Power grid project cost prediction based on improved CNN under low carbon background

Abstract

Abstract With the increasing awareness of environmental protection and the restriction of traditional energy, green energy has become the main trend in the development of power engineering. To solve the problems of the traditional power engineering cost prediction model, such as complexity, imprecision, and dependence on manual experience, this paper proposes an improved convolutional neural network (CNN)-based power transmission and transformation project cost prediction model. Firstly, the cost prediction model of power transmission and transformation project based on CNN is constructed. Then, the historical data is entered into the network model to extract data features. Finally, in order to solve the inconsistency between the expected output and the actual output, the stochastic gradient descent with momentum (SGDM) optimizer is used to optimize the weight parameters of CNN and complete the model training. This model combines the advantages of the SGDM optimizer and CNN, and it has higher prediction accuracy and stability than Levenberg-Marquardt (LM)-CNN and gradient descent (GD)-CNN models. The experimental results show that the proposed model is superior to the traditional methods in terms of prediction accuracy and efficiency, and it provides an effective reference for the reasonable budget and resource allocation of power engineering.