Abstract: The measurement scale of aircraft structural strength tests is huge. During the tests, there are often invalid strain data caused by reasons other than the test samples themselves. During test data analysis, it is required to timely screen and eliminate the invalid data to provide a complete and effective measurement database for subsequent test analysis, thereby enhancing the measurement efficiency. However, the existing data analysis methods mainly rely on manual observation and screening based on experience, which is inefficient and prone to omission. This paper establishes a database for aircraft structural strength testing, proposes a data analysis algorithm process based on statistical learning methods, including support vector machines and artificial neural networks, and conducts comparative validation tests on the data analysis algorithms. The results show that the data analysis algorithm based on statistical learning can obtain relatively accurate analysis results for data analysis tasks, and can better complete the preliminary screening of measurement data. This algorithm can effectively improve the accuracy and efficiency of measurement data analysis, providing a theoretical basis for the development of automated processing software for massive strain data in the future.