Deep neural network prediction of modified stepped double-slope solar still with a cotton wick and cobalt oxide nanofluid

Abstract

This research work intends to enhance the stepped double-slope solar still performance through an experimental assessment of combining linen wicks and cobalt oxide nanoparticles to the stepped double-slope solar still to improve the water evaporation and water production. The results illustrated that the cotton wicks and cobalt oxide (Co3O4) nanofluid with 1wt% increased the hourly freshwater output (HP) and instantaneous thermal efficiency (ITE). On the other hand, this study compares four machine learning methods to create a prediction model of tubular solar still performance. The methods developed and compared are support vector regressor (SVR), decision tree regressor, neural network, and deep neural network based on experimental data. This problem is a multi-output prediction problem which is HP and ITE. The prediction performance for the SVR was the lowest, with 70 (ml/m2 h) mean absolute error (MAE) for HP and 4.5% for ITE. Decision tree regressor has a better prediction for HP with 33 (ml/m2 h) MAE and almost the same MAE for ITE. Neural network has a better prediction for HP with 28 (ml/m2 h) MAE and a bit worse prediction for ITE with 5.7%. The best model used the deep neural network with 1.94 (ml/m2 h) MAE for HP and 0.67% MAE for ITE.

Predicting Rejection of Emerging Contaminants through RO Membrane Filtration based on ANN- QSAR Modeling Approach: Trends in Molecular Descriptors and Structures towards Rejections

“…In the QSAR method, theoretical descriptors are a group of numerical indices that are associated with the structure of molecules and encode information about the structure ( The ANN algorithms are non-linear models that make a mapping of the input and output variables, in turn, the map is utilized to predict unknown output as a function of appropriate descriptors (Sharshir et al 2022). A principal advantage of ANN methods is that they can incorporate and combine both experimental data and literature-based to solve many problems such as predicting membrane permeability and membrane rejection.…”

Predicting rejection of emerging contaminants through RO membrane filtration based on ANN-QSAR modeling approach: trends in molecular descriptors and structures towards rejections

“…41–47 The ANN algorithms are non-linear models that make a mapping of the input and output variables, in turn, the map is utilized to predict unknown output as a function of appropriate descriptors. 48 The main advantage of ANN methods is that they can incorporate and combine both experimental data and literature-based to solve many problems such as predicting membrane permeability and membrane rejection. This predictive power can be captured to virtually analyze the properties of molecules before testing them in a laboratory.…”

Deep Neural Networks Based Modeling to Optimize Water Productivity of a Passive Solar Still

“…The hybrid system resulted in a daily water production rate that was 24% higher than the production of a passive one. However, nanotechnology was proposed in recent studies [7,8,11,12,[21][22][23][24][25] to enhance the performance of triangular solar stills. Goshayeshi et al [26] conducted a novel approach involving a mixture of paraffin and graphene oxide nanoparticles.…”

“…Besides, several other studies [60][61][62][63][64][65][66][67][68] have also explored various DNN algorithms, where four machine learning methods were employed in a comparative study [22] to develop a prediction model for the performance of tubular solar stills: Support Vector Regressor (SVR), DT regressor, neural network, and DNN. The objective was to predict Hourly Freshwater Production (HP) and Instantaneous Thermal Efficiency (ITE).…”