基于哈里斯鹰优化遗传规划的钢筋混凝土地下结构硫酸盐腐蚀寿命预测

在硫酸盐腐蚀环境中，地下结构钢筋混凝土易被硫酸根离子侵蚀，从而严重影响其安全服役至结构的设计使用寿命。为预测其在硫酸盐腐蚀环境下的运营服务寿命，提出哈里斯鹰优化遗传规划(HHO-GP)新方法，该方法采用新型全局优化算法--哈里斯鹰优化算法优选遗传规划的主要算法参数。首先，基于收集到的25组工程实例数据，采用新方法建立考虑多因素影响的钢筋混凝土地下结构硫酸盐腐蚀寿命预测模型，该寿命预测模型的综合计算精度较高，计算结果表明其训练误差(5.5%)与预测误差(6.3%)均较小。然后，对该寿命预测模型的4个主要算法参数的敏感性进行分析，结果表明，综合考虑算法的精度、运行效率与模型的完备性，较小的哈里斯鹰种群数量及其最大迭代代数即可满足要求，两者的建议取值可均为15；而遗传规划种群数量与最大迭代代数的变化对模型预测精度影响不大，它们的建议取值可均为50。

Life prediction of RC underground structure by sulfate attack based on harris hawks optimizing genetic programming

In sulfate-aggressive environment, the reinforced concrete (RC) of underground structure is easily corroded by sulfate ion, thus the service life of the structure will be seriously affected. In order to predict the service life of RC underground structure in sulfate-aggressive environment, a new predicting method based on Harris hawks optimizing genetic programming (HHO-GP) is proposed, in which a new global optimization algorithm called Harris hawks optimization is adopted to optimize the main initial parameters of genetic programming. Firstly, based on the collected 25 groups of engineering data, the life prediction model of RC underground structure in sulfate-aggressive environment considering the influence of multiple factors is established. The calculation results by the proposed model show that the average relative training error (5.5%) and average relative predicting error (6.3%) of the model are acceptable. Secondly, the influence of the main initial parameters of the life prediction model on the accuracy and efficiency of the model is analyzed. The results show that, considering the accuracy and efficiency of the algorithm and model completeness, the small population number and maximum iterations of HHO is determined as 15, which can meet the requirements. The population size and maximum iterations of GP have little influence on the prediction accuracy, and their values can be recommended as 50.