AI helps make chlorine for clean water

In New Delhi, India, researchers have found a way to use ‘free residual chlorine’ in drinking water to kill bacteria and viruses that cause disease outbreaks in refugee camps and settlements.

According to the US Centers for Disease Control and Prevention (CDC), free residual chlorine that can kill germs is a measure of how safe it is to drink the water. A chlorine level of 0.5 milligrams per litre can maintain quality of water in piped distribution networks, but not enough to maintain that of water kept in a bucket for 24 hours.

Researchers say that their tool uses machine learning to predict if enough chlorine will stay in stored water to keep it potable.

Michael De Santi, a researcher at Canada’s York University’s Lassonde School of Engineering, explains further that their tool “predict[s] chlorine decay in refugee or internally displaced person settlements after water leaves the distribution system.”

The study said that refugees do not normally get piped drinking water, instead rely on water from public taps kept in containers. This increases the chances of water contamination, resulting in outbreaks of cholera, shigella, hepatitis E, among other diseases.

De Santi adds that the machine learning model uses “probabilistic forecasting”, in which the model can predict the full range of possible chlorine concentrations and their associated probabilities. Providing a probability instead of a single value allows water system operators to create an optimised chlorination strategy.