Artificial Neutral Network Research Articles

Accelerated discovery of eutectic compositionally complex alloys by generative machine learning

Eutectic alloys have garnered significant attention due to their promising mechanical and physical properties, as well as their technological relevance. However, the discovery of eutectic compositionally complex alloys (ECCAs) (e.g. high entropy eutectic alloys) remains a formidable challenge in the vast and intricate compositional space, primarily due to the absence of readily available phase diagrams. To address this issue, we have developed an explainable machine learning (ML) framework that integrates conditional variational autoencoder (CVAE) and artificial neutral network (ANN) models, enabling direct generation of ECCAs. To overcome the prevalent problem of data imbalance encountered in data-driven ECCA design, we have incorporated thermodynamics-derived data descriptors and employed K-means clustering methods for effective data pre-processing. Leveraging our ML framework, we have successfully discovered dual- or even tri-phased ECCAs, spanning from quaternary to senary alloy systems, which have not been previously reported in the literature. These findings hold great promise and indicate that our ML framework can play a pivotal role in accelerating the discovery of technologically significant ECCAs.

Prediction and optimizing of methylene blue sequestration to activated charcoal/magnetic nanocomposites using artificial neutral network and response surface methodology

Green synthesized magnetic nanoparticles (MNPs) linked with activated charcoal (AC) (AC/Fe3O4 NCs) were exploited for methylene blue (MB) confiscation in this study. The AC/Fe3O4 NCs produced were characterized using TEM, FTIR, UV/Vis and XRD spectrometry. The Response-Surface-Methodology (RSM) was utilized to improve the experimental data for the MB sorption to AC/Fe3O4 NCs, with 20 experimental runs implemented through a central composite design (CCD) to assess the effect of sorption factors-initial MB concentration, pH and sorbent dosage effects on the response (removal-effectiveness). The quadratic model was discovered to ideally describe the sorption process, with an R2 value of 0.9857. The theoretical prediction of the experimental data using the Artificial-Neural-Network (ANN) model showed that the Levenberg-Marquardt (LM) had a better performance criterion. Comparison between the modelled experimental and predicted data showed also that the LM algorithm had a high R2 of 0.9922, which showed NN model applicability for defining the sorption of MB to AC/Fe3O4 NCs with practical precision. The results of the non-linear fitting (NLF) of both isotherm and kinetic models, showed that the sorption of MB to AC/Fe3O4 NCs was perfectly described using the pseudo-second-order (PSOM) and Freundlich (FRHM) models. The estimated optimum sorption capacity was 455 mg g−1. Thermodynamically, the sorption of MB to AC/Fe3O4 NCs was shown to be non-spontaneous and endothermic.

Relative importance of global warming, the IPO, and the AMO in surface air temperature and terrestrial precipitation

AbstractThe relative importance of oceanic modes to global surface air temperature (SAT) and terrestrial precipitation during 1934–2020 was investigated using a variety of statistical and dynamical system methods. Through singular spectrum analysis, we present the distribution of decadal (10–20‐year), multidecadal (20–50‐year), and secular (>50‐year) variabilities of global SAT and terrestrial precipitation anomalies. Three sea surface temperature modes were identified by singular value decomposition that affect the low‐frequency variabilities of global SAT and terrestrial precipitation anomalies—namely, global warming (GW), the Interdecadal Pacific Oscillation (IPO), and the Atlantic Multidecadal Oscillation (AMO). The relative importance GW, the IPO, and the AMO in SAT and terrestrial precipitation was obtained through a normalized multivariate information flow analysis. Besides a high information flow percentage (60%) from GW to global SAT, especially over the northern Indian Ocean and tropical West Pacific, information flow from the AMO to the northwestern Pacific was also found. In terms of terrestrial precipitation, a large area with a wet trend was found over Eurasia at mid‐to‐high latitudes, and this trend was especially remarkable in the boreal winter half‐year (November–April), as compared with that in the boreal summer half‐year (May–October). By employing artificial neutral networks with a self‐organized map to cluster the key patterns of vertically integrated water vapour flux, we found that the synoptic circulation related to the wet trend is characterized by westerly flow that transports water vapour from the northeastern Atlantic to Eurasia, which is favourable for precipitation there both in the boreal winter and summer half‐year.

Development of Drilling Optimization Models for Autonomous Rotary Drilling Systems

The growing global energy demand and strict environmental policies, motivates the use of technology and performance improvement techniques in drilling operations. In the traditional drilling method, parameter optimization depends on the effectiveness of human-driller. Although existing work has identified the significance of upscaling from manual drilling to autonomous drilling system, but little has been done to support this transition. This work presents optimization models for an autonomous rotary drilling system, controlled by a self-tuning, multivariant controller that uses machine learning optimization strategy. The method determines the drilling medium from real-time measurement by estimating the unconfined compressive strength (UCS) from the latest data uploaded via the mud pulse telemetry (MPT) and adjust optimal setpoint based on model output. In the study, four machine learning algorithms were used to predict UCS including artificial neutral network (ANN), Category boast (CB), Support vector machine (SVR) and Randon Forest. Whilst Physics based empirical models with ANN were used to predict the drill rate. Results showed that machine learning (ML) application improves the prediction quality of drill rate and UCS with ANN and Catboast as best ML predictors. The coefficient of determination (R2) of 0.95 ROP prediction and (R2) for test dataset of 0.77 and 070 for UCS prediction using ANN and Catboast respectively. The Q-learning algorithm which uses the value function to search for optimal operating parameter at different Lithologies through dynamic programming, returns decisions for optimal drill rate at respective drilling states consequently improving the efficiency of rotary drilling process in terms of cost and time

Selection of artificial neutral networks based on cubic intuitionistic fuzzy Aczel-Alsina aggregation operators

<p>Artificial neural networks (ANNs) are the collection of computational techniques or models encouraged by the shape and purpose of natural or organic neural networks. Furthermore, a cubic intuitionistic fuzzy (CIF) set is the modified or extended form of a Fuzzy set (FS). Our goal was to address or compute the model of Aczel-Alsina operational laws under the consideration of the CIF set as well as Aczel-Alsina t-norm (AATN) and Aczel-Alsina t-conorm (AATCN), where the model of Algebraic norms and Drastic norms were the special parts of the Aczel-Alsina norms. Further, using the above invented operational laws, we aimed to develop the model of Aczel-Alsina average/geometric aggregation operators, called CIF Aczel-Alsina weighted averaging (CIFAAWA), CIF Aczel-Alsina ordered weighted averaging (CIFAAOWA), CIF Aczel-Alsina hybrid averaging (CIFAAHA), CIF Aczel-Alsina weighted geometric (CIFAAWG), CIF Aczel-Alsina ordered weighted geometric (CIFAAOWG), and CIF Aczel-Alsina hybrid geometric (CIFAAHG) operators with some well-known and desirable properties. Moreover, a procedure decision-making technique was presented for finding the best type of artificial neural networks with the help of multi-attribute decision-making (MADM) problems based on CIF aggregation information. Finally, we determined a numerical example for showing the rationality and advantages of the developed method by comparing their ranking values with the ranking values of many prevailing tools.</p>

Estimation of Solar Radiation Using Artificial Neutral Network for Different Climate Zones in Türkiye

Güneş enerjisinden gerek termal gerekse elektrik uygulamalarında elde edilen faydalı enerji yüzeye gelen ışınıma bağlıdır ve bu ışınım miktarı konum, zaman, iklimsel koşullar ve çevre koşullarına göre değişmektedir. Güneş enerji destekli enerji sistemlerinin etkin bir şekilde tasarlanması ve çalıştırılması için ışınım miktarının doğru bir şekilde belirlenmesi önem arz etmektedir. Bu çalışmada Türkiye’nin farklı iklim kuşaklarında yer alan dört il (Kayseri, Rize, Hakkari ve İzmir) için MATLAB uygulamasıyla Yapay Sinir Ağları (YSA) modeli oluşturulup güneş ışınımı tahmini yapılmıştır. Güneş ışınımı tahmini yapılırken illerin enlem, boylam ve rakımdan oluşan coğrafik verilerinin etkisinin yanı sıra ortalama sıcaklık, bağıl nem, bulutluluk ve buhar basıncı gibi meteorolojik verilerin etkisi de ele alınmıştır. Tüm bu etkenler göz önüne alındığında kurulan modelde en iyi tahminleme sonucu R2=0.93 değeriyle Hakkari ilinde olup sırasıyla İzmir, Kayseri ve Rize illeri takip etmektedir. Her bir il için YSA ile tahmin edilen değerler ile ölçülen değerler arasında uyum olduğu görülmüştür.

A Machine Learning Model for Determination of Gender Utilizing Hybrid Classifiers

One part of forensic anthropology involves investigating skeletal remains to identify corpses, and many of these remains were found incomplete, burned, broken, or destroyed, making investigation challenging. This study aims to use the pelvis and femur to identify the gender of skeletal remains. The pelvis and femur have previously been proven to be accurate indicators of a corpse's gender. The identification process is done through the measurement of the subpubic angle of the pelvis and the angle taken straight down from the top of the femur to the patella and then straight up. The two measurements were combined using the principal component analysis (PCA) method into two attributes on the x and y axes. These attributes were later used as data for the machine learning model design. The design process consisted of an Artificial Neutral Network (ANN) design model and Support Vector Machine (SVM) design model combined into a hybrid machine learning system. The ANN and SVM hybrid machine learning were tested with acquired data. The result of the test using the confusion matrix showed 83.33% accuracy, which is categorized as "good classification" based on Area Under the Curve (AUC).

Dynamic range enhancement for the sensing signals of peak-saturated fiber Bragg grating spectra.

Fiber Bragg grating (FBG) sensors applying time-delay interrogators with wavelength swept lasers (WSLs) are popular for their great potentials in high sensing resolution and power budget. In these systems, well-calibrated WSLs with reduced wavelength nonlinearity and jitter are critical for the sensing performance. However, high-performance WSLs are expensive and could significantly increase the cost of the systems. The overall cost may be reduced by maximally sharing each WSL with multiple sensing FBGs through mechanisms like power splitting, which distribute the WSL signal into multiple independently operated serial FBG chains. Under such scenarios, the sensing processing unit (SPU) of each serial FBG chain must be synchronized with the WSL for correctly estimating the FBGs' respective spectra from the signal return time delays. We previously propose a self-synchronized scheme relying on the dual-polarity spectrum signal, which reduces the synchronization labor. The dual-polarity signal has a wider dynamic range, which may limit the system response speed or accuracy, considering the amplifiers' responses or the analog-to-digital converters' (ADCs') quantization noise. In this Letter, we apply peak-saturated FBG spectra for the sensors to increase the receivers' equivalent dynamic range. The flattop waveforms of the saturated peaks result in uncertainty for the peak positions. An artificial neutral network (ANN)-based method is further studied to enhance the peak detection accuracy. We show an ∼88% receiver dynamic range improvement with an inaccuracy reduction of about a half compared to the filter-and-maximum-readout (FMR) method.

Prediction of Viscosity of Cobalt Ferrite/SAE50 Engine Oil based Nanofluids using well Trained Artificial Neutral Network (ANN) and Response Surface Methodology (RSM)

Heat transmission by ordinary fluids such as pure water, oil, and ethylene glycol is inefficient due to their low viscosity. To boost the efficiency of conventional fluids, very small percent of nanoparticles are added to the base fluids to prepare nanofluid. The impact of changing in viscosity can be used to investigate the rheological properties of nanofluids. In this paper, (CoFe2O4)/engine oil based nanofluids were prepared using two steps standard methodology. In first step, CoFe2O4 (CF) were synthesized using the sol-gel wet chemical process. The crystalline structure and morphology were confirmed using X-Ray diffraction analysis (XRD) and scanning electron microscopy (SEM), respectively. In second step, the standard procedure was adapted by taking several solid volume fractions of CF as Ø = 0, 0.25, 0.50, 0.75, and 1.0 %. Such percent of concentrations were dispersed in appropriate volume of engine oil using the ultrasonication for 5 h. After date, the viscosity of prepared five different nanofluids were determined at temperatures ranging from 40 to 80 °C. According to the findings, the viscosity of nanofluids (µnf) decreased as temperature increased while increased when the volume percentage of nanofluids Ø raised. Furthermore, total 25 experimental observations were considered to predict viscosity using an artificial neural network (ANN) and response surface methodology (RSM). The algorithm for building the ideal ANN architecture has been recommended in order to predict the fluid velocity of the CF/SAE-50 oil based nanofluid using MATLAB software. In order to determine the correctness of the predicted model, the mean square error (MSE) was calculated 0.0136.

Optimal power control in Renewable Energy sources using Intelligence algorithm

An intelligent, Adaptive Neuro-Fuzzy Inference System (ANFIS) based Xilinx Power Management system is proposed for the Hybrid renewable energy sources (HRES). The Xilinx Power Management generator block set (ANFIS-XL-PMS) controller is implemented for the purpose to control and avoid the non-linear nature opted in the HRES power system. Solar (PV), Wind Energy (WE), and Fuel cell (FC) are the primary power sources of the system, and a battery storage system (BES) is used as a backup. A comprehensive power management strategy is intended for the proposed system to control the flow of power between the various energy sources and the system's storage device. The simulation model for the HRES is elegantly designed using the MATLAB/Simulink platform and the Switching patterns are generated using the XLPMS controller. The system performance and output responses of the SMC-XL-ANFIS controller have been verified by carrying out the simulation studies and compared with the existing controller like Sliding Mode Controller (SMC-XL-PMS), and Artificial Neutral Network (ANN-XL-PMS). Simulation results show good performance in terms of voltage and current transient, settling time, load power efficiency, and power quality Also a Prototype model with a PIC microcontroller is designed and the output responses are analyzed.

Flood hazard assessment in Yemen using a novel hybrid approach of Grey Wolf and Levenberg Marquardt optimizers

This study aims to map flood susceptibility in the Qaa’Jahran watersheds located in Dhamar, Yemen, using geoprocessing and computational techniques. Historical flood data and SAR imagery were used to monitor and create a flood inventory map. The artificial neutral network (ANN) was trained using a novel algorithm called GWO_LM, which is a hybridization between the Levenberg-Marquardt algorithm (LM) and Grey Wolf Optimizer (GWO) meta-heuristic algorithm and compared the results with state of art machine learning algorithms. The GWO_LM_ANN model exhibited excellent performance in the evaluation, achieving a precision of 97.92%, sensitivity of 100%, specificity of 100%, F1 score of 98.95%, accuracy of 98.75% and AUC of 98.48. This indicates that using GWO_LM for training ANN enhanced the searching process for the optimal weights, resulting in outperforming other state-of-the-art models. The findings hold significant implications for disaster preparedness and response in the Qaa’Jahran watersheds, enabling targeted and efficient non-structural solutions to mitigate the detrimental effects of flash floods in particularly sensitive locations. The use of the previously unexplored GWO_LM model represents a notable advancement in flood susceptibility assessment, surpassing traditional methods and offering novel insights to the existing literature.

Post-blast capacity evaluation of concrete-filled steel tubular (CFST) column based on machine learning technique

In the present study, the post-blast axial load-bearing capacity of the CFST column under the close-in explosion is investigated. Dynamic response and residual axial load capacity (RALBC) of 1599 circular CFST columns are numerically studied using a validated finite element model. The effects of column dimensions, material properties, blast loading, and axial load ratio are discussed. Based on the huge database, prediction models of RALBC are developed using three commonly used machine learning techniques including Extreme Gradient Boosting (known as XGBoost), Artificial Neutral Network (ANN), and Support Vector Regression (SVR). Comparison results show that the prediction model generated by the XGBoost performs the best among the three methods followed by the ANN and SVR methods. The RALBC of the CFST columns can be estimated rapidly and accurately with the prediction model developed in the present study. Finally, the feature importance of the input variables is analyzed using the additive feature attribution method SHAP (Shapley Additive ExPlanation). The results show that geometric parameters are the main factors impacting the RALBC of the CFST columns under close-in explosion.

Use of Biometric Images to Predict Body Weight and Hot Carcass Weight of Nellore Cattle.

The objective of this study was to evaluate different methods of predicting body weight (BW) and hot carcass weight (HCW) from biometric measurements obtained through three-dimensional images of Nellore cattle. We collected BW and HCW of 1350 male Nellore cattle (bulls and steers) from four different experiments. Three-dimensional images of each animal were obtained using the Kinect® model 1473 sensor (Microsoft Corporation, Redmond, WA, USA). Models were compared based on root mean square error estimation and concordance correlation coefficient. The predictive quality of the approaches used multiple linear regression (MLR); least absolute shrinkage and selection operator (LASSO); partial least square (PLS), and artificial neutral network (ANN) and was affected not only by the conditions (set) but also by the objective (BW vs. HCW). The most stable for BW was the ANN (Set 1: RMSEP = 19.68; CCC = 0.73; Set 2: RMSEP = 27.22; CCC = 0.66; Set 3: RMSEP = 27.23; CCC = 0.70; Set 4: RMSEP = 33.74; CCC = 0.74), which showed predictive quality regardless of the set analyzed. However, when evaluating predictive quality for HCW, the models obtained by LASSO and PLS showed greater quality over the different sets. Overall, the use of three-dimensional images was able to predict BW and HCW in Nellore cattle.

Real-time monitoring and prediction method of commercial building fire temperature field based on distributed optical fiber sensor temperature measurement system

In order to accomplish the task of real-time monitoring and prediction of fire temperature in commercial buildings, this study proposes a sensor arrangement method based on sensor and fire field characteristics to better reflect the fire temperature changes of building space, and conducts high temperature sensor measurement experiments based on this method. Then the study carries out fire dynamics simulations to establish a temperature change dataset containing fire and sensor temperature changes. The study establishes several prediction models using different algorithms based on the dataset, and the models built by artificial neutral network (ANN) and long short-term memory (LSTM) respectively have the best performance. The results show that the percentage error between the output temperature of the ANN mapping model and the actual temperature of the fire key plane is 1.94%. The percentage error between the output temperature of the LSTM temperature prediction model and the actual temperature is 1.03%, which satisfies the demand for early warning of the fire temperature hazard. This study also validates the generalization of the two models for fire temperature prediction in different spatial areas, and the method can provide a good solution for real-time monitoring and early warning of fire temperature in commercial buildings.

Development of a Prediction Model for Daily PM2.5 in Republic of Korea by Using an Artificial Neutral Network

This study aims to develop PM2.5 prediction models using air pollutant data (PM10, NO2, SO2, O3, CO, and PM2.5) and meteorological data (temperature, humidity, wind speed, atmospheric pressure, precipitation, and snowfall) measured in South Korea from 2015 to 2019. Two prediction models were developed using an artificial neural network (ANN): a nationwide (NW) model and administrative districts (AD) model. To develop the prediction models, the independent variables daily averages and variances of air pollutant data and meteorological data (independent variables) were used as independent variables, and daily average PM2.5 concentration set as a dependent variable. First, the correlations between independent and dependent variables were analyzed. Second, prediction models were developed using an ANN to predict next-day PM2.5 daily average concentration, both NW and in 16 AD. The ANN models were optimized using a factorial design to determine the hidden layer layout and threshold, and a seasonal (monthly) factor was also considered. In the optimal prediction model, the absolute error in 1 σ was 91% (in-sample 91%, out-of-sample 91%) for the NW model, and the absolute error in 1 σ was 86% (in-sample 88%, out-of-sample 84%) for AD model. The accuracy of these prediction models increases further when they are developed using the next-day weather data, assuming that the weather prediction is accurate.

Machine learning based charge mobility prediction for organic semiconductors.

Transfer integral is a crucial parameter that determines the charge mobility of organic semiconductors, and it is very sensitive to molecular packing motifs. The quantum chemical calculation of transfer integrals for all the molecular pairs in organic materials is usually an unaffordable task; fortunately, it can be accelerated by the data-driven machine learning method now. In this work, we develop machine learning models based on artificial neutral networks to predict transfer integrals accurately and efficiently for four typical organic semiconductor molecules: quadruple thiophene (QT), pentacene, rubrene, and dinaphtho[2,3-b:2',3'-f]thieno[3,2-b]thiophene (DNTT). We test various forms of features and labels and evaluate the accuracy of different models. With the implementation of a data augmentation scheme, we have achieved a very high accuracy with the determination coefficient of 0.97 and mean absolute error of 4.5meV for QT, and similar accuracy for the other three molecules. We apply these models to studying charge transport in organic crystals with dynamic disorders at 300K and obtain the charge mobility and anisotropy in perfect agreement with the brutal force quantum chemical calculation. If more molecular packings representing the amorphous phase of organic solids are supplemented to the dataset, the current models can be refined to study charge transport in organic thin films with polymorphs and static disorders.

Two-stage vibration-suppression framework for optimal robust placements design and reliable PID gains design via set-crossing theory and artificial neural network

• Atwo-stage controlmodelfor design of placements and PID parametersis proposed. • The optimal placementsrobust design of uncertain vibration systemsis proposed. • The optimal PID gainsdesign withtime-dependent reliabilitylimitsis proposed. • The ANNis appliedin vibratory controloptimization to improve the efficiency. Structure vibration control is a significant problem in practical engineering, and Proportional-Integral-Differential (abbreviated as “PID”) control is widely used. The design of a PID controller contains the robust design of placements of actuators and sensors (abbreviated as “ASPD”), and the reliable design of PID gains. Because of the uncertain parameters in practical engineering, the controller designed from the deterministic system has the possibility of failure when applied to a practical system. To solve this problem, this paper proposed a two-stage framework for uncertain vibration active control systems via non-probabilistic time-dependent reliability (abbreviated as “NTDR”) and artificial neutral network (abbreviated as “ANN”). This framework decouples the ASPD and PID gains design as two stages. Considering the insufficient sample data, uncertain parameters are quantified with non-probabilistic interval forms. The optimization objective of ASPD combines the deterministic eigenvalues of controllability gramian or observability gramian, the robust theory and placements constraints to carry out the optimization. In this way the optimization objective can be less affected by uncertain parameters and has practical significance. For the reliable design of PID gains, the NTDR is adapted to evaluate the safety of an uncertain system. Then ANNs from uncertain parameters and PID gains to the optimization objective considering NTDR are established. By carrying out the optimization to the last ANN, the efficiency of optimization is improved while the accuracy is guaranteed. Two numerical examples and one experimental example show the practicability and effectiveness of the proposed method.

Degradation of PET Nanoplastic Oligomers at the Novel PHL7 Target:Insights from Molecular Docking and Machine Learning

The versatility of Polyethylene terephthalate (PET) as a material with numerous applications in the food industry and its recalcitrance to chemical and microbial degradation has recently made it an environmental nuisance. In this study, we applied computational methods to ascertain the dependence of PET nanoplastic (NP) degradation on the chain length of the oligomer. The binding affinities of the NPs on the novel enzyme Polyester Hydrolase Leipzig 7 (PHL7) were used to relate their ease of degradation at the enzyme active site. The results revealed that the binding affinity of PET NPs at the enzyme target decreased from -5.2 kcal/mol to -0.8 kcal/mol, with an increase in PET chain length from 2.18 nm to 5.45 nm (2-5 PET chains). The binding affinities became positive at chain lengths 6.54 nm (6 PET chains) and above. These findings indicated that PET NP degradation at this enzyme’s active site is most efficient as chain length decreases from 5-2 units and is not likely to occur at longer PET chains. A feedforward Artificial Neutral Network (ANN) analysis predicted that the energy of the PET NPs is a very important factor in its degradation.

The importance of data splitting in combined NOx concentration modelling

The polluted air breathed every day by those living in large conurbations poses a significant risk to their health. Through effective modelling (prediction) of concentrations of pollutants and identification of the factors influencing them, it should be possible to obtain advance information on dangers and to plan and implement measures to reduce them. This work describes two different modelling approaches: based on the NOx concentration of the previous hour (C&RT models); and based on meteorological factors, traffic flow, and past (up to two previous hours) NOx and NO2 concentrations (CA models). For each approach, three alternative machine learning methods were applied: artificial neutral network (ANN), random forest (RF), and support vector regression (SVR). The best fits were obtained for the models using ANN and RF (MAPE values in the range 18.3–18.5 %). Poorer fits were found for the SVR models (MAPE equal to 23.4 % for the C&RT approach and 29.3 % for CA). No significant preferences were identified between the C&RT and CA approaches (based on various goodness-of-fit measures). The choice should be determined by the purposes for which the forecast is to be used.

Application of ANN and RSM Techniques in Optimal Parameter Evaluation for Turbidity Removal from Abattoir Effluent using Valorized Chicken Bone Coagulant

Chicken bone coagulant (CBC) containing high grade hydroxyapatite (HPA) has been applied in the coag-flocculation of abattoir effluent. The influence of process variables (pH, initial concentration, dosage, Temperature, and settling time) on the effluent final turbidity was investigated. Also, the accuracies of two modelling techniques (Response surface methodology, RSM and Artificial neutral network, ANN) in predicting the non-linear nature of the system were compared. The optimization result indicates a final turbidity of 4.96 mg/L (corresponding to 98.28 % removal efficiency) at pH = 6.7, dosage = 1.003 g/L, initial conc. = 182.2 mg/L, coagulation temp. = 345 K and settling time of 36 min. Meanwhile, effluent pH was spotted as the most significant variable, with p-value <0.01%. Furthermore, the error analysis result portrayed the supremacy of ANN over RSM in data prediction accuracy as it signified lower error values (Mean square error, MSE = 13.11 and Absolute average relative deviation, AARD = 1.43%) when compared to those of RSM (MSE = 37.78, AARD = 5.93%). Thus, it was demonstrated that ANN is a better tool for optimization study of the present system.