In 1987, I was born … [edited for brevity] … and we finally found the hidden treasure behind the bookcase. That pushed me toward my current path of mechanical engineering – because everyone knows there’s no money in the treasure hunting business.
As I mentioned in my welcome e-mail, I’m a grad student at University of Washington. Experience gained as a researcher at Woods Hole Oceanographic Institution taught me that funding is always a concern in academic research, so I planned to apply for fellowships once I had an academic home. With guidance from Brian, my advisor at UW, and support from other previous mentors, I applied for and received a National Science Foundation Graduate Research Fellowship. I remember biking home, on the same route I had been taking for months, thinking, Holy crap! Three years of funding! Holy crap! Don’t get hit by a car. Don’t crash. Apparently, some part of me thought my life was suddenly more valuable. Or felt endangered by the possibility of immediate karmic balance. My rational mind knows that’s not how life works, but that was my reaction.
The fellowship allowed me the flexibility to undertake a research project other than those already being pursued in the lab. This was both a benefit and a curse: I struggle with decision making. Especially big decisions. Hilariously so, because I typically have a gut feeling, but need to go through all other possibilities, determine why they are also good options, but ultimately decide on the original, instinctive choice*.
Brian’s lab lives in the world of marine renewable energy, primarily focused on tidal and wave energy. Those of you who grew up or worked on the coasts will have probably felt or seen the fast currents associated with changing tides. Tidal current energy devices are like underwater wind turbines that extract energy from these water currents. Even small changes in current speed between locations can dramatically impact power output. And computer models of the flow based on site bathymetry (underwater topology), basic physics, and the delightful predictability of the tides can still be surprisingly far off. You can’t take a model to the bank — figuratively and literally. So we need to validate the model with measurements of the site. That’s the world I’m living in: instrumentation.
In these locations, we need to gather data over long time period and we can accomplish this task quite successfully with bottom packages, i.e. a bunch of sensors on an anchored frame. There are some difficulties with that, too, but I’ll leave out the boring details. We also care about making measurements over a range of possible sites, that is, spatially distributed. Well, why not put out a bunch of bottom packages? Sure, that would work. Except that the sensor suites necessary to get good data cost $30k – $100k each, thus a sensor array quickly becomes prohibitively expensive.
So what about mobile platforms? Why not use a ship or an underwater drone to ‘lawnmower’ a pattern? Great thinking, reader! You’re totally on the right track. But you might have guessed, there are some issues associated with those as well. Ship time is pretty expensive and wave movements contaminate the signal. Well, what about an underwater drone? They exist, right? Underwater drones do indeed exist and are called AUVs (Autonomous Underwater Vehicles)**, though they’re expensive at $100k – $500k and more if you kit them out with all the bells, whistles, and sensors. Now, keep in mind when we’re extracting energy, we’re looking for locations with some of the fastest moving currents on the planet. The slowest sites we consider have flows at 1 m/s (2.2 mph, 1.9 knots) and reach upwards of 5 m/s (11.2 mph, 9.7 knots) in the Bay of Fundy. I’ve got a bad feeling about this… how fast can AUVs travel? They typically have a top speed of 1 m/s – a hard fight against these currents. That half-million dollar fancy vehicle is now just a half-million dollars of drifting sensors.
And here we are at the premise of my research: skip the half million dollars and go straight to a drifting sensor. Take advantage of inexpensive hobby-grade controls and sensors to make the package cheap, allowing us to build a large number. The idea is basically an underwater version of Dorothy, from Twister: throw a bunch of small sensors out in the water and measure the currents by tracking them. These µFloats (micro floats***) are depth controlled, moving themselves up and down in the water column by adjusting their buoyancy.
The NSF fellowship I received also provides opportunity for extensions, including GROW: Graduate Research Opportunities Worldwide. Discovering this, I was stoked on the idea of working internationally for some portion of my degree, but didn’t have a clear place to apply. Then Matt Dunababin of QUT gave a seminar at UW about his work with optimal energy path planning using AUVs and depth controlled floats. Basically, we use a map of the expected flow and strategically position the device such that water flow, rather than the AUV motor, pushes you around. In the case of a float, these current are the only source horizontal movement. While tides are predictable and repeating, interactions with bathymetry produce differences of direction and water speed between ebbing and flooding tides, and path planning capitalizes on those differences. Here was the international opportunity. I approached Matt with idea of collaborating via the GROW program, we prepared and submitted a proposal to test his control algorithms using our µFloats, and so, to abbreviate this lengthy tale, here I am in Queensland.
End note: In my effort to abbreviate the life story, I’ve left out many people who have played significant roles in helping along the way. A huge thanks to all those mentors, colleagues, bosses, friends, and family.
* If there’s documented psychology associated with this phenomenon, please leave a reference in the comments.
** I cringe every time I hear the word ‘drone’ applied to underwater vehicle primarily because the word invokes connotations of political debate. But they are an accurate analogy.
*** In oceanography parlance, there is a loose custom of using ‘drifter’ to refer to surface drifting packages and ‘floats’ to refer to underwater drifting packages. Historically, drifter was first used in reference to drift cards – literally cards dispersed over the ocean surface and watched to assess currents. When underwater floats were developed in the 1950’s, another term was needed to distinguish them from their surface counterpart. Think ‘floating in the water column’ rather than ‘drifting on the surface’.
Header image: A view of the Pacific Ocean.