论文部分内容阅读
电力市场中的电价序列存在很大的随机波动和价格尖峰。文章提出根据电价序列的变化特点,通过小波变换将其分解为概貌序列和细节序列,从而在不同尺度上反映电价的变化规律。通过概貌分量找出电价的主要波动规律,并由此对电价进行预测,剔除细节分量所反映的电价的随机波动影响。建立考虑异方差的广义自回归条件异方差模型(generalized autoregressive conditional heteroscedasticity,GARCH)对概貌序列建模,并在GARCH模型中加入外生变量形成GARCHX模型,以弥补传统时间序列模型忽略外界影响的缺陷。对美国PJM电力市场的实例研究表明,所建立的W-GARCHX模型比传统时间序列模型的预测精度有明显提高。
There is a great random fluctuation and price spike in the electricity price series in the electricity market. According to the characteristics of the change of electricity price series, the article puts forward that it can be decomposed into generalized sequence and detail sequence by wavelet transform, so as to reflect the changing law of electricity price on different scales. The main fluctuation law of electricity price is found through the profile component, and then the electricity price is predicted, excluding the random fluctuations of the electricity price reflected by the detail component. A generalized autoregressive conditional heteroscedasticity (GARCH) model with heteroscedasticity was established to model the profile sequence and a GARCHX model was added to the GARCH model to make up for the defects that the traditional time series model ignores the external influence . The case study of the PJM power market in the United States shows that the W-GARCHX model has a significantly higher prediction accuracy than the traditional time series model.