本文在对特高压输电线路工程进行项目划分的基础上，分析识别出影响特高压输电线路工程造价的主要因素，并使用因子分析法对其中18 个主要影响因素进行度量，将得到的5 个因子作为输入、工程单位造价作为输出，构建了基于BP 神经网络的造价预测模型，以国内已建和在建的9 个特高压输变电工程中的75 段输电线路工程数据为样本进行了实证研究，结果证明预测模型对特高压输电线路工程造价进行预测具有可行性和较高的准确性，为特高压工程全生命周期内管理优化提供了一种新的思路和实现方法。

This paper identified and explored research on the main factors which influence UHV transmission line project cost on the basis of the detail ed subdivision and analysis of the UHV transmission line project. And the author also measured the influence factors by factor analysis and established the forecasting model based on BP neural network by measured factors as the input and the project cost as the output. Finally，comparing the output of the model and the multiple linear regression prediction model with the actual cost to demonstrate the accuracy and applicability of the model，this paper researched empirically by using the data of 75 transmission lines of the 9 UHV DC or AC power transmissions and transformation engineering which had been built or under construction at home and proved that the forecasting model was feasible to the forecast research on the project cost. And it also provided a practical method of implementation for the forecast of engineering cost and optimal management and control.