Experimental Evaluation and Data-Driven Modelling of Solar Still Performance under Coastal Climatic Conditions

In many parts of the world, the ocean offers an endless supply of water that is simply too salty to drink. For communities living along these coasts, turning that saltwater into fresh water is a matter of survival. One of the oldest and simplest ways to do this is with a solar still, a device that uses the sun's heat to boil water, capture the steam, and let it cool back into liquid form, leaving the salt behind. While the basic idea is straightforward, the actual amount of water produced depends heavily on the weather. In dry, hot deserts, these devices work well, but in humid coastal areas, the air is already full of moisture, which can slow down the process of turning steam back into water. Understanding exactly how these changing weather conditions affect the machine is crucial for designing systems that can reliably provide drinking water where it is needed most.

A researcher set out to solve this puzzle by testing a single-slope solar still in Tuticorin, a coastal city in southern India known for its hot, humid summers. They built a simple box with a black basin to hold saltwater and a slanted glass cover to catch the steam. Over the course of three months during the peak summer season, they ran the device outdoors, letting nature take its course. Every five minutes, sensors recorded the temperature of the water, the glass, and the air, along with the intensity of the sunlight, the speed of the wind, and how much moisture was in the air. At the end of each day, they weighed the fresh water collected to see exactly how much the system had produced. The goal was not just to measure the output, but to understand which specific weather factors were helping or hindering the process in this difficult, humid environment.

The researcher found that the amount of fresh water produced was not determined by a single factor, but by a complex mix of conditions. As expected, more sunlight meant more water, but the relationship was not perfectly straight. On days with similar amounts of sunshine, the water yield could vary significantly. This variation was largely due to the humidity. When the air was thick with moisture, the steam inside the still struggled to condense back into water on the glass, capping the amount of fresh water the system could make. The study confirmed that while high temperatures and strong sunlight drive the evaporation process, high humidity and a high dew point—the temperature at which moisture begins to form—act as a brake on the system. Even the wind played a role, though its effect was less dramatic than in dry areas, as the sea breeze helped cool the glass just enough to aid condensation without stripping away too much heat.

To make sense of all these interacting variables, the researcher turned to a method called Principal Component Analysis. Because weather factors like temperature, humidity, and air pressure often change together, they can confuse computer models that try to predict outcomes. This statistical technique allowed the researcher to group these related measurements into a few key patterns, effectively simplifying the data without losing the important details. They then used these simplified patterns to train several different computer models, teaching them to predict how much water the still would produce on any given day based on the weather forecast. They tested various approaches, from simple straight-line predictions to more complex, tree-like decision models.

The results showed that the most reliable predictions came from the simpler models. Specifically, a standard linear regression model and a more advanced version called Bayesian Ridge Regression proved to be the most accurate. These models could predict the daily water yield with a high degree of certainty, capturing about 89 percent of the variations seen in the real-world experiments. The error in their predictions was small, averaging about 136 milliliters per day. Interestingly, the more complex models, which are often used for difficult problems, did not perform as well in this specific case. This suggests that for solar stills in coastal climates, the relationship between the weather and the water output is strong enough that a straightforward mathematical approach is often better than a complicated one.

The study concludes that while the humid coastal climate presents a challenge, it is possible to accurately predict and optimize the performance of solar desalination systems using real-world data and modern computing tools. By understanding that humidity is the primary limiting factor, engineers can design better systems for these specific environments. The research demonstrates that combining physical experiments with data-driven modeling offers a powerful way to improve the reliability of fresh water production in regions where it is most needed, turning the unpredictable nature of the weather into a manageable variable for future technology.