kanch91 / project-energy

Time-series analysis of the electricity load consumption in the Mumbai Metropolitan Region.

project-energy

This project is for analysing the time-series algorithms via a study of electricity load forecasting on univariate and multivariate datasets in the MM Region using various classical/moving averages (SMA, WMA, CMA, EMA), exponential smoothing methods (SES, DES, TES), statistical models (ARIMA, SARIMAX) & DL models (LSTM, GRU, RNN) for univariate and multivariate datasets using Keras API

Datasets

It is divided into two primary datasets:

• Univariate dataset having only the load
• Multivariate dataset having weather (Temperature, Dew Point, Humiditiy, Wind Speed, Pressure) as well as the load

Models

Each of the dataset forecasts using the following methods:

• Classical Methods: SMA, WMA, CMA, EMA
• Exponential Methods: SES, DES, TES
• Deep Learning Models: LSTM, GRU, RNN
• Statistical Models: ARIMA, SARIMAX

project-energy's Issues

LSTM, GRU, RNN with weather data

Incorporate the weather data in the LSTM, RNN, GRU network

Hyperparameters selection

Finding the best values of hyperparameters for the LSTM model

Weather data

Scrape the weather data

Build a simple LSTM

LSTM network using the load data only

Predictions of Models with weather data

Check the output of the predicted values in the LSTM, RNN and GRU weather models

Issue using es_models and SARIMAX_model

Hello can u please help me figure this out.
using es_models
Traceback (most recent call last):
File "C:\Users\petuu\Desktop\project-energy-master\es_models.py", line 216, in
mse_tes, rmse_tes, r2_tes, y_tes = triple_exponential_smoothing(season, y_test)
File "C:\Users\petuu\Desktop\project-energy-master\es_models.py", line 121, in triple_exponential_smoothing
tes_model = ExponentialSmoothing(y_test, seasonal_periods=season, trend='add', seasonal='add').fit(use_boxcox=True)
File "G:\anaconda\envs\depp-ts\lib\site-packages\pandas\util_decorators.py", line 207, in wrapper
return func(*args, **kwargs)
File "G:\anaconda\envs\depp-ts\lib\site-packages\pandas\util_decorators.py", line 207, in wrapper
return func(*args, **kwargs)
File "G:\anaconda\envs\depp-ts\lib\site-packages\pandas\util_decorators.py", line 207, in wrapper
return func(*args, **kwargs)
File "G:\anaconda\envs\depp-ts\lib\site-packages\statsmodels\tsa\holtwinters\model.py", line 1094, in fit
raise ValueError(
ValueError: use_boxcox was set at model initialization and cannot be changed

using SARIMAX_model
Traceback (most recent call last):
File "statsmodels\tsa\statespace_kalman_smoother.pyx", line 1000, in statsmodels.tsa.statespace._kalman_smoother.dKalmanSmoother.allocate_arrays
numpy.core._exceptions._ArrayMemoryError: Unable to allocate 697. MiB for an array with shape (51, 51, 35135) and data type float64
Exception ignored in: 'statsmodels.tsa.statespace._kalman_smoother.dKalmanSmoother.reset_filter_method'
Traceback (most recent call last):
File "statsmodels\tsa\statespace_kalman_smoother.pyx", line 1000, in statsmodels.tsa.statespace._kalman_smoother.dKalmanSmoother.allocate_arrays
numpy.core._exceptions._ArrayMemoryError: Unable to allocate 697. MiB for an array with shape (51, 51, 35135) and data type float64
Traceback (most recent call last):
File "C:\Users\petuu\Desktop\project-energy-master\sarimax_model.py", line 85, in
results = model.fit()
File "G:\anaconda\envs\depp-ts\lib\site-packages\statsmodels\tsa\statespace\mlemodel.py", line 728, in fit
res = func(mlefit.params, transformed=False, includes_fixed=False,
File "G:\anaconda\envs\depp-ts\lib\site-packages\statsmodels\tsa\statespace\mlemodel.py", line 886, in smooth
result = self.ssm.smooth(complex_step=complex_step, **kwargs)
File "G:\anaconda\envs\depp-ts\lib\site-packages\statsmodels\tsa\statespace\kalman_smoother.py", line 410, in smooth
smoother = self._smooth(smoother_output, results=results, **kwargs)
File "G:\anaconda\envs\depp-ts\lib\site-packages\statsmodels\tsa\statespace\kalman_smoother.py", line 360, in _smooth
smoother()
File "statsmodels\tsa\statespace_kalman_smoother.pyx", line 1214, in statsmodels.tsa.statespace._kalman_smoother.dKalmanSmoother.call
File "statsmodels\tsa\statespace_kalman_smoother.pyx", line 1239, in statsmodels.tsa.statespace._kalman_smoother.dKalmanSmoother.next
File "statsmodels\tsa\statespace_kalman_smoother.pyx", line 1435, in statsmodels.tsa.statespace._kalman_smoother.dKalmanSmoother.initialize_smoother_object_pointers
AttributeError: Memoryview is not initialized

Scrape Load data

Scrape the load data from MLDC

Add GRU

Predict using a GRU network

Add RNN

Predict using a RNN

R-squared of LSTM, GRU, RNN of multivariate models

Increase the weather dataset

Models to add

• Naive Bayes
• Random Forest
• Gradient boosted trees