Artificial intelligence simulation of suspended sediment load with different membership functions of ANFIS

Prediction of suspended sediment load in Sungai Semenyih using extreme learning machines and metaheuristic optimization approach

Suspended sediment load (SSL) refers to sediment particles, such as silt and clay, that are suspended in water. It plays a critical role in hydrology and water quality management, influencing factors such as water quality, river erosion, sedimentation rates, dam operations, and aquatic ecosystems. Therefore, accurate prediction and monitoring of SSL is crucial for efficient water resources and environmental management. The study aims to enhance SSL prediction performance by integrating Extreme Learning Machines (ELM) with metaheuristic optimizations. The selected metaheuristic optimizations, namely Particle Swarm Optimization (PSO), Henry Gas Solubility Optimization (HGSO), Electromagnetic Field Optimization (EFO), and Nuclear Reaction Optimization (NRO), are systematically employed in ELM models for predicting SSL in Sungai Semenyih, Selangor, for a period from 2015 to 2018. The results show that the ELM-HGSO model slightly better performs than the other optimized models, achieving a Willmott Index (WI) of 0.9986, a Confidence Index (CI) of 0.996, a Mean Absolute Percentage Error (MAPE) of 0.058, a Nash-Sutcliffe Efficiency (NSE) of 0.994, and a Root Mean Square Error (RMSE) of 4.082. The ELM-HGSO model improves NSE by approximately 10 % compared to the standalone ELM model. The Taylor diagram further confirms that ELM-HGSO best fits the observed data for both training and test datasets. This study concludes that ELM-HGSO is the most suitable model for SSL prediction in the study area. The findings are expected to advance SSL prediction research, contributing to improved water management and sustainable infrastructure development.

Neural-based modeling adsorption capacity of metal organic framework materials with application in wastewater treatment

We developed a computational-based model for simulating adsorption capacity of a novel layered double hydroxide (LDH) and metal organic framework (MOF) nanocomposite in separation of ions including Pb(II) and Cd(II) from aqueous solutions. The simulated adsorbent was a composite of UiO-66-(Zr)-(COOH)2 MOF grown onto the surface of functionalized Ni50-Co50-LDH sheets. This novel adsorbent showed high surface area for adsorption capacity, and was chosen to develop the model for study of ions removal using this adsorbent. A number of measured data was collected and used in the simulations via the artificial intelligence technique. Artificial neural network (ANN) technique was used for simulation of the data in which ion type and initial concentration of the ions in the feed was selected as the input variables to the neural network. The neural network was trained using the input data for simulation of the adsorption capacity. Two hidden layers with activation functions in form of linear and non-linear were designed for the construction of artificial neural network. The model's training and validation revealed high accuracy with statistical parameters of R2 equal to 0.99 for the fitting data. The trained ANN modeling showed that increasing the initial content of Pb(II) and Cd(II) ions led to a significant increment in the adsorption capacity (Qe) and Cd(II) had higher adsorption due to its strong interaction with the adsorbent surface. The neural model indicated superior predictive capability in simulation of the obtained data for removal of Pb(II) and Cd(II) from an aqueous solution.

Predicting suspended sediment load in Peninsular Malaysia using support vector machine and deep learning algorithms

High loads of suspended sediments in rivers are known to cause detrimental effects to potable water sources, river water quality, irrigation activities, and dam or reservoir operations. For this reason, the study of suspended sediment load (SSL) prediction is important for monitoring and damage mitigation purposes. The present study tests and develops machine learning (ML) models, based on the support vector machine (SVM), artificial neural network (ANN) and long short-term memory (LSTM) algorithms, to predict SSL based on 11 different river data sets comprising of streamflow (SF) and SSL data obtained from the Malaysian Department of Irrigation and Drainage. The main objective of the present study is to propose a single model that is capable of accurately predicting SSLs for any river data set within Peninsular Malaysia. The ANN3 model, based on the ANN algorithm and input scenario 3 (inputs consisting of current-day SF, previous-day SF, and previous-day SSL), is determined as the best model in the present study as it produced the best predictive performance for 5 out of 11 of the tested data sets and obtained the highest average RM with a score of 2.64 when compared to the other tested models, indicating that it has the highest reliability to produce relatively high-accuracy SSL predictions for different data sets. Therefore, the ANN3 model is proposed as a universal model for the prediction of SSL within Peninsular Malaysia.

Utilizing Artificial Intelligence to Predict the Superplasticizer Demand of Self-Consolidating Concrete Incorporating Pumice, Slag, and Fly Ash Powders

Self-consolidating concrete (SCC) is a well-known type of concrete, which has been employed in different structural applications due to providing desirable properties. Different studies have been performed to obtain a sustainable mix design and enhance the fresh properties of SCC. In this study, an adaptive neuro-fuzzy inference system (ANFIS) algorithm is developed to predict the superplasticizer (SP) demand and select the most significant parameter of the fresh properties of optimum mix design. For this purpose, a comprehensive database consisting of verified test results of SCC incorporating cement replacement powders including pumice, slag, and fly ash (FA) has been employed. In this regard, at first, fresh properties tests including the J-ring, V-funnel, U-box, and different time interval slump values were considered to collect the datasets. At the second stage, five models of ANFIS were adjusted and the most precise method for predicting the SP demand was identified. The correlation coefficient (R²), Pearson’s correlation coefficient (r), Nash–Sutcliffe efficiency (NSE), root mean square error (RMSE), mean absolute error (MAE), and Wilmot’s index of agreement (WI) were used as the measures of precision. Later, the most effective parameters on the prediction of SP demand were evaluated by the developed ANFIS. Based on the analytical results, the employed algorithm was successfully able to predict the SP demand of SCC with high accuracy. Finally, it was deduced that the V-funnel test is the most reliable method for estimating the SP demand value and a significant parameter for SCC mix design as it led to the lowest training root mean square error (RMSE) compared to other non-destructive testing methods.

Thermal prediction of turbulent forced convection of nanofluid using computational fluid dynamics coupled genetic algorithm with fuzzy interface system

Computational fluid dynamics (CFD) simulating is a useful methodology for reduction of experiments and their associated costs. Although the CFD could predict all hydro-thermal parameters of fluid flows, the connections between such parameters with each other are impossible using this approach. Machine learning by the artificial intelligence (AI) algorithm has already shown the ability to intelligently record engineering data. However, there are no studies available to deeply investigate the implicit connections between the variables resulted from the CFD. The present investigation tries to conduct cooperation between the mechanistic CFD and the artificial algorithm. The genetic algorithm is combined with the fuzzy interface system (GAFIS). Turbulent forced convection of Al₂O₃/water nanofluid in a heated tube is simulated for inlet temperatures (i.e., 305, 310, 315, and 320 K). GAFIS learns nodes coordinates of the fluid, the inlet temperatures, and turbulent kinetic energy (TKE) as inputs. The fluid temperature is learned as output. The number of inputs, population size, and the component are checked for the best intelligence. Finally, at the best intelligence, a formula is developed to make a relationship between the output (i.e. nanofluid temperatures) and inputs (the coordinates of the nodes of the nanofluid, inlet temperature, and TKE). The results revealed that the GAFIS intelligence reaches the highest level when the input number, the population size, and the exponent are 5, 30, and 3, respectively. Adding the turbulent kinetic energy as the fifth input, the regression value increases from 0.95 to 0.98. This means that by considering the turbulent kinetic energy the GAFIS reaches a higher level of intelligence by distinguishing the more difference between the learned data. The CFD and GAFIS predicted the same values of the nanofluid temperature.