Short-Term Electricity Load Forecasting with Machine Learning

An accurate short-term load forecasting (STLF) is one of the most critical inputs for power plant units’ planning commitment. STLF reduces the overall planning uncertainty added by the intermittent production of renewable sources; thus, it helps to minimize the hydrothermal electricity production costs in a power grid. Although there is some research in the field and even several research applications, there is a continual need to improve forecasts. This research proposes a set of machine learning (ML) models to improve the accuracy of 168 h forecasts. The developed models employ features from multiple sources, such as historical load, weather, and holidays. Of the five ML models developed and tested in various load profile contexts, the Extreme Gradient Boosting Regressor (XGBoost) algorithm showed the best results, surpassing previous historical weekly predictions based on neural networks. Additionally, because XGBoost models are based on an ensemble of decision trees, it facilitated the model’s interpretation, which provided a relevant additional result, the features’ importance in the forecasting.

Download Full-text

An Extreme Gradient Boosting Algorithm for Short-Term Load Forecasting Using Power Grid Big Data

Proceedings of 2018 Chinese Intelligent Systems Conference - Lecture Notes in Electrical Engineering ◽

10.1007/978-981-13-2288-4_46 ◽

2018 ◽

pp. 479-490

Author(s):

Liqiang Ren ◽

Limin Zhang ◽

Haipeng Wang ◽

Qiang Guo

Keyword(s):

Big Data ◽

Power Grid ◽

Load Forecasting ◽

Gradient Boosting ◽

Short Term ◽

Extreme Gradient Boosting ◽

Short Term Load Forecasting ◽

Boosting Algorithm

Download Full-text

Analysis for Non-Residential Short-Term Load Forecasting Using Machine Learning and Statistical Methods with Financial Impact on the Power Market

Energies ◽

10.3390/en14216966 ◽

2021 ◽

Vol 14 (21) ◽

pp. 6966

Author(s):

Stefan Ungureanu ◽

Vasile Topa ◽

Andrei Cristinel Cziker

Keyword(s):

Machine Learning ◽

Power Systems ◽

Load Forecasting ◽

Power Market ◽

Financial Impact ◽

Electricity Production ◽

Forecast Errors ◽

Short Term ◽

Short Term Load Forecasting ◽

Residential Consumers

Short-term load forecasting predetermines how power systems operate because electricity production needs to sustain demand at all times and costs. Most load forecasts for the non-residential consumers are empirically done either by a customer’s employee or supplier personnel based on experience and historical data, which is frequently not consistent. Our objective is to develop viable and market-oriented machine learning models for short-term forecasting for non-residential consumers. Multiple algorithms were implemented and compared to identify the best model for a cluster of industrial and commercial consumers. The article concludes that the sliding window approach for supervised learning with recurrent neural networks can learn short and long-term dependencies in time series. The best method implemented for the 24 h forecast is a Gated Recurrent Unit (GRU) applied for aggregated loads over three months of testing data resulted in 5.28% MAPE and minimized the cost with 5326.17 € compared with the second-best method LSTM. We propose a new model to evaluate the gap between evaluation metrics and the financial impact of forecast errors in the power market environment. The model simulates bidding on the power market based on the 24 h forecast and using the Romanian day-ahead market and balancing prices through the testing dataset.

Download Full-text

Short and Very Short Term Firm-level Load Forecasting for Warehouses: A Comparison of Machine Learning and Deep Learning Models

10.20944/preprints202201.0107.v1 ◽

2022 ◽

Author(s):

Andrea Maria N. C. Ribeiro ◽

Pedro Rafael X. do Carmo ◽

Patricia Takako Endo ◽

Pierangelo Rosati ◽

Theo Lynn

Keyword(s):

Machine Learning ◽

Deep Learning ◽

Energy Consumption ◽

Load Forecasting ◽

Energy Performance ◽

Gradient Boosting ◽

Support Vector ◽

Learning Models ◽

Short Term ◽

Extreme Gradient Boosting

Commercial buildings are a significant consumer of energy worldwide. Logistics facilities, and specifically warehouses, are a common building type yet under-researched in the demand-side energy forecasting literature. Warehouses have an idiosyncratic profile when compared to other commercial and industrial buildings with a significant reliance on a small number of energy systems. As such, warehouse owners and operators are increasingly entering in to energy performance contracts with energy service companies (ESCOs) to minimise environmental impact, reduce costs, and improve competitiveness. ESCOs and warehouse owners and operators require accurate forecasts of their energy consumption so that precautionary and mitigation measures can be taken. This paper explores the performance of three machine learning models (Support Vector Regression (SVR), Random Forest, and Extreme Gradient Boosting (XGBoost)), three deep learning models (Recurrent Neural Networks (RNN), Long Short-Term Memory (LSTM), and Gated Recurrent Unit (GRU)), and a classical time series model, Autoregressive Integrated Moving Average (ARIMA) for predicting daily energy consumption. The dataset comprises 8,040 records generated over an 11-month period from January to November 2020 from a non-refrigerated logistics facility located in Ireland. The grid search method was used to identify the best configurations for each model. The proposed XGBoost models outperform other models for both very short load forecasting (VSTLF) and short term load forecasting (STLF); the ARIMA model performed the worst.

Download Full-text

Aggregated short-term load forecasting for heterogeneous buildings using machine learning with peak estimation

Energy and Buildings ◽

10.1016/j.enbuild.2021.110742 ◽

2021 ◽

pp. 110742

Author(s):

Amine Bellahsen ◽

Hanane Dagdougui

Keyword(s):

Machine Learning ◽

Load Forecasting ◽

Short Term ◽

Short Term Load Forecasting

Download Full-text

A hybrid method using Exponential Smoothing and Gradient Boosting for electrical short-term load forecasting

2016 IEEE Latin American Conference on Computational Intelligence (LA-CCI) ◽

10.1109/la-cci.2016.7885697 ◽

2016 ◽

Cited By ~ 7

Author(s):

Victor Mayrink ◽

Henrique S. Hippert

Keyword(s):

Hybrid Method ◽

Load Forecasting ◽

Exponential Smoothing ◽

Gradient Boosting ◽

Short Term ◽

Short Term Load Forecasting

Download Full-text

Short Term Load Forecasting of Industrial Electricity Using Machine Learning

Smart Cities - Communications in Computer and Information Science ◽

10.1007/978-3-030-38889-8_12 ◽

2020 ◽

pp. 146-161

Author(s):

Rodrigo Porteiro ◽

Sergio Nesmachnow ◽

Luis Hernández-Callejo

Keyword(s):

Machine Learning ◽

Load Forecasting ◽

Short Term ◽

Short Term Load Forecasting

Download Full-text

Ensemble and Neural Network Machine Learning Models for Short-Term Load Forecasting of Open Cast Mining Companies

Electrotechnical Systems and Complexes ◽

10.18503/2311-8318-2021-3(52)-57-65 ◽

2021 ◽

pp. 57-65

Author(s):

Dmitry Antonenkov ◽

◽

Pavel Matrenin ◽

Keyword(s):

Neural Network ◽

Machine Learning ◽

Load Forecasting ◽

Learning Models ◽

Short Term ◽

Mining Companies ◽

Open Cast Mining ◽

Short Term Load Forecasting ◽

Open Cast ◽

Machine Learning Models

Download Full-text

Review of Machine Learning for Short Term Load Forecasting

10.1007/978-981-16-7210-1_17 ◽

2021 ◽

pp. 180-190

Author(s):

Aijia Ding ◽

Huifen Chen ◽

Tingzhang Liu

Keyword(s):

Machine Learning ◽

Load Forecasting ◽

Short Term ◽

Short Term Load Forecasting

Download Full-text

HousEEC: Day-Ahead Household Electrical Energy Consumption Forecasting Using Deep Learning

Energies ◽

10.3390/en13102672 ◽

2020 ◽

Vol 13 (10) ◽

pp. 2672 ◽

Cited By ~ 1

Author(s):

Ivana Kiprijanovska ◽

Simon Stankoski ◽

Igor Ilievski ◽

Slobodan Jovanovski ◽

Matjaž Gams ◽

...

Keyword(s):

Energy Consumption ◽

Error Score ◽

Load Forecasting ◽

Electrical Energy ◽

Electrical Energy Consumption ◽

Sources Of Information ◽

Multiple Sources ◽

Short Term ◽

Benchmark Model ◽

Short Term Load Forecasting

Short-term load forecasting is integral to the energy planning sector. Various techniques have been employed to achieve effective operation of power systems and efficient market management. We present a scalable system for day-ahead household electrical energy consumption forecasting, named HousEEC. The proposed forecasting method is based on a deep residual neural network, and integrates multiple sources of information by extracting features from (i) contextual data (weather, calendar), and (ii) the historical load of the particular household and all households present in the dataset. Additionally, we compute novel domain-specific time-series features that allow the system to better model the pattern of energy consumption of the household. The experimental analysis and evaluation were performed on one of the most extensive datasets for household electrical energy consumption, Pecan Street, containing almost four years of data. Multiple test cases show that the proposed model provides accurate load forecasting results, achieving a root-mean-square error score of 0.44 kWh and mean absolute error score of 0.23 kWh, for short-term load forecasting for 300 households. The analysis showed that, for hourly forecasting, our model had 8% error (22 kWh), which is 4 percentage points better than the benchmark model. The daily analysis showed that our model had 2% error (131 kWh), which is significantly less compared to the benchmark model, with 6% error (360 kWh).

Download Full-text