sobol Sensitivity Analysis

Background: The widespread adoption of a sedentary lifestyle characterized by the consumption of a high-energy diet combined with a lack of physical activity has resulted in a rise in the prevalence of metabolic disorders like type 2 diabetes (T2D) globally. The data published by the International Diabetes Federation (IDF) every year catches the attention of the world. Insulin resistance results in the dysregulation of the control mechanism that leads to T2D. In addition to metformin as a globally prescribed drug, physical exercise is recommended for the treatment of T2D as an alternative therapy. In today’s systems biology era, sensitivity analysis plays a major role in providing deeper insights into the metabolic insulin signalling pathways (ISPs). Methods: Here, the complex in silico dynamic model of the effect of alternative therapies on T2D was examined. In order to obtain more quantitative information about the signal transduction network of the ISPs and their interaction, a global sensitivity analysis of the in silico model was carried out using the MATLAB tool. Results: This study sheds light on the effect of parameter perturbation on the ISP via GLUT4 translocation in T2D patients performing physical exercise and using metformin. It also allows for the selection of a variety of parameters for in vivo or in vitro studies in the future on the basis of the impact of parameter variations on each component of the pathway. Conclusion: The Sobol index for each condition was noted after introducing perturbations in the most sensitive parameters. These results can provide experimental guidance on how the variations in model inputs have affected the model outputs.

Predictive spatial analysis for a critical assessment of the preservation potential of Palaeolithic record in the Leiria region (central Portugal)

Predictive spatial analysis in archaeology refers to the use of geostatistical and computational techniques to assess the relationships between the distribution of archaeological sites and the environmental context. It is increasingly becoming an essential tool in scientific-academic research and rescue archaeology. In the interconnection between these two working contexts, this study aims to explore the advantages of applying an inductive approach, combining Geographical Information Systems (GIS) and geostatistical methods, to identify areas with different degrees of probability for the preservation of the Palaeolithic record in the lower valley of the Lis River (Leiria region, central Portugal). Topographic, hydrographic, geological, and geomorphological data were collected, processed, and integrated into a spatial database, using a GIS, to define the environmental conditioning factors. After applying a spatial multicollinearity procedure, slope, topographical position index, cost-weighted distance to access the watercourses, heat load index and lithology were chosen as input variables for the predictive modelling. Archaeological data, corresponding to the location of known Palaeolithic sites, were also retrieved from the official records. Three models were implemented using regression and multi-criteria methods; for Model 1 a Generalised Additive Model (GAM) was applied, while for Model 2 and Model 3 two distinct combinations of Probabilistic Frequency Ratio (PFR) and Analytic Hierarchy Process (AHP) were used. The spatial cross-tabulation between the obtained predictive maps and the location of the known Palaeolithic sites shows that all the models are acceptable. However, the combinations of PFR + AHP seem to evidence greater efficacy compared to the GAM. Model 3 offers the best performance (90%), followed by Model 2 (87%) and Model 1 (68%). Possible sources of inaccuracies in the input data and the spatial modelling results were critically discussed. Model 3 generated the most robust and reliable predictive prototype map. The maps produced can serve local and regional agencies as a complement and aid decision-making in public policies for spatial planning, and in the conservation/management of cultural heritage. These cartographic representations can provide quicker identification of areas of interest, for example, in the context of rescue archaeology, along with a substantial reduction in time and costs for project promoters, as well as helping in the environmental impact assessments.

The AMG model coupled with Rock-Eval® analysis accurately predicts cropland soil organic carbon dynamics in the Tuojiang River Basin, Southwest China

Accurate soil organic carbon models are key to understand the mechanisms governing carbon sequestration in soil and to help develop targeted management strategies to carbon budget. The accuracy and reliability of soil organic carbon (SOC) models remains strongly limited by incorrect initialization of the conceptual kinetic pools and lack of stringent model evaluation using time-series datasets. Notably, due to legacy effects of management and land use change, the traditional spin-up approach for initial allocation of SOC among kinetic pools can bring substantial uncertainties in predicting the evolution of SOC stocks. The AMG model can fulfill these conditions as it is a parsimonious yet accurate SOC model using widely-available input data. In this study, we first evaluated the performance of AMGv2 before and after optimizing the potential mineralization rate (k0) of SOC stock following a leave-one-site-out cross-validation based on 24 long-term field experiments (LTEs) in the Southwest of China. Then, we used Rock-Eval® thermal analysis results as input variables in the PARTYSOC machine learning model to estimate the initial stable SOC fraction (CS/C0) for the 14 LTEs where soil samples were available. The results showed that initializing the CS/C0 ratio using PARTYSOC combined with the optimized k0 further improved the accuracy of model simulations (R2 = 0.87, RMSE = 0.25, d = 0.90). Combining average measured CS/C0 and k0 optimization across all 24 LTEs also improved the model predictive capability by 25% compared to using default parameterization, thus suggesting promising avenue for upscaling model applications at the regional level where only a few measurement data on SOC stability can be available. In conclusion, the new version of the AMG model developed in the Tuojiang River Basin context exhibits excellent performance. This result paves the way for further calibration and validation of the AMG model in a wider set of contexts, with the potential to significantly improve confidence in SOC predictions in croplands over regional scales.

Seismic fragility analysis of steel moment frames using machine learning models

This study develops machine learning (ML) models for seismic fragility analysis of steel moment frames. Four ML methods – random forest, adaptive boosting, gradient boosting regression tree (GBRT), and extreme gradient boosting (XGBoost) – were employed for this purpose. Probabilistic seismic demand models, each representing the relationship between the seismic response of a type of structure and ground motion intensity, were used to construct the fragility curves based on nonlinear time history analyses of 616 steel moment frames subjected to 240 ground motions. The first three natural periods of steel moment frames and a capacity limit state defined by the maximum interstory drift were selected as input variables for the ML models. Two parameters (median and logarithmic standard deviation) of a fragility function were considered as output variables for the ML models. For each steel frame, the capacity limit state values considered for maximum interstory drift cover a wide range to generalize the fragility curve outcomes. The interquartile range method was used to ensure the quality of the dataset, and consequently 56,479 data points were used for the development of ML models. Based on model performance, the GBRT (R2 = 0.9986, for the testing dataset) and XGBoost (R2 = 0.9987) models are proposed as the best models for fragility analysis of steel moment frames. Finally, a graphical user interface for fragility analysis of steel moment frames was built based on the two proposed models, for easy access by practicing engineers. This study demonstrates the applicability of ML methods in practical design.

Determination of Input variables for Artificial Intelligence Models to predict the High PM2.5 concentration events in Seoul, Korea

Determination of Input variables for Artificial Intelligence Models to predict the High PM2.5 concentration events in Seoul, Korea

Prediction of interference current of buried pipeline and study on corrosion of Q235A steel

Interference current will cause severe electrochemical corrosion to buried pipes. However, the interference current flowing into the buried pipeline cannot be directly measured, and the influence of interference current on the corrosion of the buried pipeline cannot be directly observed. In this paper, the combined analysis methods of finite element analysis, partial least square method and electrochemical measurement technique are used to predict the interference current of buried pipeline and to study the effect of dynamic interference current and chloride ion concentration on the corrosion of Q235A steel. The prediction results show that increasing the input variables of the prediction model and changing the position of the input variables can effectively improve the accuracy of the model. Electrochemical measurements show that an increase in chloride ion concentration increases the corrosion current density in the polarization curve of Q235A steel and decreases the capacitive reactance arc radius in the electrochemical impedance spectroscopy. In addition, on the basis of the rail-to-ground potential monitoring points in the actual project, the number of rail-to-ground potential monitoring points was increased to improve the prediction accuracy of the buried pipeline interference currents, which indicates that the analysis method is an effective way to monitor and evaluate the influence of stray currents on the corrosion of buried pipelines.

Alighting and boarding time prediction in different types of waiting areas at subway stations

The alighting and boarding behavior of passengers at the subway station can directly affect the dwell time of trains, and then affect the service level of stations. The HPO-BP prediction model for the alighting and boarding time in scenarios with different types of waiting areas is proposed in this paper. The number of passengers, gender, whether to carry luggage or children, and the degree of passenger civilization are selected as the main influence factors after analyzing the data of field investigations, which are defined as the input variables of the prediction model. The alighting and boarding time is defined as the output variable. The HPO-BP prediction model is trained, tested and validated by using the data collected in the field investigations. The comparison results indicate that the prediction accuracies of the proposed model in a single waiting area and double waiting areas are 35.8% and 62.8% higher than that of the BP neural network, and are also significantly improved compared with other advanced algorithms. It can be concluded that the gender of passengers is the most important factor affecting passengers’ alighting and boarding time after the interpretation and analysis of the SHAP model. Moreover, the degree of passenger civilization also has a certain impact on the alighting and boarding efficiency.

New data‐driven method for estimation of net ecosystem carbon exchange at meteorological stations effectively increases the global carbon flux data

Abstract The eddy covariance (EC) flux stations have great limitations in the evaluation of the global net ecosystem carbon exchange (NEE) and in the uncertainty reduction due to their sparse and uneven distribution and spatial representation. If the EC stations are linked with widely distributed meteorological stations using machine learning (ML) and remote sensing, it will play a big role in effectively improving the accuracy of the global NEE assessment and reducing uncertainty. In this study, we developed a framework for estimating NEE at meteorological stations. We first optimized the hyperparameters and input variables of the ML model based on the optimization method called an adaptive genetic algorithm. Then, we developed 566 random forest (RF)‐based NEE estimation models by the strategy of spatial leave‐out‐one cross‐validation. We innovatively established the Euclidean distance‐based accuracy projection algorithm of the R square (R2), which could test the accuracy of each model to estimate the NEE of the specific flux at the weather station. Only the model with the highest R2 was selected from the models with a prediction accuracy of R2 > 0.5 for the specific meteorological stations to estimate its NEE. 4674 out of 10,289 weather stations around the world might match at least one of the 566 NEE estimation models with a projected accuracy of R2 > 0.5. The NEE estimation models we screened for the meteorological stations showed a reliable performance and a higher accuracy than the former studies. The NEE values of the most (96.9%) screened meteorological stations around the world are negative (carbon sink) and most (65.3%) of those showed an increasing trend in the mean annual NEE (carbon sink). The NEE dataset produced at the meteorological stations could be used as a supplement to the EC observations and quasi‐observation data to assess the NEE products of the global grid. The NEE dataset is publicly available via the figshare with https://doi.org/10.6084/m9.figshare.20485563.v1.

Predictive Modeling of Delay in an LTE Network by Optimizing the Number of Predictors Using Dimensionality Reduction Techniques

Delay in data transmission is one of the key performance indicators (KPIs) of a network. The planning and design value of delay in network management is of crucial importance for the optimal allocation of network resources and their performance focuses. To create optimal solutions, predictive models, which are currently most often based on machine learning (ML), are used. This paper aims to investigate the training, testing and selection of the best predictive delay model for a VoIP service in a Long Term Evolution (LTE) network using three ML techniques: Multilayer Perceptron (MLP), Support Vector Machines (SVM) and k-Nearest Neighbors (k-NN). The space of model input variables is optimized by dimensionality reduction techniques: RReliefF algorithm, Backward selection via the recursive feature elimination algorithm and the Pareto 80/20 rule. A three-segment road in the geo-space between the cities of Banja Luka (BL) and Doboj (Db) in the Republic of Srpska (RS), Bosnia and Herzegovina (BiH), covered by the cellular network (LTE) of the M:tel BL operator was chosen for the case study. The results show that the k-NN model has been selected as the best solution in all three optimization approaches. For the RReliefF optimization algorithm, the best model has six inputs and the minimum relative error (RE) RE = 0.109. For the Backward selection via the recursive feature elimination algorithm, the best model has four inputs and RE = 0.041. Finally, for the Pareto 80/20 rule, the best model has 11 inputs and RE = 0.049. The comparative analysis of the results concludes that, according to observed criteria for the selection of the final model, the best solution is an approach to optimizing the number of predictors based on the Backward selection via the recursive feature elimination algorithm.

Identification of Water Layer Presence in Paddy Fields Using UAV-Based Visible and Thermal Infrared Imagery

The accurate identification of the water layer condition of paddy fields is a prerequisite for precise water management of paddy fields, which is important for the water-saving irrigation of rice. Until now, the study of unmanned aerial vehicle (UAV) remote sensing data to monitor the moisture condition of field crops has mostly focused on dry crops, and research on the water status of paddy fields has been relatively limited. In this study, visible and thermal infrared images of paddy fields at key growth stages were acquired using a UAV remote sensing platform, and three model input variables were constructed by extracting the color features and temperature features of each field, while K-nearest neighbor (KNN), support vector machine (SVM), random forest (RF), and logistic regression (LR) analysis methods were applied to establish a model for identifying the water layer presence in paddy fields. The results showed that KNN, SVM, and RF performed well in recognizing the presence of water layers in paddy fields; KNN had the best recognition accuracy (89.29%) via algorithm comparison and parameter preference. In terms of model input variables, using multisource remote sensing data led to better results than using thermal or visible images alone, and thermal data was more effective than visible data for identifying the water layer status of rice fields. This study provides a new paradigm for monitoring the water status of rice fields, which will be key to the precision irrigation of paddy fields in large regions in the future.

Polymer reaction engineering meets explainable machine learning

Due to the complex polymerization technique and statistical composition of the polymer, tailoring its characteristics is a challenging task. Modeling of the polymerizations can contribute to deeper insights into the process. This study applies state-of-the-art machine learning (ML) methods for modeling and reverse engineering of polymerization processes. ML methods (random forest, XGBoost and CatBoost) are trained on data sets generated by an in house developed kinetic Monte Carlo simulator. The applied ML models predict monomer concentration, average molar masses and full molar mass distributions with excellent accuracy (R2> 0.96). Reverse engineering results delivering the polymerization recipe for a targeted molar mass distribution are less accurate, but still only minor deviations from the targeted molar mass distribution are seen. The influences of the input variables in ML models obtained by explainability methods correspond to the expert expectations.

Soil and water conservation effects of different types of vegetation cover on runoff and erosion driven by climate and underlying surface conditions

Quantitative analyses of runoff and erosion under specific climatic (precipitation characteristics) and underlying surface conditions (slope, antecedent soil moisture) with different types of vegetation cover (or plant measures hereafter) remains a challenging research topic in semi-arid regions. The research site located in the Anjiagou watershed of Gansu Province, China, and used for this study has long-term monitoring data (1987 to 2019) of precipitation conditions, antecedent soil moisture, and runoff and erosion rates from a set of modified standard erosion plots characterized by different vegetation covers or plant measures (arbor forest land, shrub forest land, artificial grassland, natural grassland, and agricultural land) at three slopes (10°, 15°, 20°). Collected data was subjected to path analysis and stepwise regression to determine the main driving factors of runoff and soil erosion, as well as proper input variables for modeling. The results showed that models could be constructed with different variables for different types of vegetation cover (P < 0.05, ra2 > 0.55). The analyses also showed that moderate rain (10 – 25 mm) accounted for >60% of the variation in runoff and plant effects on runoff and soil erosion varied with the measures. Among plant measures, shrub forest had the best performance in reducing runoff and erosion for short-term precipitation events (shorter than 24 hrs), but under the long-durations (72 hrs) precipitation, natural grassland performed the best to prevent the occurrence of runoff and soil erosion. Compared with agricultural land, the rest of the plant measures had an average erosion reduction rate of 73%, which was about twice the runoff reduction rate (37%). Slope was found to play some role in controlling runoff and erosion variation. Compared to slopes of 10° and 20°, runoff and erosion reduction rates at 15° plots were lower. The key variables determining runoff rates under different plant measures included precipitation and maximum 30-minute precipitation intensity, followed by types of vegetation cover. In contrast, erosion rates mostly depended upon runoff and maximum 30-min precipitation intensity. We also found that coupling precipitation characteristics with antecedent soil moisture in the modelling could help produce a more accurate assessment of runoff and soil and water conservation as part of ecological restoration efforts involving typical plant measures under different topographical conditions in semi-arid areas.

Data-driven optimization of pollutant emission and operational efficiency for circulating fluidized bed unit

To achieve the economic and environmentally friendly operation of circulating fluidized bed (CFB) units, it is imperative to conduct optimization to obtain an economical mode of pollutant removal. This article focuses on the multi-objective optimization between SO2-NOx emissions and thermal efficiency for CFB units. According to the operation data and production mechanism of pollutants, models of SO2-NOx emission concentration, bed temperature, and oxygen content based on a convolutional neural network–bidirectional long short-term memory–attention mechanism (CNN-BiLSTM-Attention) were established. Then, an improved quantum genetic algorithm was used to find the optimal input variables of the SO2-NOx emission model. The proposed modeling method was evaluated, and it more accurately simulated the trends of actual operation data than other models under different operating conditions. Combining these models with economic calculations, the operating costs under typical conditions were reduced by 3.20% and 1.82% respectively, and the thermal efficiency increased by 0.72% and 1.07%, which contributes to economical and intelligent operation of the unit.

Linear mixed model to identify the relationship between grain yield and other yield related traits and genotype selection for sorghum

Sorghum is the most popular crop in arid and semi-arid areas, especially in Sub-Saharan African countries. Genotype effects, environmental and the interaction of genotype by environmental factors have an influence on phenotypic traits. The aim of the study is to identify the relationship between grain yield and other yield-related traits and select the genotypes which perform better in grain yield as well as to examine the association between the uncorrelated phenotypic traits and grain yield via mixed model. The data was generated using a lattice square design. Principal component analysis was used to generate uncorrelated variables for the mixed model. The study revealed that there was a difference in grain yield due to the treatment and there was a pairwise relationship among the phenotypic variables. 77.12% of the total variance of the original phenotypic variables was explained by the first three principal components and decided to use PCAs as input variables for the mixed model. All PCs had significant effects on grain yield as well as grain yield variability due to random effects associated with genotypes, genotype interaction by treatment, and replication within the treatment. The variability of grain yield due to genotype effect was explained about 45.73%, the variation of grain yield due to the interaction of genotype by the treatment was also explained about 39.06% and 1.55% of the variation of grain was explained by replication within treatment. The best performer genotypes recommended for mass production were G40 (Genotype 40), G186 (Genotype 186) and G196 (Genotype 196) without any constraint of environment. The genotypes recommended for mass production under irrigation conditions were G40 (Genotype 40), G62 (Genotype 62) and G192 (Genotype 192). G26 (Genotype 26), G55 (Genotype 55) and G49 (Genotype 49) were the genotypes recommended for mass production under stress conditions. Overall, the study recommends using a mixed model to fit the grain yield, and future work will focus on to evaluate the performance of genotypes under different environments and years of production.

Prediction of Compressive Strength of Self-Compacting Concrete using Machine Learning Techniques

The paper deals with the use of Deep Neural Networks (DNN), Artificial Neural Network (ANN), and Random Forest (RF) for estimating the 28- day compressive strength of self-compacting concrete (SCC) containing silica and filler (fly ash, marble powder, and lime powder) with a comparative performance analysis of all techniques. A total of 179 data were taken from literature already published with eight input variables for modelling. The evaluation and comparison of the performance of predicted models were made using the same datasets in training and testing based on correlation coefficient (CC), Coefficient of determination (R2), root mean square error (RMSE), and mean absolute error (MAE). The results showed that proposed model's performance could be improved when training takes place in a Deep Neural Network model with multiple hidden layers. Sensitivity analysis was used to quantify the effect of different variables on concrete strength with coarse aggregate greatly affecting the compressive strength of SCC, followed by fine aggregate content and quantity of silica. A dependable prediction tool is provided through this investigation which suggests that the present model can help scientists and engineers in the optimization of the mixture design of SCC.