Partial decomposition approach to generate load curve forecasting scenarios

被引：6

作者：

Carmo, D. L. ^{[1
]}

Souza, R. C. ^{[1
]}

Barbosa, C. R. H. ^{[2
]}

机构：

[1] Pontifical Catholic Univ Rio de Janeiro, Ind Engn Dept, Rio De Janeiro, RJ, Brazil

[2] Pontifical Catholic Univ Rio de Janeiro, Postgrad Program Metrol, Rio De Janeiro, RJ, Brazil

来源：

INTERNATIONAL JOURNAL OF ELECTRICAL POWER & ENERGY SYSTEMS | 2020年 / 115卷

关键词：

Load curve forecasting; Load demand; Hourly data; Bottom-up approach; Partial decomposition approach;

D O I：

10.1016/j.ijepes.2019.105436

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

Detailed information about hourly load has always been considered essential when the subject is meeting the energy demand. Thus, in order to propose a support tool to the Brazilian energy market, this work presents a Scenarios Generation Framework. Such procedure uses bottom-up approach as an annual demand projection provider. The Specific Profile Generation Method is used as a way to overcome the lack of hourly data in Brazil. Not only that, Partial Decomposition Approach, a well-known methodology is used to adapt annual load demand into hourly load curves, closing all existing gaps of other methodologies. Finalizing the Scenarios Generation Framework, Monte Carlo simulation is applied over different obtained results and confidence intervals indicate the possible values of load behavior in the future, thus turning a deterministic forecasting method into a scenarios generation framework. In order to check the results, the Framework is applied and validated using the southeast Brazilian region, developing consistent scenarios which were able to keep relevant historical behavior, at the same time it inset projected behavior changes on the demand.

引用

页数：11

共 31 条

[11] Divide and Conquer? k-Means Clustering of Demand Data Allows Rapid and Accurate Simulations of the British Electricity System [J].

Green, Richard ;

Staffell, Iain ;

Vasilakos, Nicholas .

IEEE TRANSACTIONS ON ENGINEERING MANAGEMENT, 2014, 61 (02) :251-260

[12] Construction of the hourly load curves and detecting the annual peak load of future Syrian electric power demand using bottom-up approach [J].

Hainoun, A. .

INTERNATIONAL JOURNAL OF ELECTRICAL POWER & ENERGY SYSTEMS, 2009, 31 (01) :1-12

[13] Improved Short-Term Load Forecasting Based on Two-Stage Predictions with Artificial Neural Networks in a Microgrid Environment [J].

Hernandez, Luis ;

Baladron, Carlos ;

Aguiar, Javier M. ;

Calavia, Lorena ;

Carro, Belen ;

Sanchez-Esguevillas, Antonio ;

Sanjuan, Javier ;

Gonzalez, Alvaro ;

Lloret, Jaime .

ENERGIES, 2013, 6 (09) :4489-4507

[14] Quantifying uncertainties of neural network-based electricity price forecasts [J].

Khosravi, Abbas ;

Nahavandi, Saeid ;

Creighton, Doug .

APPLIED ENERGY, 2013, 112 :120-129

[15] Comprehensive Review of Neural Network-Based Prediction Intervals and New Advances [J].

Khosravi, Abbas ;

Nahavandi, Saeid ;

Creighton, Doug ;

Atiya, Amir F. .

IEEE TRANSACTIONS ON NEURAL NETWORKS, 2011, 22 (09) :1341-1356

[16]

Koreneff G., 2009, VTT Tied-Valt. Tek. Tutkimusk, P1

[17] Empirical Comparison of Neural Network and Auto-Regressive Models in Short-Term Load Forecasting [J].

Lopez, Miguel ;

Sans, Carlos ;

Valero, Sergio ;

Senabre, Carolina .

ENERGIES, 2018, 11 (08)

[18] Modeling hourly electricity dynamics for policy making in long-term scenarios [J].

Pina, Andre ;

Silva, Carlos ;

Ferrao, Paulo .

ENERGY POLICY, 2011, 39 (09) :4692-4702

[19] Residual load, renewable surplus generation and storage requirements in Germany [J].

Schill, Wolf-Peter .

ENERGY POLICY, 2014, 73 :65-79

[20] Short-Term Load Forecasting with Multi-Source Data Using Gated Recurrent Unit Neural Networks [J].

Wang, Yixing ;

Liu, Meiqin ;

Bao, Zhejing ;

Zhang, Senlin .

ENERGIES, 2018, 11 (05)

← 1 2 3 4 →