微型电脑应用

随着电网系统的智能技术的复杂化，电力系统中能量也具有越来越复杂的工作形式。针对智能电网系统协调管理的问题，提出了一种结合时间序列的方法。方法采用了自回归模型来捕捉电力负荷数据中与时间相关联的变化规律，并通过构建最小二乘法的目标函数来优化预测结果。此外，为了提高电力数据的质量，引入变分模态分解技术，分离并剔除数据中的噪声干扰。变分模态分解通过非递归的形式对数据进行分解，识别不同模态下的噪声分量，并通过样本熵的计算来确定最优的分解个数。实验结果表明，在预测数据相对误差分析中，研究方法的最大相对误差仅为0.12%。说明所提方法能够有效智能电网系统短期负荷预测完成，为协调管理提供支持，在实际应用中可以为智能电网管理方提供更高质量的参照数据。

With the complexity of intelligent technology in power system, energy in power system has more and more complicated working forms. Aiming at the issue of coordinated management in smart grid systems, a method combining time series is proposed. The method adopts an autoregressive model to capture the time-related variation patterns in electricity load data and optimizes the prediction results by constructing the objective function of the least square method. In addition, in order to improve the quality of electricity data, variational mode decomposition technology is introduced to separate and eliminate noise interference in the data. Variational mode decomposition decomposes data in a non-recursive form, identifies noise components in different modes, and determines the optimal number of decompositions through the calculation of sample entropy. The experimental results show that the maximum relative error of the research method in the analysis of relative error of predicted data is only 0.12%. It is shows that the proposed method can effectively predict the short-term load of smart grid systems, provide support for coordinated management, and provide higher quality reference data for smart grid management in practical applications.