论文部分内容阅读
电力行业在放松管制、引入竞争机制、提高效率的同时,也带来了电价的剧烈波动问题,成为威胁电力市场稳定安全经济运行的一个重要方面。在确定价格钉判别标准的基础上,分析价格钉的主要分布特点。针对价格钉与正常价格分布的不平衡性,为提高价格钉发生预测的准确性,对标准的支持向量机进行修正:一方面运用少数类样本综合仿制技术(SMOTE)增加价格钉这一少数类样本的数量;另一方面对价格钉与正常价格这2类分别设置不同的惩罚系数。方法应用于澳大利亚昆士兰电力市场中价格钉预测,取得了较好的效果。
In the meantime, the power industry has also brought the problem of drastic fluctuations in electricity prices while deregulating and introducing competition mechanisms to improve efficiency. This has become an important aspect threatening the stable, safe and economical operation of the power market. On the basis of determining the criterion of price nails, the main distribution characteristics of price nails are analyzed. In order to improve the accuracy of price nails forecasting, aiming at the imbalance of price nails and normal price distribution, standard support vector machines are used to amend: On the one hand, the use of SMOTE to increase the price of nails The number of samples; the other hand, the price of nails and the normal price of these two categories were set different penalties. The method is applied to the price spike prediction in Queensland electricity market in Australia and has achieved good results.