Global Power Plant Database¶
https://github.com/wri/global-power-plant-database/
Initial Exploration with Power BI¶
Thanks to Timo for the nice introduction and demonstration of Power BI. Power BI is freely available for personal use: https://www.microsoft.com/en-US/download/details.aspx?id=58494
Questions:¶
- Generic
- What kind of information is stored in the table?
- How much data is missing?
- Is the dataset clean or are there any clear outliers?
- Specific
- Which country has the highest number of powerplants (how many, top 10)?
- Which is the country with the highest power preduction (MW) per continent (how much, top 5)
- What is the overall proportion of each energy source (by plant number, by MW produced)
- Which is the greenest country (highest fraction of power from renewable sources)?
- How does the distribution of power plants look on a worldmap (by type, interactive)?
Rotation¶
- Markus
- Stefan
- Timo
In [1]:
library(tidyverse)
In [2]:
power_plants <- readr::read_csv('https://raw.githubusercontent.com/wri/global-power-plant-database/master/output_database/global_power_plant_database.csv')
What kind of information is in the table?¶
In [3]:
str(power_plants)
Missing Values¶
In [4]:
power_plants %>% select(everything()) %>%
summarise_all(list(~sum(is.na(.)))) %>%
pivot_longer(country:estimated_generation_gwh)%>%filter(value<1000)
Out[4]:
Outliers? Distribution¶
In [5]:
ggplot(data=power_plants) + geom_density(aes(x=capacity_mw)) + scale_x_log10()
Out[5]:
Top power plant and producers (country)¶
In [6]:
power_plants %>% filter(capacity_mw==max(capacity_mw))
Out[6]:
In [7]:
power_plants %>% count(country_long, sort=T) %>% head(10)
Out[7]:
In [8]:
power_plants %>% group_by(country_long,) %>% summarize(capacity = sum(capacity_mw)) %>% arrange(-capacity) %>% head(20)
Out[8]:
Include primary fuel type¶
In [9]:
unique(power_plants$primary_fuel)
Out[9]:
In [10]:
power_plants %>%
group_by(country_long,primary_fuel) %>%
summarize(capacity = sum(capacity_mw), number_powerplants=n()) %>% arrange(-capacity) %>% head
Out[10]:
In [11]:
options(repr.plot.res = 250, repr.plot.width = 14)
In [12]:
power_plants %>%
mutate(
country_long = fct_lump(country_long, 9),
primary_fuel = fct_lump(primary_fuel, 8, w=capacity_mw)
) %>%
group_by(country_long,primary_fuel) %>%
summarize(capacity = sum(capacity_mw), number_powerplants=n()) %>% arrange(-capacity) %>%
ungroup() %>%
mutate(country_long = fct_reorder(country_long, capacity, .fun=sum)) %>%
ggplot(aes(y=country_long, x=capacity, fill=primary_fuel)) +
geom_col() +
scale_fill_viridis_d(option = "F") +
theme_minimal() +
labs(y="Country", x = "Capacity") #+ facet_wrap(~primary_fuel, scale="free_x")
Out[12]:
In [13]:
power_plants %>%
mutate(
country_long = fct_lump(country_long, 9),
#primary_fuel = fct_lump(primary_fuel, 8, w=capacity_mw)
) %>%
group_by(country_long,primary_fuel) %>%
summarize(capacity = sum(capacity_mw), number_powerplants=n()) %>% arrange(-capacity) %>%
ungroup() %>%
mutate(primary_fuel = fct_reorder(primary_fuel, capacity, .fun=sum)) %>%
ggplot(aes(y=primary_fuel, x=capacity, fill=primary_fuel)) +
geom_col() +
scale_fill_viridis_d(option = "F", direction = -1) +
theme_minimal() +
labs(y="Country", x = "Capacity")
Out[13]:
In [14]:
power_plants %>%
group_by(country_long,primary_fuel) %>%
summarize(number_powerplants=n()) %>%
pivot_wider(names_from=primary_fuel, values_from=number_powerplants, values_fill=0) %>%
identity %>% head
Out[14]: