Tane-P5.Rmd

---
title: "P5"
author: "Tane Koh"
date: "2024-04-01"
output: html_document
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
library(tidyverse)
library(readr)
library(lubridate)
library(maps)
library(ggmap)
library(tmap)
library(tmaptools)

register_stadiamaps("8d3c6033-7875-4644-b843-3156014c7e0a", write = TRUE)
```

```{r}
data <- read_csv("camdens big dataset.csv")
cities <- read_csv("MA cities.csv")
```

```{r}
data <- data %>% 
  rename_at('Date In Service', ~'Date') %>% 
  rename_at('Total Cost with Design Fees', ~'Cost') %>%
  rename_at('Total Grant', ~'Grant')

data$Date <- as.Date(data$Date, format = "%m/%d/%Y")
data$Cost <- as.numeric(gsub("[^0-9.]", "", data$Cost))
data$Grant <- as.numeric(gsub("[^0-9.]", "", data$Grant))

cities$name <- str_replace(cities$name, " town$", "")

cities$name <- str_replace(cities$name, " Town$", "")

data <- left_join(data,cities, by = c("City" = "name"))

```

```{r}
city_counts <- data %>%
  group_by(City) %>%
  summarise(count = n()) %>% 
  filter(count > 200)

merged_data <- left_join(city_counts, cities, by = c("City"="name"))

merged_data$ratio <- merged_data$count / merged_data$pop2024

map_MA = get_stadiamap(
  bbox = c(left = -73.1387, bottom = 41.4696, right = -69.8813, top = 42.8599),
  maptype = "stamen_toner_lite",
  zoom = 9
)

ggmap(map_MA)+
  geom_point(
    data = merged_data,
    aes(x = lng, y = lat),
    alpha = .80,
    size = merged_data$ratio*10
  )+
  scale_color_gradient(low = "yellow", high = "red")+
  labs(title = "Solar Installations in MA between 2000 and 2024 per capita.",caption="Restricted to cities with over 200 installations.")
```

```{r}
just_residential <- data %>%
  filter(`Facility Type` == "Commercial / Office")

city_counts <- data %>%
  group_by(City) %>%
  summarise(count = n())

merged_data <- left_join(city_counts, cities, by = c("City"="name"))

merged_data$ratio <- merged_data$count / merged_data$pop2024

map_MA = get_stadiamap(
  bbox = c(left = -73.1387, bottom = 41.4696, right = -69.8813, top = 42.8599),
  maptype = "stamen_toner_lite",
  zoom = 9
)

ggmap(map_MA)+
  geom_point(
    data = merged_data,
    aes(x = lng, y = lat),
    alpha = .80,
    size = merged_data$count/500
  )+
  scale_color_gradient(low = "yellow", high = "red")+
  labs(title = "All Solar Installations in MA between 2000 and 2016.",caption="Not per capita.")
```
```{r}
cost <- data %>% 
  filter(`Facility Type` %in% c("Multi-family residential (4 or more dwelling units per building)", "Residential (3 or fewer dwelling units per building)")) %>% 
  filter(Cost > 1) %>% 
  filter(!is.na(`Capacity (DC, kW)`)) %>% 
  filter(!is.na(`Date`)) %>% 
  filter(!is.na(`Grant`)) %>% 
  filter(!is.na(`Cost`)) %>%
  mutate(Cost_per_kW = `Cost` / `Capacity (DC, kW)`) %>% 
  mutate(Grant_per_kW = `Grant` / `Capacity (DC, kW)`)

average_cost_per_year <- cost %>%
  group_by(Year = format(Date, "%Y")) %>%
  summarise(average_Cost_per_kW = mean(Cost / `Capacity (DC, kW)`))

average_cost_per_year2 <- cost %>%
  group_by(Year = format(Date, "%Y")) %>%
  summarise(mean_cost = mean(Cost / `Estimated Annual Production (kWhr)`))

average_capacity <- cost %>%
  group_by(Year = format(Date, "%Y")) %>%
  summarise(capacity = mean(`Capacity (DC, kW)`))

average_cost_per_city <- cost %>% 
  group_by(City) %>%
  summarise(average_Cost_per_kW = mean(Cost / `Capacity (DC, kW)`))

average_cost_per_city <- left_join(average_cost_per_city, cities, by = c("City"="name"))

ggplot(average_cost_per_year2,aes(x=Year,y=mean_cost,group=1))+
  geom_line()+
  geom_point()+
  labs(title="Yearly Average Cost Per kWhr of Annual Energy Production",y="Average Cost Per kWhr of Annual Energy Production")

ggplot(average_cost_per_year,aes(x=Year,y=average_Cost_per_kW,group=1))+
  geom_line()+
  geom_point()+
  labs(title="Yearly Average Cost Per KW",y="Average Cost Per kW")

ggplot(average_capacity,aes(x=Year,y=capacity,group=1))+
  geom_line()+
  geom_point()+
  labs(title="Trends in Residential Solar Installations Over TIme",y="Average capcity of solar panels (kW)")

ggplot(cost,aes(x=Cost_per_kW))+
  geom_density()+
  xlim(0,10000)+
  labs(title = "Distribution of Cost per kW",x = "Cost per kW ($)")

map_MA = get_stadiamap(
  bbox = c(left = -73.1387, bottom = 41.4696, right = -69.8813, top = 42.8599),
  maptype = "stamen_toner_lite",
  zoom = 9
)



ggmap(map_MA)+
  geom_point(
    data = average_cost_per_city,
    aes(x = lng, y = lat),
    alpha = .80,
    size = average_cost_per_city$average_Cost_per_kW/2000
  )+
  scale_color_gradient(low = "yellow", high = "red")+
  labs(title = "Average Cost per kW of Solar Installations Across MA",caption="")
```

```{r}
Grant_per <- cost %>%
  group_by(City) %>%
  summarise(average_grant_per_kW = mean(Grant / `Capacity (DC, kW)`))
Grant_per <- left_join(Grant_per, cities, by = c("City"="name"))

Grants <- cost %>%
  group_by(City) %>%
  summarise(average_grant = mean(Grant))

Grants <- left_join(Grants, cities, by = c("City"="name"))

ggmap(map_MA)+
  geom_point(
    data = Grant_per,
    aes(x = lng, y = lat),
    alpha = .80,
    size = Grant_per$average_grant_per_kW/500
  )+
  scale_color_gradient(low = "yellow", high = "red")+
  labs(title = "Average Grant per kW of Capacity Across MA",caption="")

ggmap(map_MA)+
  geom_point(
    data = Grants,
    aes(x = lng, y = lat),
    alpha = .80,
    size = Grants$average_grant/100000
  )+
  scale_color_gradient(low = "yellow", high = "red")+
  labs(title = "Average Grant per kW of Capacity Across MA",caption="")

ggplot(cost,aes(x=Grant))+
  geom_density()+
  xlim(0,10000)+
  labs(title = "Distribution of Value of Grants",x = "Value of Grant ($)")

ggplot(cost,aes(x=Grant_per_kW))+
  geom_density()+
  xlim(0,5000)+
  labs(title = "Distribution of Value of Grants per kW",x = "Value of Grant per kW ($)")

average_grant_per_kW_per_year <- cost %>%
  group_by(Year = format(Date, "%Y")) %>%
  summarise(average_grant_per_kW = mean(Grant / `Capacity (DC, kW)`))

average_grant_year <- cost %>%
  group_by(Year = format(Date, "%Y")) %>%
  summarise(grant = mean(`Grant`))

total_grant_year <- cost %>%
  group_by(Year = format(Date, "%Y")) %>%
  summarise(grant = sum(`Grant`))

total_grant_city <- cost %>%
  filter(`Facility Type` == "Residential (3 or fewer dwelling units per building)") %>% 
  group_by(City) %>%
  summarise(grant = sum(`Grant`))
total_grant_city <- left_join(total_grant_city, cities, by = c("City"="name"))

ggplot(average_grant_per_kW_per_year,aes(x=Year,y=average_grant_per_kW,group=1))+
  geom_line()+
  geom_point()+
  labs(title="Yearly Average Grant Per kW",y="Average Grant Per kW ($)")

ggplot(average_grant_year,aes(x=Year,y=grant,group=1))+
  geom_line()+
  geom_point()+
  ylim(0,75000)+
  labs(title="Yearly Average Grant",y="Average Grant ($)")

ggplot(total_grant_year,aes(x=Year,y=grant/1000000,group=1))+
  geom_line()+
  geom_point()+
  ylim(0,40)+
  labs(title="Total Grants Awarded By Year",y="Total Value of Grants Awarded ($M)")

ggmap(map_MA)+
  geom_point(
    data = total_grant_city,
    aes(x = lng, y = lat),
    alpha = .80,
    size = total_grant_city$grant/(100*total_grant_city$pop2024)
  )+
  scale_color_gradient(low = "yellow", high = "red")+
  labs(title = "Total Residential Grants Awarded by City per capita",caption="")
```
```{r}
lm_model <- lm(Grant ~ `Estimated Annual Production (kWhr)`, data = data)

# Create a scatter plot with regression line
ggplot(data, aes(y = Grant, x = `Estimated Annual Production (kWhr)`)) +
  geom_point() +  # Add points
  geom_smooth(method = "lm", se = FALSE, color = "blue") +  # Add linear regression line
  xlim(0,750000)+
  labs(x = "Estimated Annual Production (kWhr)", y = "Grant", title = "Linear Model: Estimated Annual Production vs. Grant") +
  theme_minimal()

lm_model2 <- lm(`Grant` ~ pop2024, data = data)

# Create a scatter plot with regression line
ggplot(cost, aes(x = pop2024, y = `Grant`)) +
  geom_point() +  # Add points
  geom_smooth(method = "lm", se = FALSE, color = "blue") +  # Add linear regression line
  ylim(0,750000)+
  labs(x = "Population", y = "Grant", title = "Linear Model: Estimated Annual Production vs. Grant") +
  theme_minimal()

lm_model3 <- lm(`Grant` ~ pop2024 + `Estimated Annual Production (kWhr)`, data = cost)

intercept <- coef(lm_model3)[1]
slope_production <- coef(lm_model3)[2]
slope_pop2024 <- coef(lm_model3)[3]

# Create the scatter plot with the linear regression line
ggplot(cost, aes(x = `Estimated Annual Production (kWhr)`, y = Grant)) +
  geom_point() +  # Add points
  geom_smooth(method = "lm", se = FALSE, color = "blue", formula = y ~ x + pop2024) +  # Add linear regression line
  labs(x = "Estimated Annual Production", y = "Grant", title = "Linear Model: Grant ~ Estimated Annual Production + pop2024") +
  theme_minimal() +
  xlim(0,700000)+
  annotate("text", x = max(data$`Estimated Annual Production (kWhr)`), y = max(data$Grant), 
           label = paste("Grant =", round(intercept, 2), "+", round(slope_production, 2), "* Production +", round(slope_pop2024, 2), "* pop2024"))
```

```{r}
# make visualization of newer implementations via color
City_color <- cost %>%
  mutate(Year = lubridate::year(Date)) %>% 
  group_by(City) %>%
  summarise(avg_year = mean(Year))
City_color <- left_join(City_color, cities, by = c("City"="name"))

ggmap(map_MA) +
  geom_point(
    data = City_color,
    aes(x = lng, y = lat, color = avg_year),
    alpha = 0.80,
    size = 3
  ) +
  scale_color_gradient(
    low = "white",
    high = "blue",
    limits = range(City_color$avg_year),
    breaks = pretty(range(City_color$avg_year), n = 5)
  ) +
  labs(
    title = "Average Year of Solar Project Installation",
    caption = ""
  ) +
  theme(legend.position = "bottom") +
  guides(color = guide_colorbar(title = "Average Year"))

# visualization of price rate change specific to commercial/residential facet plot?
cost2 <- cost %>%
  mutate(`Facility Type` = case_when(
    `Facility Type` %in% c("Multi-family residential (4 or more dwelling units per building)", 
                         "Residential (3 or fewer dwelling units per building)") ~ "Residential",
    `Facility Type` %in% c("Industrial", "Commercial / Office", "Restaurant / Food Service", "Retail") ~ "Industrial",
    `Facility Type` %in% c("College / University", "Community Solar", "Federal", 
                         "Hospital / Health Care", "Municipal - K-12 School", 
                         "Municipal / Government / Public", "School (K-12)", "State", 
                         "State - College/University") ~ "Government",
    `Facility Type` %in% c("Other", "Agricultural", "Religious", "Mixed use (commercial & residential)") ~ "Other"
  ))

facet_data <- cost2 %>%
  mutate(Year = lubridate::year(Date)) %>% 
  group_by(`Facility Type`, Year) %>%
  summarise(average_cost = mean(Cost / `Estimated Annual Production (kWhr)`))

ggplot(facet_data, aes(x = Year, y = average_cost, color = `Facility Type`)) +
  geom_line() +
  labs(x = "Year", y = "Average Cost per Annual Production", title = "Cost per Annual Production by Facility Type") +
  theme_minimal()+
  ylim(0,12)

facet_grant <- cost2 %>%
  mutate(Year = lubridate::year(Date)) %>% 
  group_by(`Facility Type`, Year) %>%
  summarise(avg_grant = mean(Grant / `Estimated Annual Production (kWhr)`))

ggplot(facet_grant, aes(x = Year, y = avg_grant, color = `Facility Type`)) +
  geom_line() +
  labs(x = "Year", y = "Average Grant per Annual Production", title = "Grant per Annual Production by Facility Type") +
  theme_minimal()+
  ylim(0,7)

## Grants awarded by city per capita but limited to residential
```

```{r}
cost$Year <- format(cost$Date, "%Y")

# Aggregate data by year to calculate average Capacity and Production
aggregate_cost <- cost %>%
  group_by(Year) %>%
  summarise(avg_capacity = mean(`Capacity (DC, kW)`),
            avg_production = mean(`Estimated Annual Production (kWhr)`))

ggplot(aggregate_cost, aes(x = Year)) +
  geom_line(aes(y = avg_capacity*1000, color = "Capacity"), group = 1) +
  geom_line(aes(y = avg_production, color = "Production"), group = 1) +
  labs(title = "Capacity and Production of Residential Solar Installations Over Time",
       x = "Date",
       y = "Average Estimated Annual Production (kWhr)",
       color = "Variable") +
  scale_color_manual(values = c("Capacity" = "blue", "Production" = "red")) +
  theme_minimal() +
  # Adding a secondary y-axis for Production
  scale_y_continuous(
    sec.axis = sec_axis(~./1000, name = "Average Capacity (kW)")
  )
```
```{r}

aggr_money <- cost %>%
  group_by(Year) %>%
  summarise(cost = mean(Cost/`Estimated Annual Production (kWhr)`),
            grant = mean(Grant/`Estimated Annual Production (kWhr)`))

ggplot(aggr_money, aes(x = Year)) +
  geom_line(aes(y = cost, color = "Average Cost of Installation"), group = 1) +
  geom_line(aes(y = grant*2, color = "Average Awarded Grant Value"), group = 1) +
  labs(title = "Cost of Residential Solar Installations and Grant Values per kWhr over time",
       x = "Date",
       y = "Cost ($/kWhr)",
       color = "Variable") +
  scale_color_manual(values = c("Average Cost of Installation" = "blue", "Average Awarded Grant Value" = "red")) +
  # Adding a secondary y-axis for Production
  scale_y_continuous(
    sec.axis = sec_axis(~./2, name = "Grant ($/kWhr)")
  )

```
```{r}
# average per year amount grant vs number of grants
grant_per <- cost %>% 
  

```

# number per funding type faceted by year