10 kV中压配电网应用电力空间负荷密度特性研究

针对传统方法在研究水电配电网电力空间负荷密度的特性时存在电力负荷密度预测精度较低、用电高峰期出现供电紧张、运行稳定性较弱、安全性较低等问题,提出一种基于DLBAN模型的电力空间负荷密度预测方法。利用跨小波空间算法对水电配电网电力扰动信号进行去噪,获取电力空间滤波信号,根据DLBAN构建电力空间负荷预测模型,对待测区块指定最合适的类标签进行修正。利用DLBAN预测模型完成配电网电力空间负荷密度预测,得到其属性呈正相关性,从而实现对水电配电网电力空间负荷密度特性研究。结果表明,在水电配电网的应用中,城市第二产业的负荷密度的稳定性较强;不同用户的休息时间与负荷密度具有较强的关联性;第三产业的用电时间具有周期性,且负荷密度较大,对整个水电配电网区域负荷密度的贡献度较大,且具有较高的预测精度。

Study on power spatial load density characteristics of 10 kV medium voltage distribution network

Aiming at the problems，such as lower prediction accuracy of power load density，power supply tension in peak power consumption period，weaker operation stability，lower safety，etc．，during studying the characteristics of the power spatial load of hydropower distribution network with the conventional methods，a kind of DLBAN ( double-level Bayesian network augmented naive Bayes) model-based prediction method for power spatial load is proposed herein． With the cross-wavelet spatial algorithm，the power disturbance signal is denoised and then the wave filtering signal is obtained，while the power spatial load prediction model is established in accordance with DLBAN and the most suitable class label assigned to the block to be measured is modified． Furthermore，the prediction of the power spatial load density of the power distribution network is completed with the BLBAN prediction model，from which it is obtained that the attributes are positively correlated，thus the study on the power spatial load density characteristics is realized． The study result shows that in the application of hydropower distribution network，the stability of the power load density from the urban secondary industry is stronger and the break time of different users have stronger correlation with the load density，while the power consumption time of the tertiary industry is periodic with larger load density，thus has higher contribution rate to the regional load density of the whole hydropower distribution network with a higher accuracy of the prediction．