基于经验模式分解和最小二乘支持向量机的短期负荷预测

Short-term load forecasting based on empirical mode decomposition and least square support vector machine

电力负荷是具有一定的周期性和随机性的非平稳时间序列,传统的预测方法是建立在负荷是平稳序列的前提下,难以精确的预测。为了进行有效的预测,提高预测精度,提出将经验模式分解EMD(Empirical Mode Decomposition)和最小二乘支持向量机LS-SVM(Least Square Support Vector Machine)相结合对短期负荷进行预测。首先,运用EMD将负荷序列自适应地分解成一系列不同尺度的本征模式分量IMF(intrinsic mode function),分解后的分量突出了原负荷的局部特征,能更明显地看出原负荷序列的周期项、随机项和趋势项;然后,根据各个IMF的变化规律,采用合适的核函数和超参数构造不同的LS-SVM进行预测,最后对各分量的预测值进行相加得到最终的预测值。仿真试验表明,此方法具有较高的精度和较强的推广能力。

The power load is inherently non-stationary time series so that it is difficult to construct the model of accurate forecast. In order to improve forecast precision, a hybrid forecasting method based on Empirical Mode Decomposition （EMD） and Least Square Support Vector Machine （LS-SVM） is presented in this paper. Firstly, the power load series is adaptively decomposed into a series of stationary intrinsic mode functions （IMF） in different scale space. The local features of original load series are prominent in the IMF so that it is more obvious to observe the cycle, random and trend parts of the original load sequence. Secondly, according to the change regulation of each IMF, the right parameter and kernel functions are chosen to build different LS-SVM respectively to forecast each IMF. Finally, these forecasting results of each IMF are combined to obtain final forecasting result. The simulation results show that the hybrid method has faster speed, higher precision and greater generalization ability than that of the single LS-SVM method and that of the BP neural network method, which proves that it is an effective method.