What question did this study set out to answer?

To develop and evaluate a multi-layer framework based on deep extreme learning machines for accurate hourly energy consumption prediction in residential buildings.

September 28, 2018ElectronicsOpen Access

A Prediction Methodology of Energy Consumption Based on Deep Extreme Learning Machine and Comparative Analysis in Residential Buildings

View Full Paper

Ask AI

Bookmark

Authors

MFMuhammad FayazSejong University DKDo‐Hyeun KimJeju National University

Discussion

Member takes

Implication

Computational study demonstrates superior hourly energy prediction with deep extreme learning machines in residential buildings, indicating high accuracy over traditional neural network models.

Key Points

To develop and evaluate a multi-layer framework based on deep extreme learning machines for accurate hourly energy consumption prediction in residential buildings.
Collected real-world energy consumption data from four multi-storied residential buildings and implemented data-cleaning schemes to eliminate anomalies.
Constructed a four-layer architecture (data acquisition, preprocessing, prediction, and performance evaluation) testing deep extreme learning machines (DELM), artificial neural networks (ANN), and adaptive neuro-fuzzy inference systems (ANFIS) across multiple activation and membership functions.
Evaluated model accuracy for one-week and one-month hourly forecasting intervals using mean absolute error (MAE), root mean square error (RMSE), and mean absolute percentage error (MAPE).
DELM demonstrated substantially lower forecasting errors than ANN and ANFIS across both one-week and one-month hourly prediction periods.
Optimizing the number of hidden layers, hidden neurons, and activation functions achieved superior predictive performance in DELM relative to the alternative models.

Cite This Study

Fayaz et al. (2018) studied this question.

synapsesocial.com/papers/6a1acc424bf744cf5a8555a5 https://doi.org/10.3390/electronics7100222

View Full Paper

Ask AI

Bookmark

Implication

Computational study demonstrates superior hourly energy prediction with deep extreme learning machines in residential buildings, indicating high accuracy over traditional neural network models.

A Prediction Methodology of Energy Consumption Based on Deep Extreme Learning Machine and Comparative Analysis in Residential Buildings

Discussion

Member takes

Implication

Key Points

Cite This Study

A Prediction Methodology of Energy Consumption Based on Deep Extreme Learning Machine and Comparative Analysis in Residential Buildings

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider