The functions in ELA can be used independently of the user interface widgets to explore the energy generation and storage data and the K nearest neighbors model. Selected functions in the ela and mapping modules are described and demonstrated here.



In [1]:

    
import ela
import folium
%matplotlib inline

Lookup & prediction functions: `ela.py`

1. get_latlon_from_zip ()

This function needs only one input which is the zipcode.
This function enables the user to input a zipcode and find the respective latitude and longitude for it.



In [2]:

    
ela.get_latlon_from_zip(98105)









    Out[2]:





(47.66377, -122.30118)

2. get_state_from_zip()

This function needs only 1 input which is the zipcode.
This function enables the user to input a zipcode and find the respective state for it.



In [3]:

    
ela.get_state_from_zip(98105)









    Out[3]:





'WA'

3. get_closest_facility()

This function needs two inputs, a tuple of length equal to 2 and a string 'gen' or 'stor'
This function enables the user to input a tuple of latitude and longitude and find the closest generation or storage facility for it.
The energy generation or storage facility is returned as a pandas dataframe row object, but attributes of the facility can be selected for output.



In [4]:

    
ela.get_closest_facility((47.66377, -122.30118),'gen')









    Out[4]:





Unnamed: 0                                        8179
state                                               WA
name              University of Washington Power Plant
lat                                            47.6539
lon                                           -122.304
type                                               Gas
capacity_MW                                         15
production_GWh                                 12.3075
Name: 8142, dtype: object

Alternatively the get_latlon_from__zip function can be called and that will provide the tuple of lat-lon.



In [5]:

    
ela.get_closest_facility(ela.get_latlon_from_zip(98105),'stor')









    Out[5]:





Unnamed: 0                             110
name          Seattle City Light Microgrid
lat                                47.6062
lon                               -122.332
type2                     Electro-chemical
type                      Electro-chemical
power_kW                                 0
duration                               n/a
status                           Announced
usecase         Electric Energy Time Shift
city                               Seattle
state                                   WA
Name: 110, dtype: object

Printing formatted output from the returned dataframe row:



In [6]:

    
my_fac = ela.get_closest_facility((47.66377, -122.30118),'gen')
print("The nearest energy generation facility is {}, which is a {} \
facility with a capacity of {} MW.".
      format(my_fac['name'], my_fac['type'], my_fac.capacity_MW))









    



The nearest energy generation facility is University of Washington Power Plant, which is a Gas facility with a capacity of 15.0 MW.

4. get_state_breakdown()

This function needs two inputs, a state in the two letter abbreviation and a pandas dataframe containing energy generation or storage facility data.
This function enables the user to find the statewise energy generation and storage breakdown. This gives the percentage of energy facilities (not total energy production or storage capacity) of each type in that state.



In [7]:

    
ela.get_state_breakdown('WA', ela.gen_data)









    Out[7]:






  
    
      
      Hydro
      Wind
      Biomass
      Gas
      Oil
      Coal
      Other
      Nuclear
    
  
  
    
      WA
      53%
      15%
      13%
      11%
      3%
      1%
      0%
      0%



In [8]:

    
ela.get_state_breakdown('WA', ela.stor_data)









    Out[8]:






  
    
      
      Electro-chemical
      Pumped Hydro Storage
    
  
  
    
      WA
      94%
      5%

5. get_energy_breakdown()

This function needs one input,a pandas dataframe containing energy generation or storage facility data.
This function enables the user to find the energy generation or storage breakdown in the U.S. This gives the percentage of energy facilities (not total energy production or storage capacity) of each type in the U.S.



In [9]:

    
ela.get_energy_breakdown(ela.gen_data)









    Out[9]:






  
    
      
      Gas
      Solar
      Hydro
      Wind
      Oil
      Biomass
      Coal
      Other
      Geothermal
      Nuclear
    
  
  
    
      US
      22%
      18%
      17%
      12%
      10%
      9%
      6%
      1%
      0%
      0%



In [10]:

    
ela.get_energy_breakdown(ela.stor_data)









    Out[10]:






  
    
      
      Electro-chemical
      Thermal Storage
      Pumped Hydro Storage
      Electro-mechanical
    
  
  
    
      US
      66%
      21%
      7%
      5%

6. state_type()

This function needs one input,a pandas dataframe containing energy generation or storage facility data.
This function returns the same information as get_state_breakdown(), but as a pandas dataframe containing the energy type percentages for all U.S. states, rather than just for one state.
This gives the percentage of energy facilities (not total energy production or storage capacity) of each type in that state.



In [11]:

    
ela.state_type(ela.gen_data).head()

7. get_predicted_type()

This function needs two inputs, a tuple of length equal two for the lat-lon and a string 'gen' or 'stor'.
This function enables the user to predict the type of generation or storage for that location, based on ELA's built-in KNN model.



In [12]:

    
ela.get_predicted_type((47.66377, -122.30118),'gen')









    Out[12]:





'Gas'



In [13]:

    
ela.get_predicted_type((47.66377, -122.30118),'stor')









    Out[13]:





'Electro-chemical'

8. graph_state_breakdown()

This function needs two inputs, a state in the two letter abbreviation and a pandas dataframe 'gen' or 'stor'
This function enables the user to visualize the statewise energy generation or storage breakdown in a pie chart.
This gives the percentage of energy facilities (not total energy production or storage capacity) of each type in that state.



In [14]:

    
ela.graph_state_breakdown('WA', 'gen')



In [15]:

    
ela.graph_state_breakdown('WA', 'stor')

Mapping functions: `mapping.py`

Most of the functions in mapping are for processing GeoJSON files and making predictions using ela.get_predicted_type. The mapping functions of primary interest to a general user are described below.

The GeoJSON objects needed for visualizing state- or county-level predictions are included with ELA and can be accessed as ela.states and ela.counties.

1. prediction_map()

Two inputs are provided for the function, a GeoJSON object containing state or county boundaries and a string, either gen or stor.
This function then returns a Folium map layer showing the input geographic features colored according to their predicted energy types.



In [16]:

    
m = folium.Map(location=[35,-97], zoom_start=4)
f = ela.prediction_map(ela.states, 'gen')
m.add_child(f)









    Out[16]:



In [17]:

    
m = folium.Map(location=[35,-97], zoom_start=4)
f = ela.prediction_map(ela.counties, 'stor')
m.add_child(f)









    Out[17]:

2. facility_map()

Two inputs are added to this function, a string containing the two letter abbreviation of the state and the string 'gen' or 'stor'.
This function then returns a folium map layer containing a marker for each facility.



In [18]:

    
m = folium.Map(location=ela.get_latlon_from_zip(98105), zoom_start=6)
f = ela.facility_map('WA', 'stor')
m.add_child(f)









    Out[18]:



In [19]:

    
m = folium.Map(location=ela.get_latlon_from_zip(98105), zoom_start=6)
f = ela.facility_map('WA', 'gen')
m.add_child(f)









    Out[19]:

3. state_map()

This function needs a single input, a two letter abbreviation of a state.
This function generates a blank folium map centered at the input state.



In [20]:

    
ela.state_map('NY')









    Out[20]:

Combined mapping functions

These functions can be combined to display any combination of generation or storage and prediction or facilities.



In [21]:

    
m = ela.state_map('WA')
prediction = ela.prediction_map(ela.counties, 'gen')
facilities = ela.facility_map('WA', 'gen')
m.add_child(prediction)
m.add_child(facilities)









    Out[21]:



In [ ]:

	Biomass	Coal	Gas	Geothermal	Hydro	Nuclear	Oil	Other	Solar	Wind
AK	0%	3%	9%	0%	22%	0%	58%	1%	0%	2%
AL	13%	15%	32%	0%	30%	2%	3%	2%	0%	0%
AR	15%	7%	38%	0%	30%	1%	4%	0%	1%	0%
AZ	4%	4%	25%	0%	9%	0%	1%	0%	48%	4%
CA	9%	0%	23%	3%	19%	0%	1%	1%	32%	9%

	Biomass	Coal	Gas	Geothermal	Hydro	Nuclear	Oil	Other	Solar	Wind
AK	0%	3%	9%	0%	22%	0%	58%	1%	0%	2%
AL	13%	15%	32%	0%	30%	2%	3%	2%	0%	0%
AR	15%	7%	38%	0%	30%	1%	4%	0%	1%	0%
AZ	4%	4%	25%	0%	9%	0%	1%	0%	48%	4%
CA	9%	0%	23%	3%	19%	0%	1%	1%	32%	9%

Lookup & prediction functions: ela.py