论文部分内容阅读
构建沥青路面试验检测数据验证方法体系,进而探索出识别和监控虚假数据的计算流程,以防范和遏制虚假数据的蔓延。将经验判断和数学分析相结合,从试验检测数据的逻辑联系和正态性、方差可分解性等基本特征着手,综合异常数据识别、统计检验、误差分析等数据验证方法,提出了沥青路面试验检测数据验证一般流程。工程应用表明,逻辑性检查和异常数据识别可为统计检验提供较好的数据环境,而统计检验需按K-S检验、F检验和t检验3步依次进行,方能保证数据验证的有效性。误差分析作为验证方法的重要补充,有利于促进质量抽检争议的解决,保证试验检测向规范抽样、科学分析的方向发展。
Asphalt pavement test data verification system is constructed, and then the computational process of identifying and monitoring the fake data is explored to prevent and curb the spread of fake data. Based on the combination of experience judgment and mathematical analysis, this paper starts from the basic characteristics of the test data, such as the logical relation and the normality and variance decomposability. Based on the data verification methods such as identification of abnormal data, statistical test and error analysis, the asphalt pavement test Test data to verify the general process. The engineering application shows that the logic check and abnormal data identification can provide a better data environment for statistical test, and the statistical test should be conducted in three steps according to K-S test, F test and t test in order to ensure the validity of data verification. Error analysis as an important complement to the verification method is conducive to the promotion of quality sampling dispute resolution, to ensure that the test to the standard sampling, testing, scientific analysis of the direction of development.