针对日趋严重的电网谐波污染亟需大量谐波数据支撑分析和治理及电网谐波监测能力不足的问题,提出一种改进减法平均优化(subtraction average based optimizer,SABO)算法优化反向传播(back-propagation,BP)神经网络实现谐波预测,以缓解当前谐波数据匮乏的问题.为了克服现有SABO算法易于陷入局部最优解,初始化时使用Logistic混沌映射替代随机数,同时迭代搜索中利用黄金正弦优化算法辅助SABO跳出局部最优,从而提高BP神经网络预测准确率.最后,以某省实际运行数据验证所提改进SABAO-BP模型在谐波电压畸变率及单次谐波电压含有率预测中均具有较高准确性.
Harmonic Prediction of Power Grid Based on Improved SABO-BP Algorithm
Aiming at the increasingly serious harmonic pollution of the power grid,which urgently needs a large amount of harmonic data to support the analysis and management of the lack of harmonic monitoring ability of power grid,this paper proposes an improved subtraction average based optimizer(SABO)algorithm to optimize the back propagation(BP)neural network to realize harmonic prediction,so as to alleviate the current lack of harmonic data.In order to overcome the problem that the existing SABO algorithm is easy to fall into the local optimal solution,the Logistic chaotic map is used to replace the random number during initialization.At the same time,the golden sine optimization algorithm is used to assist SABO to jump out of the local optimum in the iterative search,thus improving the prediction accuracy of BP neural network.Finally,the actual operation data of one province is used to verify that the proposed improved SABAO-BP model has high accuracy in the prediction of harmonic voltage distortion rate and single harmonic voltage content rate.
power qualityharmonic predictionimproved BP neural networksubtraction-average-based optimizer
万方数据
维普
