# Your project title
# Your name
Due on November 23, Monday at 11:59 PM.
The final project is an individual project, where you apply one or more machine learning techniques to some
power system planning/operation problem. A good project either has the potential to be polished into a
esearch paper, or is suitable to be transformed into teaching materials. I will give you some ideas in the
lectures on October 26.
This template is for your reference. You do not have to strictly follow it. You can remove the instructions.
You need to submit a .zip file, which includes the .Rmd file, the .html file, a data folder, and a figs
folder (if applicable).
The project will be graded by both the instructor and the TA. Please feel free to contact (at least one of) us
(by email or attending office hours) as you work on the project.
The outline of the project is similar to an IEEE research paper.
Introduction
Do not include a reference list in the appendix. Instead, include a link to cite a reference, e.g., a benchmark
model is developed in [Hong11], another model is developed in [Xie18].
Problem Statement
Formulate a problem in power systems from the data analytics/machine learning perspective. Clearly
specify the available data sets, the objective, the performance metric, and the state of the art (best
performance in the cu
ent literature for the same or a similar problem). You can load all the needed
packages here and have a glance at the data:
li
ary(tidyverse)
se <- read_csv("data
se_clean.csv", col_types = cols(Hour = col_factor())) %>%
print()
## # A ti
le: 51,192 x 4
## Date Hour Load T
##
## XXXXXXXXXX XXXXXXXXXX
## XXXXXXXXXX XXXXXXXXXX
## XXXXXXXXXX XXXXXXXXXX
## XXXXXXXXXX XXXXXXXXXX
## XXXXXXXXXX XXXXXXXXXX
## XXXXXXXXXX XXXXXXXXXX
## XXXXXXXXXX XXXXXXXXXX
## XXXXXXXXXX XXXXXXXXXX
## XXXXXXXXXX XXXXXXXXXX
## XXXXXXXXXX XXXXXXXXXX
## # … with 51,182 more rows
You can include code for data preprocessing in either this section or later sections.
https:
doi.org/10.1109/PES XXXXXXXXXX
https:
doi.org/10.1007/s XXXXXXXXXX
Methodology
Describe your proposed methods (e.g., a classical machine learning method and a deep neural network).
Intuitively argue why your methods may be better than the existing methods (in some aspects, of course).
The instructor and the TA are aware that some topics may be more sophisticated, and even reproducing the
state of the art is already challenging enough (which is acceptable in this project; we will take into account
all the factors and grade your work based on the overall quality).
You can include a figure like this:
You can create a math equation like this: \[ E = mc^2. \]
You can include some key code in this section to highlight the main idea and contributions of your work.
Results
Most of your code is included here, with the generated results. You may want to compare your results with
the state of the art, and discuss the results.
Conclusion
Main takeaways of your work, and future directions of your work.