Rejump probability prediction optimization method and device for power distribution network line based on gradient descent

The invention relates to the technical field of power grid equipment manufacturing, in particular to a rejump probability prediction optimization method and device for a power distribution network line based on gradient descent. The method comprises the following steps: constructing a rejump probability prediction model based on power distribution network data; generating a target function throughfunction mapping; based on the target function, calculating by a rejump probability prediction model to obtain a prediction value; constructing a loss function through the absolute error of the predicted value and the actual value, and assigning hyper-parameters; obtaining the minimum loss value of the loss function through a stochastic gradient descent method; and determining model parameters based on the minimum loss value to obtain an optimal rejump probability prediction model. 本申请涉及电网设备制造技术领域，特别地，涉及一种基于梯度下降的配电网线路重跳概率预测优化方法和装置。所述方法包括：基于配电网数据构建重跳概率预测模型；通过函数映射生成目标函数；基于所述目标函数、重跳概率预测模型计算得到预测值；通过预测值和实际值的绝对误差构造损失函数，赋值超