论文部分内容阅读
针对混凝土结构劣化评价的复杂性,提出了一种基于最小二乘支持向量机(LS_SVM)的混凝土结构劣化评价方法,并结合搜集到的既有混凝土结构劣化评价实例数据,选取其中的一部分作为训练样本对LS_SVM进行训练,利用网格搜索和交叉验证法确定LS_SVM模型的最优参数;另一部分作为检验样本进行测试。实例计算结果表明,该方法在劣化评价中耗时短、结果准确,适用于已有较多评价实例的同类混凝土结构或构件的劣化评价。
Aiming at the complexity of the evaluation of concrete structural deterioration, a method based on least square support vector machine (LS_SVM) is proposed to evaluate the structural deterioration of concrete. Based on the collected data of existing structural deterioration evaluation of concrete structure, some of them are selected as training The sample trained LS_SVM, using grid search and cross validation method to determine the optimal parameters of LS_SVM model; the other part as a test sample for testing. The calculation results of examples show that this method is short in time and accurate in the evaluation of deterioration, and is suitable for the degradation evaluation of similar concrete structures or components that have more evaluation examples.