Prediction of tubular solar still performance by machine learning integrated with Bayesian optimization algorithm

Wang, Yunpeng; Kandeal, A.W.; Swidan, Ahmed; Sharshir, Swellam W.; Abdelaziz, Gamal B.; Halim, M. A.; Kabeel, A.E.; Yang, Nuo

doi:10.1016/j.applthermaleng.2020.116233

Cited by 98 publications

(20 citation statements)

References 54 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The applied method and algorithm are among the most important factors that influence the exactness of the datadriven methods in forecasting the outputs of solar stills (Mashaly and Alazba, 2015;Mashaly and Alazba, 2017b;Mashaly and Alazba, 2018a;Mashaly and Alazba, 2018b;Mashaly and Alazba, 2019a). For instance, Wang et al (2021) used random forest (RF), ANN, and multilinear regression to forecast the productivity of the system based on time, solar radiation intensity, wind speed, temperatures of feed water, basin plates, salt water, cover, and ambient temperature. They found that using RF led to the prediction with the least error compared with others.…”

Section: Applications Of Data-driven Methods In Solar Desalinationsmentioning

confidence: 99%

Utilization of Data-Driven Methods in Solar Desalination Systems: A Comprehensive Review

Nazari¹,

Salem²,

Mahariq³

et al. 2021

Front. Energy Res.

View full text Add to dashboard Cite

Renewable energy sources have been used for desalination by employing different technologies and mediums due to the limitations of fossil fuels and the environmental issues related to their consumption. Solar energy is one of the most applicable types of renewable sources for desalination in both direct and indirect ways. The performance of solar desalination is under effects of different factors which makes their performance prediction difficult in some cases. In this regard, data-driven methods such as artificial neural networks (ANNs) would be proper tools for their modeling and output forecasting. In the present article, a comprehensive review is provided on the applications of different data-driven approaches in performance modeling of solar-based desalination units. It can be concluded that by employing these methods with proper inputs and structures, the outputs of the solar desalination units can be reliably and accurately forecasted. In addition, several recommendations are produced for the upcoming work in the relevant areas of the study.

show abstract

Section: Applications Of Data-driven Methods In Solar Desalinationsmentioning

confidence: 99%

Utilization of Data-Driven Methods in Solar Desalination Systems: A Comprehensive Review

Nazari¹,

Salem²,

Mahariq³

et al. 2021

Front. Energy Res.

View full text Add to dashboard Cite

show abstract

“…Based on the literature, especially as reported by Wang et al, [ 32 ] it has been found that RF has more robust performance than ANN predicting the freshwater yield of tubular SS, whereas the ANN is more sensitive to hyperparameters and more amenable to be enhanced by BOA than RF. Considering these findings, the current work aimed to confirm that based on another design of SS, which is DSSS to determine the validity of these results to be generalized on the field of solar desalination via SSs.…”

Section: Introductionmentioning

confidence: 97%

“…For example, using a double‐layer square wick can lead to 114% yield improvement, [ 29 ] using a concave‐shaped wick can result in good production up to 4.1 L m −2 , [ 30 ] and integrating a wick‐wrapped pin‐finned basin can get enhanced yield by 23%. On the contrary, the SS performance can be enhanced via developing new designs, such as tubular, [ 31–34 ] spherical, [ 35 ] pyramid, [ 36–38 ] stepped solar still, [ 39 ] pyramid with evacuated tubes, [ 40 ] and double slope SSs (DSSSs). [ 19 ] Among these designs, DSSS has been preferable due to its simple design and cost‐effective construction and operation.…”

Section: Introductionmentioning

confidence: 99%

Productivity Modeling Enhancement of a Solar Desalination Unit with Nanofluids Using Machine Learning Algorithms Integrated with Bayesian Optimization

Kandeal

Chen

et al. 2021

Energy Tech

Self Cite

View full text Add to dashboard Cite

Herein, double slope solar still (DSSS) performance is accurately forecast with the aid of four different machine learning (ML) models, namely, artificial neural network (ANN), random forest (RF), support vector regression (SVR), and linear SVR. Furthermore, the tuning of ML models is optimized using the Bayesian optimization algorithm (BOA) to get the optimal performance of all models and identify the best predictive one. All the models are trained, tested, and validated depending on experimental data acquired under Egyptian climatic conditions. The results reveal that ML models can be a powerful tool to forecast DSSS performance. Among them, RF is the most potent ML model obtaining the highest determination coefficient (R2) and the lowest absolute error percentage of 0.997% and 2.95%, respectively. Furthermore, the experimental results also show that the mean value of accumulated (daily) freshwater productivity from DSSS is 4.3 L m−2.

show abstract

“…Zhang et al, 2021), medicine (Dhamala et al, 2020), and engineering (Y. Wang et al, 2021), (Wu et al, 2019), (Q. Zhang et al, 2020).…”

Section: Introductionmentioning

confidence: 99%

Machine Learning Models Based on Random Forest Feature Selection and Bayesian Optimization for Predicting Daily Global Solar Radiation

Chaibi

Benghoulam

Tarik

et al. 2021

Int. J. Renew. Energy Dev.

View full text Add to dashboard Cite

Prediction of daily global solar radiation with simple and highly accurate models would be beneficial for solar energy conversion systems. In this paper, we proposed a hybrid machine learning methodology integrating two feature selection methods and a Bayesian optimization algorithm to predict H in the city of Fez, Morocco. First, we identified the most significant predictors using two Random Forest methods of feature importance: Mean Decrease in Impurity (MDI) and Mean Decrease in Accuracy (MDA). Then, based on the feature selection results, ten models were developed and compared: (1) five standalone machine learning (ML) models including Classification and Regression Trees (CART), Random Forests (RF), Bagged Trees Regression (BTR), Support Vector Regression (SVR), and Multi-Layer Perceptron (MLP); and (2) the same models tuned by the Bayesian optimization (BO) algorithm: CART-BO, RF-BO, BTR-BO, SVR-BO, and MLP-BO. Both MDI and MDA techniques revealed that extraterrestrial solar radiation and sunshine duration fraction were the most influential features. The BO approach improved the predictive accuracy of MLP, CART, SVR, and BTR models and prevented the CART model from overfitting. The best improvements were obtained using the MLP model, where RMSE and MAE were reduced by 17.6% and 17.2%, respectively. Among the studied models, the SVR-BO algorithm provided the best trade-off between prediction accuracy (RMSE=0.4473kWh/m²/day, MAE=0.3381kWh/m²/day, and R²=0.9465), stability (with a 0.0033kWh/m²/day increase in RMSE), and computational cost.

show abstract

Prediction of tubular solar still performance by machine learning integrated with Bayesian optimization algorithm

Cited by 98 publications

References 54 publications

Utilization of Data-Driven Methods in Solar Desalination Systems: A Comprehensive Review

Utilization of Data-Driven Methods in Solar Desalination Systems: A Comprehensive Review

Productivity Modeling Enhancement of a Solar Desalination Unit with Nanofluids Using Machine Learning Algorithms Integrated with Bayesian Optimization

Machine Learning Models Based on Random Forest Feature Selection and Bayesian Optimization for Predicting Daily Global Solar Radiation

Contact Info

Product

Resources

About