# -*- coding: utf-8 -*-
"""
Created on Tue Sep 10 12:51:48 2019
@author: IST_1
"""
import helper
import pandas as pd
import numpy as np
solar = helper.pv_data(address='Lancaster, Massachusetts',
solar_size_kw_dc=6032,
tilt=23.45)
solar['year'] = 2018
total_output = solar.solar_output_kw.sum()
print(total_output)
date = pd.DataFrame()
date['year'] = solar.year
date['month'] = solar.month
date['day'] = solar.day
date['hour'] = solar.hour
power = 2400
discharge_time = 2
peak_duration = 4
assumed_duration_multiplier = peak_duration / discharge_time
#this is to calculate utilization based on the number of peak hours and discharge time
#calculated by number of peak hours / discharge_time
eff = 0.9
import pandas as pd
import pvlib
import datetime
import helper
import numpy as np
from pvlib import clearsky, atmosphere, solarposition
from pvlib.location import Location
from pvlib.iotools import read_tmy3
from pvlib.forecast import GFS, NAM, NDFD, HRRR, RAP
import ml
date = pd.date_range(start='1/1/2018', end='12/31/2018', freq='H')
solcast = pd.read_csv('new_york_solcast_data.csv')
pv = helper.pv_data(address='New York, New York',
solar_size_kw_dc=1000,
tilt=0)
dates = pd.to_datetime(solcast.Date)
solcast.Date = dates
#dirint_dni = pvlib.irradiance.dirint(solcast.Ghi,solcast.Zenith,times=pd.to_datetime(date))
solcast['Month'] = solcast.Date.dt.month
solcast['Day'] = solcast.Date.dt.day
solcast['Hour'] = solcast.Date.dt.hour
solcast = solcast.drop(columns=['Date'])
solcast = solcast.reindex(index=np.roll(solcast.index, -4)).reset_index(
drop=True)
solcast['solar_output_kw'] = pv.solar_output_kw
solcast = solcast.drop(solcast.loc[(solcast.Dni == 0) & (solcast.Dhi == 0) &
(solcast.Ghi == 0)].index)
target_output = ['Dhi', 'Ghi', 'Dni']
#solcast.loc[solcast.Azimuth<=-1] = solcast.loc[solcast.Azimuth<=-1] + 360
"""
Created on Thu Aug 8 14:37:04 2019
@author: IST_1
"""
import csv
import pandas as pd
import numpy as np
import helper
txt_file = r"s3159989.txt"
csv_file = r"solar_data.csv"
with open(txt_file, 'r') as infile, open(csv_file, 'w') as outfile:
stripped = (line.strip() for line in infile)
lines = (line.split(",") for line in stripped if line)
writer = csv.writer(outfile)
writer.writerows(lines)
df = pd.read_csv('solar_data.csv')
tilt = -20
angle_of_array = 180
aoi = np.arccos((np.cos(df['Zenith (refracted)']) * np.cos(tilt)) +
(np.sin(df['Zenith (refracted)']) * np.sin(tilt) *
np.cos(df['Azimuth angle'] - angle_of_array)))
dni_df = helper.pv_data(address='New York, NY', solar_size_kw_dc=1230, tilt=20)
dni_df.direct_normal_irradiance = dni_df.direct_normal_irradiance * aoi
output = ml.load_models('entersolar_model.h5')
import importlib
importlib.reload(ml)
start_idx = 0
length = 2000
target_names = target_output
y_pred, y_true,x,mse = ml.plot_comparison(x_test_scaled,y_test,y_scaler,
target_names,output,
start_idx = start_idx,
length=length,
verbose=True)
solar_pvwatts = helper.pv_data(coordinates=(lat,lon),solar_size_kw_dc = 1750,tilt=25)
index = pd.date_range(start='1/1/'+str(2018), periods=8760, freq='60T')
solar_pvwatts.index = index
solar_pvwatts = solar_pvwatts.resample('15T').asfreq()
#solar_pvwatts = solar_pvwatts.drop(columns={'month','day','hour'})
solar_pvwatts = solar_pvwatts.interpolate()
solar_pvwatts = solar_pvwatts.reset_index(drop=True)
solar_pvwatts = solar_pvwatts[seg_start:seg_end]
segment_pvwatts = solar_pvwatts[-(int(np.ceil(len(segment)-0.9*(len(segment))))+1):-1]
plt.figure(figsize=(15,5))
#plt.plot(segment_pvwatts.index,segment_pvwatts.solar_output_kw,'bo',alpha=0.5)
plt.plot(segment_pvwatts.index,y_pred)
import ml
import matplotlib.pyplot as plt
import tensorflow as tf
import numpy as np
import pandas as pd
import os
from sklearn.preprocessing import MinMaxScaler
from tensorflow.python.keras.models import Sequential
from tensorflow.python.keras.layers import Input, Dense, GRU, Embedding, LSTM
from tensorflow.python.keras.optimizers import RMSprop
from tensorflow.python.keras.callbacks import EarlyStopping, ModelCheckpoint, TensorBoard, ReduceLROnPlateau
import keras
import helper
import numpy as np
solar = helper.pv_data(address = 'New York, NY', solar_size_kw_dc = 200, tilt = 20)
import pandas as pd
data = pd.read_csv('nyc_data2017.csv')
data.columns = ['year','month','day','hour','minute','dhi','dni','ghi','x','surface_albedo','xx','pressure']
data = data.drop([0,1])
data = data.reset_index(drop=True)
data = data.drop(columns=['x','xx','minute','year','month','day','hour'])
data['solar_output_kw'] = solar.solar_output_kw
#data['solar_irradiance'] = solar.solar_irradiance
#solar_prev = solar.reindex(index=np.roll(solar.index,1*24)).reset_index(drop=True)
#solar_2prev = solar.reindex(index=np.roll(solar.index,2*24)).reset_index(drop=True)
#dni_prev = data.reindex(index=np.roll(data.index,1*24)).reset_index(drop=True).dni
#dni_2prev = data.reindex(index=np.roll(data.index,2*24)).reset_index(drop=True).dni
#dhi_prev = data.reindex(index=np.roll(data.index,1*24)).reset_index(drop=True).dhi
from sklearn.ensemble import RandomForestClassifier
import numpy as np
df = pd.read_excel('File 2.xlsx')
df = df[['Device name', 'Unnamed: 8']]
df = df.drop([0, 1, 2, 3, 4, 5, 6])
df = df.rename(columns={
'Device name': 'date',
'Unnamed: 8': 'original_building_power_kw'
})
df.date = pd.to_datetime(df.date)
df.original_building_power_kw = df.original_building_power_kw / 1000
df['original_building_energy_kwh'] = df.original_building_power_kw / 4
df = df.sort_values(by='date')
df = df.reset_index(drop=True)
solar = helper.pv_data(address='Chirin Gita, Mexico', solar_size_kw_dc=80)
df = helper.add_solar_to_df(df, solar, 4)
#f['y'] = 'yes'
#df['target'] = df['y'].apply(lambda x: 1 if x=='yes' else 0)
df.isnull().mean().sort_values(ascending=False) * 100
#df.drop('y',axis=1,inplace=True)
df.original_building_power_kw = df.original_building_power_kw.astype(float)
df.original_building_energy_kwh = df.original_building_energy_kwh.astype(float)
df.power_post_solar_kw = df.power_post_solar_kw.astype(float)
df.energy_post_solar_kwh = df.energy_post_solar_kwh.astype(float)
#df['month'] = df.date.dt.month.astype(float)
#df['day'] = df.date.dt.day.astype(float)
#df['hour'] = df.date.dt.hour.astype(float)
corr = df.corr()