Interpretable Short-Term Electrical Load Forecasting Scheme Using Cubist DOI Creative Commons
Jihoon Moon, Sungwoo Park, Seungmin Rho

et al.

Computational Intelligence and Neuroscience, Journal Year: 2022, Volume and Issue: 2022, P. 1 - 20

Published: Feb. 8, 2022

Daily peak load forecasting (DPLF) and total daily (TDLF) are essential for optimal power system operation from one day to week later. This study develops a Cubist-based incremental learning model perform accurate interpretable DPLF TDLF. To this end, we employ time-series cross-validation effectively reflect recent electrical trends patterns when constructing the model. We also analyze variable importance identify most crucial factors in Cubist In experiments, used two publicly available building datasets three educational cluster datasets. The results showed that proposed yielded averages of 7.77 10.06 mean absolute percentage error coefficient variation root square error, respectively. confirmed temperature holiday information significant external factors, loads ago internal factors.

Language: Английский

An ensemble of differential evolution and Adam for training feed-forward neural networks DOI
Yu Xue,

Yiling Tong,

Ferrante Neri

et al.

Information Sciences, Journal Year: 2022, Volume and Issue: 608, P. 453 - 471

Published: June 16, 2022

Language: Английский

Citations

110
An optimal hybrid multiclass SVM for plant leaf disease detection using spatial Fuzzy C-Means model DOI
Santosh Kumar Sahu, Manish Pandey

Expert Systems with Applications, Journal Year: 2022, Volume and Issue: 214, P. 118989 - 118989

Published: Oct. 17, 2022

Language: Английский

Citations

92
Influence of PV/T waste heat on water productivity and electricity generation of solar stills using heat pipes and thermoelectric generator: An experimental study and environmental analysis DOI
Shahin Shoeibi,

Mohammad Saemian,

Mehdi Khiadani

et al.

Energy Conversion and Management, Journal Year: 2022, Volume and Issue: 276, P. 116504 - 116504

Published: Nov. 30, 2022

Language: Английский

Citations

91
Proposing hybrid prediction approaches with the integration of machine learning models and metaheuristic algorithms to forecast the cooling and heating load of buildings DOI
He Dasi, Ying Zhang,

MD Faisal Bin Ashab

et al.

Energy, Journal Year: 2024, Volume and Issue: 291, P. 130297 - 130297

Published: Jan. 15, 2024

Language: Английский

Citations

24
A combined system based on data preprocessing and optimization algorithm for electricity load forecasting DOI

Lei Gu,

Jianzhou Wang, Jingjiang Liu

et al.

Computers & Industrial Engineering, Journal Year: 2024, Volume and Issue: 191, P. 110114 - 110114

Published: April 5, 2024

Language: Английский

Citations

16
MFformer: An improved transformer-based multi-frequency feature aggregation model for electricity load forecasting DOI

Hing-Yip Tong,

Jun Liu

Electric Power Systems Research, Journal Year: 2025, Volume and Issue: 243, P. 111492 - 111492

Published: Feb. 8, 2025

Language: Английский

Citations

2
Review of load forecasting based on artificial intelligence methodologies, models, and challenges DOI
Hui Hou, Chao Liu, Qing Wang

et al.

Electric Power Systems Research, Journal Year: 2022, Volume and Issue: 210, P. 108067 - 108067

Published: May 11, 2022

Language: Английский

Citations

56
Tree-Based Machine Learning Models with Optuna in Predicting Impedance Values for Circuit Analysis DOI Creative Commons
Jung-Pin Lai, Ying-Lei Lin,

Ho-Chuan Lin

et al.

Micromachines, Journal Year: 2023, Volume and Issue: 14(2), P. 265 - 265

Published: Jan. 20, 2023

The transmission characteristics of the printed circuit board (PCB) ensure signal integrity and support entire system, with impedance matching being critical in design high-speed PCB circuits. Because factors affecting are closely related to production process, designers manufacturers must work together adjust target maintain integrity. Five machine learning models, including decision tree (DT), random forest (RF), extreme gradient boosting (XGBoost), categorical (CatBoost), light (LightGBM), were used forecast values. Furthermore, Optuna algorithm is determine forecasting model hyperparameters. This study applied tree-based techniques predict impedance. results revealed that five models can generate satisfying accuracy terms three measurements, mean absolute percentage error (MAPE), root square (RMSE), coefficient determination (R2). Meanwhile, LightGBM outperformed other models. In addition, by using tune parameters be increased. Thus, this suggest a viable promising alternative for predicting values analysis.

Language: Английский

Citations

31
Current status, challenges, and prospects of data-driven urban energy modeling: A review of machine learning methods DOI Creative Commons
Prajowal Manandhar, Hasan Rafiq, Edwin Rodríguez-Ubiñas

et al.

Energy Reports, Journal Year: 2023, Volume and Issue: 9, P. 2757 - 2776

Published: Feb. 2, 2023

Urban energy modeling is essential in planning electricity generation and efficiently managing electric power systems. Various urban models were developed for several energy-driven applications, including emission reduction, retrofit analysis, forecasting. Electricity load forecasts help to estimate the demand effectively aid system operation balancing. The accuracy of at high temporal spatial resolution can impact operation. Therefore, it know factors that affect these how they be improved regarding current state art. This article reviews recent literature on data-driven three steps. First, different phases review process are explained select analyze machine learning-based short-term forecasts. Then various aspects forecasting techniques have been reviewed, addressing their advantages, disadvantages, resolution, performance. Finally, covers challenges describes reasons performance degradation lower accuracy. Based reviewed literature, was found temperature, user profiles, proper management input data highly forecast In addition, shortcomings existing evaluation metrics make applicability those questionable. we conclude by highlighting necessary actions improve relatively unexplored used as a reference future research accurate

Language: Английский

Citations

30
Predicting hourly heating load in residential buildings using a hybrid SSA–CNN–SVM approach DOI Creative Commons

Wenhan An,

Bo Gao, Jianhua Liu

et al.

Case Studies in Thermal Engineering, Journal Year: 2024, Volume and Issue: 59, P. 104516 - 104516

Published: May 8, 2024

This study proposes a hybrid prediction model using sparrow search algorithm (SSA) to optimize the convolutional neural network (CNN) and support vector machine (SVM), in order perform accurate of secondary supply temperature (Ts2). The historical operation data Weifang residential building thermal station was adopted reasonable preprocessing performed suppress interference abnormal data. input variables were screened correlation analysis method, taking influence hysteresis effect into consideration. SSA-CNN-SVM then developed for prediction. performance evaluated by root mean square error, absolute percentage error (MAPE), value relative each time step. results obtained demonstrated that has high accuracy. MAPE values two heat exchange stations between 2.28% 2.4%. indoor significantly affected accuracy Ts2. After introduction temperature, predicted reduced 0.35%. maximum reduction 1.5% compared with other models.

Language: Английский

Citations

12