改进GM(1,1)自适应滤波组合新陈代谢模型在电力远期价格预测中的应用

Application of Combined Adaptive Filtering and Improved GM(1,1) Metabolism Model for Electricity Forward Price Forecasting

随着电力工业市场化的日益深入,电力远期价格的预测日益重要。电力远期价格受实时电价、利率、负荷需求、社会发展等多种因素影响,变化趋势复杂,无法建立一个准确的数学模型进行描述。本文提出了基于自适应滤波算法和改进灰色 GM(1,1)模型的组合新陈代谢预测算法。用此模型进行的预测能不断将系统新信息引入算法,使预测更接近最新的变化趋势。

In the market of power industry, the forecasting of electricity forward price is very important. The electricity forward price is affected by many factors such as real time electricity price, interest rate, power demands, development degree of the society and so on. It is difficult to set up an accurate math model to describe its overall movement tendencies. A metabolism model combined adaptive filtering and improvedGM（1, 1） modal is presented in this paper. This model can input new information continuously into algorithm to foUow the new tendency to gain higher forecasting effects.