Irrigation Monitoring Device

Background: Dr. Wendy O'Meara, a Professor and Researcher at Duke, found out during her research that some especially dry regions in Kenya have had a large number of Malaria cases due to mismanagement of rudimentary irrigation canals. In order to assist the communities around these canals, she created a project intended in designing and building a low-cost device to monitor these canals.

After applying and interviewing for this position, I was invited to join the team and I worked with them during the 2021 - 2022 academic year. This team consisted of 8 undergraduate students and 2 graduate students. It was primarily student-led, but we had 5 faculty advisors that attended our meetings and provided feedback for us. Here is a summary of our work.

My Responsabilities:

• Research about Malaria and mosquito breeding through literature review and conducting interviews with experts.

• Brainstorm and analyze potential solutions to the problem through thorough and genuine feedback.

• Conduct and document team meetings.

• Lead the design and development of the electronic components of our design.

• Lead general integration of all device parts.

• Test device quality and performance through iterative documented processes.

• Implement and test the device's performance in Kenya with a subteam of 3 other students.

Device specifications: After much research, interviews and discussions, we came with a main design idea. We wanted to create a device that would detect if a specific irrigation canal had water going through it and how fast it was moving. The more detailed criteria list was as follows:

NC - Non-constraint. In other words, "Nice to have".

We constantly updated and reviewed these criterias as we moved on through brainstorming, prototyping and testing.

Brainstorming:

We had several brainstorming sessions where we would all come up with possible solutions to the problem, and had others give feedback to them. The main goal was to see what would be the best mechanism for the device. These sessions happened after we all just put possible solutions in a jam board, and then had to choose one to ellaborate.

After all of our sessions, we decided 2 top ideas. One of them was to use a hinge connected to an accelerometer and use the angle of deflection as a indicator of how fast the water was moving. It would be something like this:

The other idea was to use a hall-effect sensor to measure rotations in a turbine. These rotations would be translated to water flow by using a adjustable model and then outputted in a screen. It would be something like this:

In order to evaluate which idea was best, we divided the group in two teams and each team was responsible for designing and testing each idea. My team was responsible for the second idea.

Initially our device had a promising future as it had good testing results when we used sink water flow, however when we tested in an apparatus closer to the what the water canals are like, our sensor did not work. One day, I was in the lab trying to find all of the possible ways of solving this, and I decided to open up the sensor and adjust its internal features. It became something like this:

And it worked out! We gathered good data and were ready to finally show to our mentors (after 1 month of prototyping and testing) what we had. During this meeting, everyone was pleased with our progress and how we creatively solved this issue. However, since we only were able to fix it 1 week prior the meeting and the other device, while not very accurate, had more consistency, our mentors voted to choose the hinge as our mechanism. And so we proceeded.

I don't want to bother you with too many details, but I am happy to talk with you any other time about all of the stories behind this project. In short, I was very disappointed and frustrated about the mechanism selection, but after thinking about it, I decided to own up to it, and do my best to make this project work, even if it was not my original idea. And it worked!

Results:

The final version had the following features:

• Structure

  • PVC pipes as the transvering structure of our device.

  • Metal rods to attach to the sides of the canal and add rigidity.

  • Plastic containers to allo

  • A plastic see-through box with the electronics in it.

  • A 3d-printed case attached to the acrylic hinge.

• Electronics

  • Arduino Every as the microcontroller.

  • An accelerometer to record angle measurements.

  • A moisture sensor to detect water presence.

  • SD-card to store all data.

  • A LCD screen to display the data.

  • A breadboard to connect the components.

  • A Power bank to power the system.

This is us in Kenya implementing our device!

This is me troubleshooting some unexpected problems we had on the spot:

And this is a video showing the device working:

The device would be left measuring data every minute overnight, and we would check it on the next day. We were then able to see the accelerometer values and deduce how it behaved overnight, including periods of increasing water volume and movement orientation. It still had some noise and our data collection model needs to be refined, but it was a great start!

And this project is still going! While the original participants in it are not actively working, we are now mentors to a group of freshmen that are now taking the lead as an effecitve way to make this project move forward even with students graduating. Stay tune to Duke News to hear more about it!

Awards:

• 2nd Place - Best Exhibition, Bass Connections Symposium (30+ teams).

• Bass Connections Student Research Award.

• URS Independent Study Award.

Thank you for checking this out!