Prediction of standard aeration efficiency of a propeller diffused aeration system using response surface methodology and an artificial neural network

Abstract

Aeration experiments were conducted in a masonry tank to study the effects of operating parameters on standard aeration efficiency (SAE) of a propeller diffused aeration (PDA) system. The operating parameters include the rotational speed of shaft (N), submergence depth (h), and propeller angle (α). The response surface methodology (RSM) and artificial neural network (ANN) were used for modelling and optimizing the standard aeration efficiency (SAE) of a PDA system. The results of the both approaches were compared for their modelling abilities in terms of coefficient of determination (R2), root mean square error (RMSE), and mean absolute error (MAE), computed from experimental and predicted data. ANN models were proved to be superior to RSM. The results indicate that for achieving the maximum standard aeration efficiency (SAE), N, h and α should be 1,000 rpm, 0.50 m, and 12°, respectively. The maximum SAE was found to be 1.711 kg O2/ kWh. The cross-validation results show that the best approximation of optimal values of input parameters for maximizing SAE is possible with a maximum deviation (absolute error) of ±15.2% between the model predicted and experimental values.

Modelling and prediction of aeration efficiency of the venturi aeration system using ANN-PSO and ANN-GA

“…The PSO requires fewer parameter adjustments than any other optimisation approach available and is simpler to implement (Eberhart and Shi, 2001). Furthermore, if the target function includes a single attribute and several local minima/maxima, the ideal value may be reached using the PSO method (Roy et al, 2021b). As PSO is a naturalistic exploration method as opposed to gradient-based optimisation, it has the drawback of being a sluggish procedure (Roy et al, 2021b).…”

“…Furthermore, if the target function includes a single attribute and several local minima/maxima, the ideal value may be reached using the PSO method (Roy et al, 2021b). As PSO is a naturalistic exploration method as opposed to gradient-based optimisation, it has the drawback of being a sluggish procedure (Roy et al, 2021b). On the other hand, complicated non-linear optimisation issues are frequently solved using mathematical techniques such as genetic algorithm-based optimisation (ANN-GA) and ANN modelling.…”

“…According to a thumb rule, the optimal values of DO for the majority of aquatic species at sea level are considered as 50% saturation concentration in freshwater bodies, or 5 mg/L and 4 mg/L at temperatures of 15°C and 26°C, respectively (Roy et al, 2022;Yadav and Roy, 2023). In particular, the cold water and warm water species require DO concentrations between 7 mg/L and 9 mg/L and between 4 mg/L and 7 mg/L for proper growth and reproduction (Roy et al, 2021b(Roy et al, , 2022. The existence possibility of aquatic species can increase using artificial aeration techniques in pond culture (Pawar et al, 2009).…”

“…Recently, according to the review of literature, several research studies have been conducted on soft computing techniques based on artificial neural network (ANN), particle swarm optimisation (PSO), and genetic algorithm (GA) modelling in a variety of applications of aeration system (Bagatur and Onen, 2014;Kumar et al, 2018;Garg and Jain, 2020;Roy et al, 2021b). The ANN is a useful tool for simulating complex processes with dynamic interaction between the components that affect outcomes (Omid et al, 2009;Pareek et al, 2021).…”

“…A validation cheque halted the learning process of the neural network. A variety of validation tests will be carried out based on how frequently the neural network is iterated(Roy et al, 2023). The gradient's magnitude and the number of validation cheques needed to end the training are displayed in Figure7A.…”