Probabilistic Forecasting of Hourly Electricity Price by Generalization of ELM for Usage in Improved Wavelet Neural Network

Rafiei, Mehdi; Niknam, Taher; Khooban, Mohammad‐Hassan

doi:10.1109/tii.2016.2585378

Cited by 118 publications

(40 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Studies where univariate statistical time series models are used include Nogales et al (2002), Contreras et al (2003), Conejo et al (2005), Zareipour et al (2006), Paraschiv et al (2015) and Ziel et al (2015a), while papers where neural networks are put to work include Rodriguez and Anders (2004), Amjady (2006), Pao (2007), Amjady et al (2010), Abedinia et al (2015), Kim (2015), Dudek (2016), Keles et al (2016) and Rafiei et al (2017), among others.…”

Section: Introductionmentioning

confidence: 99%

Day-ahead electricity price forecasting with high-dimensional structures: Univariate vs. multivariate modeling frameworks

2018

View full text Add to dashboard Cite

We conduct an extensive empirical study on short-term electricity price forecasting (EPF) to address the long-standing question if the optimal model structure for EPF is univariate or multivariate. We provide evidence that despite a minor edge in predictive performance overall, the multivariate modeling framework does not uniformly outperform the univariate one across all 12 considered datasets, seasons of the year or hours of the day, and at times is outperformed by the latter. This is an indication that combining advanced structures or the corresponding forecasts from both modeling approaches can bring a further improvement in forecasting accuracy. We show that this indeed can be the case, even for a simple averaging scheme involving only two models. Finally, we also analyze variable selection for the best performing high-dimensional lasso-type models, thus provide guidelines to structuring better performing forecasting model designs.or a univariate framework, the latter generally perform better for the first half of the day, whereas the former are better in the second half of the day. However, there has been no through, empirical study to date, involving many fine-tuned specifications from both groups. With this paper we want to fill the gap and provide much needed evidence. In particular we want to address three pertinent questions:1. Which modeling frameworkmultivariate or univariate -is better for EPF? 2. If one of them is better, is it better across all hours, seasons of the year and markets? 3. How many and which past values of the spot price process should be used in EPF models?The remainder of the paper is structured as follows. In Section 2 we thoroughly discuss the univariate and multivariate modeling frameworks, which are driven by different data-format perspectives. This is a crucial, conceptual part of the paper, which sets ground for the empirical analysis in the following Sections. In Section 3 we briefly describe the 12 price series used and present the area hyperbolic sine transform for stabilizing the variance of spot price data. In Section 4 we define 10 forecasting models representing eight model classes: (C1) the mean values of the past prices, (C2) similar-day techniques, (C3) sets of 24 parsimonious, interrelated autoregressive (AR) structures (so-called expert models), (C4) sets of 24 univariate AR models, (C5) vector autoregressive (VAR) models, (C6) sets of 24 parameter-rich, interrelated AR models estimated using the least absolute shrinkage and selection operator (i.e., lasso or LASSO; which shrinks to zero the coefficients of redundant explanatory variables), (C7) univariate AR models and (C8) univariate, parameter-rich AR models estimated using the lasso. In Section 5 we evaluate their performance on the basis of the Mean Absolute Error (MAE), the mean percentage deviation from the best (m.p.d.f.b.) model and using two variants of the Diebold and Mariano (1995) test for significant differences in the forecasting performance. We also discuss variable selection for the best performin...

show abstract

Section: Introductionmentioning

confidence: 99%

Day-ahead electricity price forecasting with high-dimensional structures: Univariate vs. multivariate modeling frameworks

2018

View full text Add to dashboard Cite

show abstract

“…In this formulation, the prediction value of the PIs is considered as a misleading result, especially when the deviation is large, so the width can be closer to that of the primary without any negative effects of the value out of PIs. Considering that the wind power data is nonlinear, highly dimensional, and strongly coupled, the Wavelet Neural Network (WNN) is a more suitable and flexible neural network that can deal with the data of wind power efficiently, because it can instigate a superior system model for complex and seismic applications in comparison to the NN with a sigmoidal activation function [18,19], so it is used to construct the PI model.…”

Section: Introductionmentioning

confidence: 99%

Wind Power Forecasting Using Multi-Objective Evolutionary Algorithms for Wavelet Neural Network-Optimized Prediction Intervals

2018

View full text Add to dashboard Cite

Abstract:The intermittency of renewable energy will increase the uncertainty of the power system, so it is necessary to predict the short-term wind power, after which the electrical power system can operate reliably and safely. Unlike the traditional point forecasting, the purpose of this study is to quantify the potential uncertainties of wind power and to construct prediction intervals (PIs) and prediction models using wavelet neural network (WNN). Lower upper bound estimation (LUBE) of the PIs is achieved by minimizing a multi-objective function covering both interval width and coverage probabilities. Considering the influence of the points out of the PIs to shorten the width of PIs without compromising coverage probability, a new, improved, multi-objective artificial bee colony (MOABC) algorithm combining multi-objective evolutionary knowledge, called EKMOABC, is proposed for the optimization of the forecasting model. In this paper, some comparative simulations are carried out and the results show that the proposed model and algorithm can achieve higher quality PIs for wind power forecasting. Taking into account the intermittency of renewable energy, such a type of wind power forecast can actually provide a more reliable reference for dispatching of the power system.

show abstract

“…Table 2 shows the values of the model parameters. The data in the table are based on the general standard of [40].…”

Section: Basic Datamentioning

confidence: 99%

An Optimization Model and Modified Harmony Search Algorithm for Microgrid Planning with ESS

Jiao

Tan

et al. 2017

Discrete Dynamics in Nature and Society

View full text Add to dashboard Cite

To solve problems such as the high cost of microgrids (MGs), balance between supply and demand, stability of system operation, and optimizing the MG planning model, the energy storage system (ESS) and harmony search algorithm (HSA) are proposed. First, the conventional MG planning optimization model is constructed and the constraint conditions are defined: the supply and demand balance and reserve requirements. Second, an ESS is integrated into the optimal model of MG planning. The model with an ESS can solve and identify parameters such as the optimal power, optimal capacity, and optimal installation year. Third, the convergence speed and robustness of the ESS are optimized and improved. A case study comprising three different cases concludes the paper. The results show that the modified HSA (MHSA) can effectively improve the stability and economy of MG operation with an ESS.

show abstract

Probabilistic Forecasting of Hourly Electricity Price by Generalization of ELM for Usage in Improved Wavelet Neural Network

Cited by 118 publications

References 18 publications

Day-ahead electricity price forecasting with high-dimensional structures: Univariate vs. multivariate modeling frameworks

Day-ahead electricity price forecasting with high-dimensional structures: Univariate vs. multivariate modeling frameworks

Wind Power Forecasting Using Multi-Objective Evolutionary Algorithms for Wavelet Neural Network-Optimized Prediction Intervals

An Optimization Model and Modified Harmony Search Algorithm for Microgrid Planning with ESS

Contact Info

Product

Resources

About