LinkNLearn / homedata

Sensor data from a home. More information in the ReadMe file.
2 stars 0 forks source link

Future Data Sources Discussion - for Future Quarters #35

Open pwdel opened 5 years ago

pwdel commented 5 years ago

Possible Source: https://opendata.cityofnewyork.us/

pwdel commented 5 years ago

@nicholaspoggioli I thought you may find this repo interesting...a few years ago I built a super simple solar panel connected to a small 12V Sealed Lead Acid (12V SLA) battery, and I collected data on basically the battery voltage level over time. I had the battery steadily discharging by basically having it run current through a resistor to simulate an electronic draw.

There was a pretty clear linear correlation between the "next day voltage" of the battery and the, "previous day cloudiness" derived from a cloudiness average rating that I calculated taken from a Yahoo weather API. Yahoo Weather gave hourly cloudiness reports for a zone covering the Twin Cities on a scale from 0-100. If the Yahoo Weather cloudiness rating was 0, the battery voltage would be "greater" the next day.

This was not even corrected for temperature and time of year. This data was collected from late summer through the following early summer.

https://github.com/pwdel/Solar-Arduino-Honeybee

What was fun about this project is that it was a device that I both built as well as data that I collected and I was able to create a correlation out of that from a dataset that exists online in the open interwebs.

Since I first worked on this I have learned a lot more about various systems that could be used to essentially build a self-powered, completely independent solar collector device and be able to have the system automatically shut down if it knows that bad weather is coming, similar to how NASA may send a shut down signal to a various parts of the Mars Rover when there is bad weather and dust storms coming on Mars which will eliminate solar energy.

Originally this project was called, "solar Arduino Honeybee" because I had thought that there may be many different applications for independent solar powered units sitting in remote locations for measurement purposes, including measurement of honeybee activity through video (since at the time, there was a lot of news about honeybee collapse being a big concern). Ultimately I did end up talking with a professor from the UMN who specializes in honeybees and she told me that you can't just measure honeybee collapse by counting the number of bees going in and out of a hive - because there are about a thousand reasons why honeybee colonies collapse, and many of those reasons humanity doesn't even yet understand yet. So I guess the "arduino honeybee" moniker is just a code name at this point.

pwdel commented 5 years ago

MN Traffic Data:

https://www.dot.state.mn.us/traffic/data/