中文摘要：

提出了一种基于负荷时间序列相空间重构与量子粒子群优化支持向量机的混合超短期负荷预测方法.首先利用G-P算法和C-C算法分别确定超短期负荷数据关联维数和延迟时间,对数据进行相空间重构,并获取预测模型的输入输出数据.接着采用量子粒子群（QPSO）对支持向量机（SVM）进行优化,构建了QPSO-SVM预测模型.最后利用相空间重构获得的模型输入输出数据对QPSO-SVM进行训练获得负荷预测模型.对某电网模拟负荷预测试验结果表明,方法有效提高了负荷预测精度,具有一定的科学意义及工程价值.

英文摘要：

After elaborating the characteristics of power loads, this paper analyzes the chaotic characteristics of load powers. Phasespace is reconstructed after getting correlation dimension and delay-time with G-P and C-C algorithms. In the phase-space, support vector prediction model is established. To improve the prediction accuracy, this paper optimizes SVM parameters with QPSO. SVM prediction results are compared with BP prediction results. It is concluded that the algorithm has a better effect.