基于支持向量机的省级电网中长期投资规模预测模型研究

Research on Medium and Long-term Investment Scale Prediction Model of Provincial Power Grid Based on SVM

本文提出了基于支持向量机的省级电网中长期投资规模预测模型。围绕新形势下以电量驱动的电网投资规模预测模型参数单一、线型拟合精度较差等问题,本文分析了电网投资新环境、新方向、新要求;探索了电网投资外部驱动因素指标体系,从电力需求因素、电网安全因素、能源转型因素和技术创新因素4个方面,结合灰色关联理论,筛选关键驱动因素,构建省级电网中长期投资规模预测模型;最后通过支持向量机算法,预测了省级电网2019~2022年投资规模,实证结果表明,本文构建的预测模型是有效可行的。

A medium and long-term investment scale prediction model for provincial power grid based on support vector machine is proposed.Focusing on the problem of single-parameter model of power grid-driven grid investment forecasting scale and poor linear fitting accuracy under the new situation,the new environment,new directions and new requirements of grid investment are analyzed firstly.Secondly,the external driving factor index system of grid investment is explored.From the four aspects of power demand factor,grid security factor,energy transformation factor and technological innovation factor,combined with the grey correlation theory,the key driving factors are screened,and the medium and long-term investment scale prediction model of provincial power grid is constructed.Finally,the support vector machine algorithm is used to predict.The investment scale of the provincial power grid in 2019~2022,the empirical results show that the prediction model constructed in this paper is effective and feasible.