摘要
电力负荷表现为一组非线性时间序列,通过对电力负荷的准确预测,避免电力负荷过载和用电集中拥堵,保障电网稳定可靠运行。传统方法采用Lyapunove指数分岔预测算法,由于Lyapunove指数对电力负荷的初始状态特征的敏感性,导致负荷采样样本较少时预测效果不好。提出一种基于关联维指数分析的电力负荷预测算法,构建了电力负荷时间序列的信号模型,采用级联FIR滤波器实现对电力负荷数据信息流的抗干扰滤波处理,进行信号提纯,然后对电力负荷时域信号模型进行关联维特征提取,采用关联维特征在递归图中的指数分岔性实现对负荷时间序列走势的准确预测,实现电力负荷预测算法改进。仿真实验结果表明,采用该算法进行电力负荷预测具有较好的预测准确性,指向性较好,且具有较好的抗干扰能力,在电力管理和调度中具有较好的应用性。
Power load performance is a set of nonlinear time series, whose accurate prediction can avoid power overload and consumption congestion and ensure the stable and reliable grid operation. The traditional Lyapunove exponent bifurcation prediction algorithm may produce poor forecasting result if the load samples are unsuffiecient because of the sensitivity of the initial state of the Lyapunove index to the initial state of the power load. This paper proposes a power load forecasting algorithm based on correlation dimension index analysis. The signal model of power load time series is constructed, and the FIR filter is used to realize the anti-interference filtering processing of power load data, and then the correlation dimension is extracted. The simulation results show that the proposed method has good accuracy in forecasting power load, good directivity, good anti-disturbance ability, and has good application in electric power management and scheduling.
出处
《电力与能源》
2016年第2期202-206,共5页
Power & Energy
关键词
关联维
电力负荷
预测算法
correlation dimension
power load
forecasting algorithm
