Abstract: Big data and artificial intelligence technology have promoted the all-round innovation of engineering cost.In order to improve the level of decision-making of distribution network engineering cost, aiming at the problems of many influencing cost factors and low prediction accuracy in distribution network, this paper proposed a distribution network overhead line engineering cost combination prediction model.Firstly, the missing of important data in the overhead line project of the distribution network was analyzed and processed.Seconelly, the important cost factors of overhead lines of the distribution network were selected based on the random forest algorithm.Finally, the least squares support vector machine model was used for cost prediction based on parameter optimization.By the comparison of the results of different prediction methods, it is confirmed that the cost prediction model constructed is closest to the measured value and can effectively improve the prediction speed and accuracy.The prediction model can provide an effective and practical method for realizing the cost prediction of overhead line project in distribution network.