基于SGDM优化IWOA-CNN的配电网工程造价控制研究

Research on cost control of distribution network engineeringbased on SGDM optimization IWOA-CNN

为了控制配电网工程项目的成本,需准确预测配电网工程造价,本文提出一种基于带动量因子的随机梯度下降(stochastic gradient descent with momentum factor, SGDM)优化的改进鲸鱼算法-卷积神经网络工程造价预测模型。首先,考虑回路数、杆塔数、导线、地形、地质、风速、覆冰、导线截面、混凝土杆、塔材、绝缘子(直线)、绝缘子(耐张)、基坑开方、基础钢材、底盘和水泥对配电网工程造价的影响,建立了非线性函数关系;采用SGDM优化器改进的卷积神经网络对函数进行逼近,并用贝叶斯方法优化卷积神经网络的超参数;利用改进的鲸鱼算法(improved whale optimization algorithm, IWOA)优化卷积神经网络,找出卷积神经网络的最优学习率。数值算例表明,新模型预测效果较好,并提出相应的控制策略。

In order to control the distribution network engineering cost,it is necessary to accurately predict the distribution network engineering cost.Therefore,an improved whale algorithm convolutional neural network engineering cost prediction model based on stochastic gradient descent with momentum factor,SGDM optimization is proposed.First,considering the influence of the number of circuits,towers,conductors,terrain,geology,wind speed,icing,conductor cross-sections,concrete poles,tower materials,insulators(straight lines),insulators(tensile),foundation pits,foundation steel,chassis and cement on the construction cost of the distribution network is established,and the nonlinear function relationship is established.Then,the function is approximated by the convolutional neural network improved by SGDM optimizer,and the hyperparameters of the convolutional neural network are optimized by Bayesian method.Finally,the improved whale optimization algorithm(IWOA)is used to optimize the convolutional neural network,and the optimal learning rate of the convolutional neural network is found.The numerical example shows that the new model works well,and put forward the corresponding control strategy.