make example notebooks for each fault type and better documentation

### TODO

General feed back

General Comments:

Notebook wants an outline (or table of contents) that explains what you're going to learn and in what order. It feels like it jumps around a lot.
In general the a shorter notebook with a specific fault type would be a better introduction. People are overwhelmed and lose interest easily.
I would start each notebook or section with an explanation of the rule: what equipment it works on, explain what the fault condition is, explain how it is detected, and then what input are needed. This will help people follow along rather than feel like they need to memorize steps. The fault for DX cooling control implies what the logic is but doesn't actually explain the rule.
I would break out the trouble shooting stuff into it's own notebook.
The AHU readme does a better job introducing someone to the project.

Other Comments:

I like the fake errors. Very thoughtful. Interesting way to teach.
Might consider showing how to resample and merge data in pandas so people can get a tabular data set like the one used in the example.
Consider demonstrating how to localize timestamps
You launch into ERV fault detection process without explaining what the rule is and why use need SA_FanSpeed and Oa_Temp.
No explanation of why you print all the max values? What do you want the reader to learn from that?

bbartling / open-fdd