K-Nearest Neighbor Regression for Forecasting Electricity Demand

Atanasovski, Metodija; Kostov, Mitko; Arapinoski, Blagoja; Spirovski, Mile

doi:10.1109/icest49890.2020.9232768

Cited by 11 publications

(2 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The KNN algorithm is a non-parametric method used for regression (and classification). It predicts the value of a new point based on the 'K' nearest points in the training dataset [28,29]. The output is typically the average of the values of its nearest neighbors.…”

Section: Methodsmentioning

confidence: 99%

Comparison of machine learning models in forecasting different ENSO types

Ibebuchi,

Rainey,

Obarein

et al. 2024

Phys. Scr.

View full text Add to dashboard Cite

Accurate forecasting of the El Niño Southern Oscillation (ENSO) plays a critical role in mitigating the impacts of extreme weather conditions linked to ENSO variability on ecosystems. This study evaluates the performance of six machine learning models in forecasting two ENSO types: the Central Pacific El Niño (Niño 4 index) and the East Central Pacific El Niño (Niño 3.4 index). The models analyzed include the Feed Forward Neural Network (FFNN), Long Short-term Memory (LSTM) neural network, eXtreme Gradient Boosting Regressor, K-Nearest Neighbors Regressor, Gradient Boosting Regressor, and Support Vector Regressor, using the ENSO index lagged by six months as the predictor. The models were trained on the monthly ENSO indices from 1870 to 1992 and tested from 1993 to 2023. We also assess the relative predictability of the two ENSO types. Events were defined as when the ENSO index exceeded ±0.4. Our evaluation during the testing period reveals that for the analyzed models, the deep neural network models (LSTM and FFNN) demonstrated superior performance in forecasting ENSO at a 6-month lead time. Furthermore, all models achieved impressive all-season correlations ranging from 0.93 to 0.97 and threat score for the ENSO phases between 0.71 to 0.88 for Niño 3.4 events, and 0.72 to 0.93 for Niño 4 events. The predictability of the two ENSO types depended on the model and strength of the ENSO event. Considering both ENSO phases, La Niña events were forecasted with a higher accuracy relative to El Niño events, and all models, besides the deep learning models, notably fell short in capturing the extreme 2015/2016 Central Pacific El Niño event. These results highlight the potential of machine learning models, particularly the deep learning approaches, for skillful ENSO forecasting, by leveraging its historical data.

show abstract

Section: Methodsmentioning

confidence: 99%

Comparison of machine learning models in forecasting different ENSO types

Ibebuchi,

Rainey,

Obarein

et al. 2024

Phys. Scr.

View full text Add to dashboard Cite

show abstract

“…The KNN algorithm assumes that similar features characterize the points that are near each other. An example of using KNN in short-term load forecasting can be found in [17,18]. Decision Tree (DT) and its derivatives form an important class of ML algorithms.…”

Section: Machine Learning Techniquesmentioning

confidence: 99%

Energy Load Forecasting Techniques in Smart Grids: A Cross-Country Comparative Analysis

Hachache,

Labrahmi,

Grilo

et al. 2024

Energies

View full text Add to dashboard Cite

Energy management systems allow the Smart Grids industry to track, improve, and regulate energy use. Particularly, demand-side management is regarded as a crucial component of the entire Smart Grids system. Therefore, by aligning utility offers with customer demand, anticipating future energy demands is essential for regulating consumption. An updated examination of several forecasting techniques for projecting energy short-term load forecasts is provided in this article. Each class of algorithms, including statistical techniques, Machine Learning, Deep Learning, and hybrid combinations, are comparatively evaluated and critically analyzed, based on three real consumption datasets from Spain, Germany, and the United States of America. To increase the size of tiny training datasets, this paper also proposes a data augmentation technique based on Generative Adversarial Networks. The results show that the Deep Learning-hybrid model is more accurate than traditional statistical methods and basic Machine Learning procedures. In the same direction, it is demonstrated that more comprehensive datasets assisted by complementary data, such as energy generation and weather, may significantly boost the accuracy of the models. Additionally, it is also demonstrated that Generative Adversarial Networks-based data augmentation may greatly improve algorithm accuracy.

show abstract