基于台区10 kV配线负荷的组合预测方法及应用

Combination Prediction Method and Application Based on Substation Area 10 kV Distribution Load

线路重过载问题不仅影响供电设备的使用寿命,而且极易引发一系列的安全事故。为实现10 kV线路重过载问题有效预警的目标,从用电信息采集台区入手,搭建基于遗传神经网络和自适应权重的10 kV配线负荷组合预测模型,进而结合线变关系,实现配线负荷预测,提前一天掌握各10 kV配线负荷情况,针对存在重过载风险的配线线路及时采取调整用电方案或负荷转供等手段,可有效缓解线路重过载压力,确保配电网安全、稳定运行。实验表明,所提模型具有较高的预测精度和应用价值。

The problem of heavy line overload not only affects the service life of power supply equipment,but also easily leads to a series of safety accidents.In order to achieve the goal of effective early warning of the problem of heavy overload of 10 kV lines,starting from the power consumption information collection substation area,the combination prediction model of 10 kV distribution load based on genetic neural network and adaptive weight is built.Then combinating with the line change relationship,the distribution load is predicted and the load situation of each 10 kV distribution line is grasped one day in advance.Taking the measures such as adjusting the power consumption scheme or load transfer in time for the distribution line with heavy overload risk,the heavy overload pressure of the line can be effectively relieved and the safe and stable operation of the distribution network can be ensured.Experiments show that the proposed model has high prediction accuracy and application value.