机构地区: 华南理工大学
出 处: 《电力建设》 2004年第3期4-6,共3页
摘 要: 电力负荷预报是电力系统的重要工作之一 ,电力系统年负荷为非线性时间序列 ,文章根据时间序列的理论与方法 ,在分析电力系统历史负荷非线性时间序列运动轨迹的基础上 ,得出了年电力负荷数据由时间序列的典型分解式及选择期望值为噪声项的概率模型 ,并采用最小二乘法拟合趋势项 ,再根据参数估计和数据拟合优度的检验结果对时间序列未来取值进行预报。对电力负荷时间序列R/S分形分析进一步表明了预报值的合理性 ,也预示了今后 2 0~3 0年内电力负荷会持续。 The power load prediction is one of important works in power system a nd the annual load of the power system is a nonlinear time sequence. According t o the time sequence theory and method and analysis on moving tracking of nonline ar time sequence for history load of power system, the paper has obtained that t he typical decomposition formula represents the annual power load data, expectin g selected value is the probabilistic model of the noise item. The minimum two-m ultiplication method is used to draft tendency item to predict the annual power load. The analysis on the time sequence of power load R/S shapes indicates furth er the rationality of the predicted values and shows that the power load in foll owing 20~30 years will continuous grow up.
关 键 词: 电力系统 电力负荷 非线性预报 概率模型 数学模型
领 域: [电气工程]