³ÉÈËÊÓƵ

Clean water is a global priority, but monitoring quality is becoming a worsening challenge as water bodies become more contaminated.

Sungyong Jung, South Dakota State University professor and head of the McComish Department of Electrical Engineering, and his research team are working on a solution utilizing high-tech sensors and artificial intelligence.

In daily life, humans are using more chemicals, which eventually make their way into water sources. These chemicals are not only potentially harmful to human health, but they are also difficult to detect. 

On top of that, traditional water quality monitoring methods are labor intensive, time consuming and expensive. To monitor South Dakota's 11,929 miles of rivers and streams and 577 lakes and reservoirs, the state's Department of Agriculture and Natural Resources collects samples from water bodies on a regular or semiregular basis. The sample is then brought to the lab, where the water is separated and tested for different chemicals.

Monitoring all these water bodies in a timely and efficient manner is a challenge. That’s why Jung is developing a method that will be able to efficiently monitor water quality in real time.

Jung's high-tech sensors are submerged underneath the water and measure various chemicals, including nitrate, nitrite, ammonium, nitrogen and phosphorus. After being placed in the water, the sensors will give an "instant result" to pertaining to the water quality.

Commercial sensors can be used for this work, but those are highly expensive. Instead, Jung — with an extensive background in integrated circuits and embedded systems for chemical and biosensing applications — often builds his own sensors. To detect the trace amount of chemicals, highly sensitive optical sensors must be developed and deployed.

"We have to utilize nanotechnology to develop such sensors," Jung said. "We then need to build electronic systems in conjunction with the sensor in order to accurately detect the chemicals. Our work is focused on the development of the readout circuit for the electrochemical sensors."

The sensors have an electronic readout circuit that remotely sends long-range wireless data transmission signals back to a mobile application. This allows for real-time monitoring of water bodies and eliminates the need to retrieve the sensor frequently.

Artificial intelligence must also be utilized for these systems to work. Machine learning techniques analyze the data collected from the sensors to provide a level for each chemical detected. The data is then sent to the mobile application through the embedded system.

Jung and his research team, which includes post-docs and doctoral candidates, develops the mobile application, allowing the user to easily interpret the data from the convenience of their phone or computer.

"We develop the user interface and mobile app," Jung said. "Users can control certain settings to read the data."

Jung began this type of work while as a faculty member at the University of Texas-Arlington. There, he developed efficient low-power, integrated circuits for hearing aids. Eventually, his work expanded into agriculture-based research and sensors, but the low-power, integrated circuits have remained as the research's basis.

Overall, Jung believes this work could be an asset to natural resource managers or could even be used by farmers and other private landowners to monitor their water bodies.

"This is a tool for decision-making optimization," Jung added.