Support Vector Machines with Simulated Annealing Algorithms in Electricity Load Forecasting

By Ping-Feng Pai, Wei‐Chiang Hong. 2005

388 citations recorded by the source corpus.

Subfields: Electrical and Electronic Engineering, Environmental Engineering, Management Science and Operations Research

View Paper: https://www.semanticscholar.org/paper/cfb701ba64f45b5b831ec166b677d22270cc0e01

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Electrical and Electronic Engineering

1. Junyi Zhou
2. Jing Shi
3. Gong Li
4. Kwok-Wing Chau
5. Lin Qiu
6. Ravi Sankar
7. K. Rajurkar
8. James W. Taylor

In Environmental Engineering

1. Kwok-Wing Chau
2. Lin Qiu
3. Wenchao Wang
4. Chun-tian Cheng
5. Jing Shi
6. Gong Li
7. James W. Taylor
8. Yi Zeng

In Management Science and Operations Research

1. Kwok-Wing Chau
2. Lin Qiu
3. Ravi Sankar
4. Jing Shi
5. Gong Li
6. James W. Taylor
7. Yi Zeng
8. Wenchao Wang

Citations in the corpus

• Fine tuning support vector machines for short-term wind speed forecasting (2011, background): “In principle, the ranges of the SVM parameters are chosen to be wide so that the common values and ranges reported in literature are well covered [16,19,25,29,30].”
• Back propagation neural network with adaptive differential evolution algorithm for time series forecasting (2015, methodology): “…signed-rank test is one of the most commonly adopted tests for evaluating the predictive capabilities of two different models and for determining whether statistically significant differences exist among the results (Ju & Hong, 2013; Lu, Lee, & Chiu, 2009; Pai & Hong, 2005; Zhang et al., 2012).”
• A comparison of performance of several artificial intelligence methods for forecasting monthly discharge time series (2009, methodology): “Recently, there are several methods developed to identify the parameters, such as the simulated annealing algorithms (Pai and Hong, 2005), GA (Pai, 2006) and the shuffled complex evolution algorithm (SCE-UA) (Lin et al., 2006; Yu et al., 2004).”
• Short-Term Load Forecasting With Exponentially Weighted Methods (2012, background): “casting is to decompose the load into weather sensitive and weather insensitive components [3], and for the latter univariate”
• Triple seasonal methods for short-term electricity demand forecasting (2010, methodology): “Pai and Hong (2005) implement an ARIMA model and a neural network as benchmarks in their study of support vector machines, which also enable nonlinear regression modelling.”
• Using support vector machines for long-term discharge prediction (2006, methodology): “More recently, several other methods have been developed to identify the parameters, such as the stepwise search (Dong et al., 2005), genetic algorithms (Chen et al., 2004), the shuffled complex evolution algorithm (Yu et al., 2004), and the simulated annealing algorithms (Pai & Hong, 2005).”
• Hourly cooling load forecasting using time-indexed ARX models with two-stage weighted least squares regression (2014, background): “Artificial neural networks (ANN) and support vector machines (SVM) are also common models for cooling load prediction [15] [16] [17] [18] [19] [20] [21] [22] [23] [24].”
• Time Series Prediction Using Support Vector Machines: A Survey (2009, background): “The key to this approach was the use of both a genetic algorithm and the use of feedback (recurrent network architecture) to aid in the selection of the free parameters of the SVR. Hong and Pai [77] compared the SVR to three other models (Duane, ARIMA, and GRNN) for the prediction of engine failure.”
• Parameter determination of support vector machine and feature selection using simulated annealing approach (2008, methodology): “Pai and Hong [24,25] also presented an SA approach to obtain parameter values for SVM and applied it to real data.”