论文部分内容阅读
通过室内试验测得不同路面结构下冷再生材料力学特性数据,运用遗传算法(GA)优化灰色神经网络模型进行数据分析和预测,采用灰关联理论进行冷再生材料力学特性影响因素(冷再生层厚度和模量、水泥稳定碎石厚度和模量以及土基模量)的敏感性分析。结果表明,运用遗传算法(GA)优化灰色神经网络组合预测值与试验实测值最大误差仅为6.281%,能有效预测乳化沥青冷再生材料力学特性,可对不同因素下冷再生力学特性进行量化预测分析,可减少试验量。通过灰关联理论敏感性分析得到,水泥稳定碎石模量对乳化沥青力学性能影响较大。
The mechanical properties of cold reclaimed material under different pavement structures were measured by laboratory tests. The gray neural network model was optimized by genetic algorithm (GA) for data analysis and prediction. The gray correlation theory was used to analyze the mechanical properties of cold reclaimed material And modulus, cement stabilized macadam thickness and modulus, and modulus of subgrade). The results show that the maximum error between the combined forecasting value and the experimental value of genetic algorithm (GA) optimized gray neural network is only 6.281%, which can effectively predict the mechanical properties of cold reclaimed asphalt emulsion material, and can quantitatively predict the mechanical characteristics of cold reclamation under different factors Analysis, can reduce the test volume. Through the gray correlation theory sensitivity analysis, the cement stabilized macadam modulus has a great influence on the mechanical properties of emulsified asphalt.