Search

An effective interdisciplinary study between machine learning and solar-thermal desalination requires a sufficiently large and well-analyzed experimental datasets. This study develops a modified dataset collection and analysis process for studying solar-thermal desalination by machine learning. Based on the optimized water condensation and collection process, the proposed experimental method collects over one thousand datasets, which is ten times more than the average number of datasets in previous works, by accelerating data collection and reducing the time by 83.3%. On the other hand, the effects of dataset features are investigated by using three different algorithms, including artificial neural networks, multiple linear regressions, and random forests. The investigation focuses on the effects of dataset size and range on prediction accuracy, factor importance ranking, and the model's generalization ability. The results demonstrate that a larger dataset can significantly improve prediction accuracy when using artificial neural networks and random forests. Additionally, the study highlights the significant impact of dataset size and range on ranking the importance of influence factors. Furthermore, the study reveals that the extrapolation data range significantly affects the extrapolation accuracy of artificial neural networks. Based on the results, massive dataset collection and analysis of dataset feature effects are important steps in an effective and consistent machine learning process flow for solar-thermal desalination, which can promote machine learning as a more general tool in the field of solar-thermal desalination.

Solar still is a convenient off-grid device for desalination, which can provide fresh water for families, ships, islands and so on. The conventional inclined solar still (ISS) suffers from low efficiency and low productivity. To improve the performance of solar still, a flat solar still (FSS) is proposed, which has a working principle similar to the solar cell. The condensate water in FSS is collected by the capillary grid attached under the ultra-hydrophilic glass cover, instead of by gravity. Therefore, FSS avoids the inclined structure and is much more compact than ISS. The daily productivity of FSS reaches up to 4.3 kg/m2. Theoretical analysis shows that the enhanced mass transfer in FSS by the compact structure is an important factor for high performance. More interestingly, FSS can also be easily extended to more stage for latent heat recovery. The results show that the daily productivity of a double-stage FSS reaches up to 7 kg/m2, which is much higher than the conventional solar still. FSS paves a new way in designing and optimizing of solar still., Comment: 34

Presented is a new generation prediction model of a tubular solar still (TSS) productivity utilizing two machine learning (ML) techniques, namely:Random forest (RF) and Artificial neural network (ANN). Prediction models were conducted based on experimental data recorded under Egyptian climate. Meteorological and operational thermal parameters were utilized as input layers. Moreover, Bayesian optimization algorithm (BOA) was used to obtain the optimal performance of RF and ANN models. In addition, these models results were compared to those of a multilinear regression (MLR) model. As resulted, experimentally, the average value accumulated productivity was 4.3 L/(m2day). For models results, RF was less sensitive to hyper parameters than ANN as ANN performance could be significantly improved by BOA more than RF. In addition, RF achieved better prediction performance of TSS on the current dataset. The determination coefficients (R2) of RF and ANN were 0.9964 and 0.9977, respectively, which were much higher than MLR models, 0.9431. Based on the robustness performance and high accuracy, RF is recommended as a stable method for predicting the productivity of TSS.

In last decades, solar stills, as one of the solar desalination technologies, have been well studied in terms of their productivity, efficiency and economics. Recently, to overcome the bottleneck of traditional solar still, improving solar still by optimizing the solar evaporation process based on micro/nanomaterials have been proposed as a promising strategy. In this review, the recent development for achieving high-performance of solar still and solar evaporation are discussed, including materials as well as system configurations. Meanwhile, machine learning was used to analyze the importance of different factors on solar evaporation, where thermal design was founded to be the most significant parameter that contributes in high-efficiency solar evaporation. Moreover, several important points for the further investigations of solar still and solar evaporation were also discussed, including the temperature of the air-water interface, salt rejecting and durability, the effect of solid-liquid interaction on water phase change., Comment: 53pages

Solar evaporation is important for many applications such as desalination, power generation and industrial drying. Recently, some studies on evaporation reported obtaining high energy efficiency and evaporation rate, which are based on floating evaporation setup (FES) with nanomaterials. Here, we proposed a new cheap and simple setup, named as airing evaporation setup (AES). It shows that the energy efficiency of AES reaches up to 87 % under 1 kW/m2 of solar irradiation, which is 14% higher than that of FES. Meanwhile, the total evaporation rate of AES is about 20% higher than that of FES. The theoretical analysis reveals that the main reason for a better performance of AES is the increasing evaporation area. More interesting, AES could be used for designing portable systems due to its simplicity and flexibility. Furthermore, we show that AES and the corresponding wick material can be used in solar desalination, textile quick-drying and warm-keeping., Comment: 21pages, 5 figures