Model
In order to test the robustness of the controller, several scenarios detailed in the controller section need to be run to determine controller success. Success will be indicated by non-stop sourcing of power to the end user, regardless of source, in a profitable manner.
Solar and Wind Generation Profiles
​
The next problem is source material for wind and solar generation. For wind, we intend on using Wunderground's historical information from 2016. Wunderground.com is able to provide hour by hour data about wind speed, which will directly correlate to wind generation data for the LPPs. Solar generation data will follow the daily generation curve provided in the model with peak power during the day being represented by the solar array's maximum capacity. The solar output will then be scaled by clouds to simulate non-linear behavior in the model.
Average temperature and wind data for August 2016, Midland, Mi
Daily temperature and wind data for August 15, 2016, Midland, Mi
Load Profiles for Commercial, Industrial, and Home​
Our simulation and model will include load profiles for 5 different houses, 1 small scale manufacturing industry and a small shopping complex. Having this different load will help to generate better results based on various simulation models.
​
Residential Load - A single small scale house generally uses 30kWh per day. We are using 5 houses for our micro grid simulation and so we will be taking 150kWh per day for the residential load.
This data can be found at https://www.electricchoice.com/blog/electricity-on-average-do-homes/
​
​
​
​
​
​
​
​
​
​
​
​
​
​
Industrial Load - A small scale industry in Michigan would be using an average of 650kWh per day. This would be a small scale industry working only for one morning shift. The industrial load will vary on an individual day depending on productivity of the factory.
https://www.electricitylocal.com/states/michigan/midland/
​
Commercial Load - A small scaled commercial mall in Michigan uses about 265kWh per day. This mall load would be simulated in our micro grid model as constant load for 10 hour assuming the mall is open from 10 am to 8pm.
Wind generation function. Model reflects a 3m/s cut in and a 25m/s cutout
High level overview of the RENU uGrid model.
Simplified sourcing and loading output. Note, powerSecAndRenew should equal powerLoad, thus adding a level of error checking.