基于神经网络算法的10 kV配电变压器容载比

Capacity-load Ratio of 10 kV Power Distribution Transformer Based on Neural Network Algorithm

根据相关标准、导则及10 kV配电网的特点,分析配电变压器容载比各参数并进行适当修正。利用反向传播(backpropagation,BP)神经网络算法得到负荷同时率与主要影响因素的映射关系,得到最优负荷分散系数,并结合具体地区实例展示10 kV配电变压器容载比的确定方法。该方法充分考虑了地区负荷特点,避免以往盲目按照导则或标准全省、市一刀切执行的情况。

According to relevant standards, guide rules and characteristics of 10 kV power distribanalyzes parameters of capacity-load ratio of the power distribution transformer and makes suitable amendment. By usingback propagation （BP） neural network algorithm, it obtains mapping relation between load coincidence factor and major in-fluencing factors and the optimal load dispersion coefficien. Combining with regional actual tion method for capacity-load ratio of the 10 kV power distribution transformer which fully considers characteristics of re-gional load and avoids blind execution in the whole province and cities according to guide rulcs or standards.