Comparison of different artificial neural network architectures in modeling of Chlorella sp. flocculation

Abstract

Biodiesel production from microalgae feedstock should be performed after growth and harvesting of the cells, and the most feasible method for harvesting and dewatering of microalgae is flocculation. Flocculation modeling can be used for evaluation and prediction of its performance under different affective parameters. However, the modeling of flocculation in microalgae is not simple and has not performed yet, under all experimental conditions, mostly due to different behaviors of microalgae cells during the process under different flocculation conditions. In the current study, the modeling of microalgae flocculation is studied with different neural network architectures. Microalgae species, Chlorella sp., was flocculated with ferric chloride under different conditions and then the experimental data modeled using artificial neural network. Neural network architectures of multilayer perceptron (MLP) and radial basis function architectures, failed to predict the targets successfully, though, modeling was effective with ensemble architecture of MLP networks. Comparison between the performances of the ensemble and each individual network explains the ability of the ensemble architecture in microalgae flocculation modeling.

Modeling of microalgal shear‐induced flocculation and sedimentation using a coupled CFD‐population balance approach

“…The recovery efficiency was calculated for a range of different PAC flocculant dosages in order to find the optimum amount which tended to rise with increasing the dosage, as depicted in Figure , and then decreased after reaching an optimum point. This is because addition of the flocculant initially facilitates the flocculation by neutralizing cells, while further addition leads to an inhibition effect due to accumulation of positive charge in the suspension preventing cells from attaching to one another …”

“…This is because addition of the flocculant initially facilitates the flocculation by neutralizing cells, while further addition leads to an inhibition effect due to accumulation of positive charge in the suspension preventing cells from attaching to one another. 49 The main purpose of the experimental work was to investigate the average floc size in steady state conditions. To do so, the experimental procedure explained earlier was repeated three times for shear rate of 182 s 21 to minimize possible errors and ensure accuracy of the results.…”

<b>Artificial neural networks to control chlorine dosing in a water treatment plant

“…The performance of the artificial neural network is measured by the mean squared error (MSE) and mean absolute error (MAE), respectively shown in Equation 1 and 2 (Ayodele & Auta, 2012;Zenooz et al, 2017):…”

“…Artificial Neural Networks (ANN), a technique related to computational intelligence, improves processes with multivariate and non-linear characteristics, such as water treatment and effluent processes (Han & Qiao, 2014;Liu & Chung, 2014). Likewise, ANN is capable of extracting information from processes that are not well understood or detailed (Na, Ren, Shang, & Guo, 2012;Zenooz, Ashtiani, Ranjbar, Nikbakht, & Bolour, 2017).…”

Chlorella sp. protective effect on acetaminophen-induced liver toxicity in ICR mice

“…is a good prophylactic-therapeutic agent against obesity-related complications. [ 10 ] Acetaminophen, or N-acetyl-para-amino-phenol (APAP), is widely used analgesic-antipyretic drugs. Although they are considered safe drugs,[ 11 ] they cause hepatic necrosis and renal failure when given in high doses.…”