基于粒子群优化的高速公路路面性能评价

Performance evaluation of highway pavement based on particle swarm optimization

为提高高速公路路面性能评价的科学性和准确性,以2022年京沪高速济南段实测数据为基础构建路面性能评价指标体系,提出一种基于粒子群优化(particle swarm optimization,PSO)的反向传播(back propagation,BP)神经网络路面性能评价算法——PSO-BP算法.通过与传统BP神经网络对比,验证本算法的准确性和有效性.结果表明,PSO-BP算法在训练集上的预测准确率达到了99.7%,在测试集上的预测准确率达到了99.4%,与传统BP神经网络相比分别提升19.2%和19.1%,说明利用粒子群算法对BP神经网络的初始权值和阈值进行优化,能够提高模型的预测能力以及准确性;PSO-BP算法的预测结果与实际评价高度一致,具有较好的可靠性和稳定性,能够准确地对高速公路沥青路面性能等级进行评价预测.研究成果可为高速公路的养护决策提供重要依据.

In order to improve the scientific and accurate evaluation of expressway pavement performance,we construct a pavement performance evaluation index system based on the measured data from the Jinan section of the Beijing-Shanghai Highway in 2022.A back propagation(BP)neural network pavement performance evaluation algorithm based on particle swarm optimization(PSO),referred to as the PSO-BP algorithm,is proposed.The results show that the PSO-BP algorithm achieves the prediction accuracy of 99.7%on the training set and 99.4%on the test set,which is 19.2%and 19.1%higher than that of the traditional BP neural network,respectively.This indicates that using the particle swarm optimization algorithm to optimize the initial weights and thresholds of the BP neural network can improve the prediction ability and accuracy of the model.The prediction results of the PSO-BP algorithm are highly consistent with the actual evaluation,demonstrating good reliability and stability.The PSO-BP algorithm can accurately evaluate and predict the performance grade of highway asphalt pavement,providing an important basis for highway maintenance decision-making.