Optimization of support vector machine power load forecasting model based on data mining and Lyapunov exponents

Abstract

According to the chaotic and non-linear characters of power load data, the time series matrix is established with the theory of phase-space reconstruction, and then Lyapunov exponents with chaotic time series are computed to determine the time delay and the embedding dimension. Due to different features of the data, data mining algorithm is conducted to classify the data into different groups. Redundant information is eliminated by the advantage of data mining technology, and the historical loads that have highly similar features with the forecasting day are searched by the system. As a result, the training data can be decreased and the computing speed can also be improved when constructing support vector machine (SVM) model. Then, SVM algorithm is used to predict power load with parameters that get in pretreatment. In order to prove the effectiveness of the new model, the calculation with data mining SVM algorithm is compared with that of single SVM and back propagation network. It can be seen that the new DSVM algorithm effectively improves the forecast accuracy by 0.75%, 1.10% and 1.73% compared with SVM for two random dimensions of 11-dimension, 14-dimension and BP network, respectively. This indicates that the DSVM gains perfect improvement effect in the short-term power load forecasting.

Keywords

power load forecasting / support vector machine (SVM) / Lyapunov exponent / data mining / embedding dimension / feature classification

References

[1] ChristianseW. R.. Short term load forecasting using general exponential smoothing [J]. IEEE Trans Power Apparatus Syst, 1971, 90(2): 900-911

[2] ParkJ. H., ParkY. M., LeeK. Y.. Composite modeling for adaptive short-term load forecasting [J]. IEEE Transactions on Power Systems, 1991, 6(2): 450-457

[3] PaiP.-f., HongW.-chang.. Support vector machines with simulated annealing algorithms in electricity load forecasting [J]. Energy Conversion and Management, 2005, 46: 2669-2688

[4] NiuD.-x., CaoS.-h., LuJ.-c., ZhaoLei.Technology and application of power load forecasting [M], 20092nd ed.Beijing, China Power Press

[5] MbamaluG. A. N., El-hawaryM. E.. Load forecasting via suboptimal seasonal autoregressive models and iteratively reweighted least squares estimation [J]. IEEE Transactions on Power Systems, 1993, 8(1): 343-348

[6] DouglasA. P., BreipohlA. M., LeeF. N., AdapaR.. Impacts of temperature forecast uncertainty on Bayesian load forecasting [J]. IEEE Transactions on Power Systems, 1998, 13(4): 1507-1513

[7] SadownikR., BarbosaE. P.. Short-term forecasting of industrial electricity consumption in Brazil [J]. Journal of Forecasting, 1999, 18(3): 215-224

[8] SuC.-jian.. Study of nonlinear behavior on price and volatility of Chinese stock markets [J]. Mathematics in Practice and Theory, 2006, 36(2): 141-148

[9] KatarzynaB., ArkadiuszO.. Application of bootstrap to detecting chaos in financial time series [J]. Physica A: Statistical Mechanics and its Applications, 2004, 344(2): 317-321

[10] NuJ.-hu.Chaos time series analysis and its application [M], 2002, Wuhan, Wuhan University Press

[11] ChenS.-y., Wangwei.. Chaos forecasting for traffic flow based on Lyapunov exponent[J]. China Civil Engineering Journal, 2004, 37(9): 96-99

[12] DouglasA. P., BreipohlA. M., LeeF. N., AdapaR.. Risk due to load forecast uncertainty in short term power system planning [J]. IEEE Trans Pow Syst, 1998, 13(4): 1493-1499

[13] VapnikV. N.The nature of statistical learning theory [M], 1995, New York, Springer

[14] CortesC., VapnikV.. Support vector networks [J]. Mach Learn, 1995, 20(3): 273-297

[15] HuangW., NakamoriY., WangS. Y.. Forecasting stock market movement direction with support vector machine [J]. Computers and Operations Research, 2005, 32(10): 2513-2522

[16] PaiP.-f., HongW.-chang.. Software reliability forecasting by support vector machines with simulated annealing algorithms [J]. J Syst Software, 2006, 79(6): 747-755

[17] CatalaoJ. P. S., MarianoS. J. P. S., MendesV. M. F., FerreiraL. A. F. M.. Short-term electricity prices forecasting in a competitive market: A neural network approach [J]. Electric Power Systems Research, 2007, 77(10): 1297-1304

[18] WolfA., SwiftJ. B., SwinneyH. L.. Determining Lyapunov exponents from a time series [J]. Physics D, 1985, 116: 285-317

[19] LiangZ.-s., WangL.-m., FuD.-p.. Short-term power load forecasting based on Lyapunov exponents [J]. Proceeding of the CSEE, 1998, 118: 368-472

[20] NiuD.-x., WangY.-l., WuD.-sheng.. Power load forecasting using support vector machine and ant colony optimization [J]. Expert Systems with Applications, 2010, 37(3): 2531-2539

[21] NiuD.-x., WangY.-l., DuanC.-m., XingMian.. A new short-term power load forecasting model based on chaotic time series and SVM [J]. Journal of Universal Computer Science, 2009, 15(13): 2726-2745