Automated Machine Learning pipelines. Builds the Open Short Term Energy Forecasting package.

A curated list of awesome energy forecasting resources such as, code libraries, datasets, courses, tutorials, research papers, competitions and communities.

This study considers the prediction and forecasting of solar and wind power generation on a country-wide basis for the Greek energy grid.

Comparison of LSTM models and FCNN on Energy forecasting project using ASHRAE's data base for the Great Energy Predictor III competition on Kaggle.

Kernel quantile regression

Overview of how the market operates and EDA of some of the variables in it

Solar power production calculator from forecast data.

Energy Consumption Forecating using Stack LSTM, hosted online with Binder

This is a time-series forecast project I did with Max Wang during the Summer 2022 to benchmark Facebook AI's Neuroprophet along other forecasting techniques.

Optimized demand forecasting using time series modeling with Prophet and NeuralProphet. Includes autoregressive memory, holiday effects, time-aware cross-validation, and hyperparameter tuning. Delivers interpretable, multi-horizon predictions for short-term accuracy and long-term grid planning purposes.

Neuro-fuzzy is a repository focused on implementing Adaptive Neuro Fuzzy Inference System (ANFIS) for two distinct applications: Capacitive Deionization and Power Prediction.

A machine learning project to predict household energy consumption using historical smart meter data. Implements data preprocessing, XGBoost regression, and performance evaluation with RMSE, MSE, and MAE metrics.

WattsNext is a time series forecasting project that leverages LSTM neural networks to predict daily electricity consumption using six years of historical data from Finland’s national grid. Designed to enhance energy management and planning.

ioBroker Adapter for solar forecasts from solarprognose.de

Hybrid Renewable Energy Forecasting and Trading Competition: Team Skro

Time-series forecasting of hourly energy consumption using Long Short-Term Memory (LSTM) neural networks. This project explores deep learning models for energy forecasting and compares their performance to traditional statistical models like ARIMA and SARIMA. Part of my transition into computational energy systems research.

ML-based web app that predicts Karnataka's district-wise monthly household electricity consumption using demographic & housing data.

A machine learning project predicting building energy consumption using features like temperature, occupancy, and building type. Includes exploratory data analysis, visualization, and regression modeling. Ideal for understanding key factors affecting energy usage in residential, commercial, and industrial buildings.

A machine learning model that predicts energy consumption (in megawatts) based on time-based input features such as hour of the day and day of the week.