Final Blog Post, by Naveen Bahadur

Research Abstract:

When planning to deploy grid sensing devices on hundreds of power poles, it is very important
to ensure that these devices can be powered reliably. In my case, this meant designing the most
robust solar system possible for Gridware's devices. Using solar power for these devices presents
many unique challenges such as variance in solar irradiance due to weather conditions and
foliage along with the risk of efficiency losses due to dust or oddities like bird droppings. In
order for the devices to excel at detecting faults in real-time, they must be operational 99.9% of
the time, which translates to a downtime of only 10 minutes per week. This does not leave much
room for error in our power budget, so it is essential that the solar system is as optimized as
possible and exceeds expectations to avoid any potential drops in power.

To understand the options currently available, I began surveying the solar panel market. I started
aquiring cost and lead time estimates from major distributors and eventually ordered a few
different sample panels for testing purposes. While waiting for the panels to arrive, I built a
testing circuit that allowed me to vary the panel voltage by turning a potentiometer and measure
the current to generate P-V curves. Once the panels starting coming in, I began characterizing
them under varied conditions. These conditions included time of day, shading, angle from
vertical, and orientation. In the end, I generated a total of around 40 P-V curves that evaluated
the performance of the sample panels under a large variety of conditions. These curves will help
the company set the optimal panel voltage to achieve maximum power.

In the final two weeks of my internship, I focused all of my time on building a system that can
record solar panel data across an entire day. I built the device using a Raspberry Pi, current
sensor, and a modification of the testing circuit I built before. I programmed the device such that
it would record time, voltage, current, and power into a .csv file every minute, starting from the
moment it is powered on. I built three of these devices total, so that I could compare performance
when facing South, West, and East. Although there were some technical difficulties with the
West-facing device on the day of the test, I was able to determine that the South facing panel
generates about 3 watt-hours per day more than the East facing panel (11Wh vs 8Wh).

Although I was able to obtain a fair amount of data regarding the device's power, the one thing I
would have wanted to do if given more time is investigate coating options for the solar panels.
There are some coatings that have the ability to repel water, dust, and dirt from the surface of the
panels. This would improve their efficiency while also reducing the need for manual cleaning
and maintenance. It would be interesting to test these coatings and see if the benefits outweigh
the costs.

Personal Reflection:

Considering that I had never worked in the industry before, I could not be more grateful for the
opportunity I had to work at Gridware this past summer. I know it's cliché, but I truly believe I
gained more applicable knowledge in my three months at Gridware than I did in my first couple
years in college. I really enjoyed experiencing firsthand the differences between working and
being a student. On one hand, it can be more stressful because the company is relying on you to
do your part for the product to be successful. However, it is so reassuring to be working among a
team of individuals who are just as passionate as I am about energy solutions and are happy to
help guide me through any obstacles I face.

Working virtually for the first two weeks felt a little bit isolating at times since I only had about
an hour or two of face to face meetings per day. It was difficult to get caught up to speed with
how the company operates, but once I started working in person at the office, I was able to
connect with the other engineers and learn a ton from just our casual lunch conversations. I was
also especially lucky to be assigned a project that I am very passionate about and have some
expertise in from various classes. I truly can not thank my director and mentors enough for
setting me up so well for a successful project!

One of my main goals for the internship was to learn how to effectively apply technical
knowledge to an industry setting, since this isn't really something that is taught in school. Having
a full hardware lab on-site was really convenient and made it easy to conduct small research
projects and experiment with building different devices to help acquire the data I needed. I felt
much more motivated to work hard on my internship projects than my school projects because I
was more eager to obtain meaningful results that would help the company progress.

Although I have no reference to an experience at a larger company, I really enjoyed the flexible
work culture at Gridware and would definitely want to work for a startup in the future. I was
very fond of the fast-paced environment and the ability to execute my vision for the project in
such a short amount of time by avoiding the red tape that comes with working for a larger
corporation. The summer went by extremely fast, but I'm really proud of the knowledge I gained
and the contributions I made to the company. In three months I went from barely knowing how
to solder to building a system that can record solar panel data across a full day. I can't wait to
learn more in future industry experiences and continue to progress my career in energy
engineering.