Abstract: In order to meet the demand of detecting moisture content of aggregates in industrial production, an analysis model was constructed based on microwave transmission and multiple linear regression model theory, its significance and the reliability of its detection results were studied. The comparison deviation and its root mean square error, standard deviation and coefficient of variation evaluation indexes were established to represent the repeatability and accuracy of the off-line and on-line measurement results of the water content of aggregates with different particle size ranges of 0~4. 75 mm, 4. 75~16. 00 mm and 16. 00~31. 50 mm. The results showed that the multiple linear regression analysis model had a high goodness of fit, the deviation range of offline and online detection data was ±0. 3%, and the root mean square error was 0. 148%, 0. 13%, 0. 147%. The test results were highly consistent and meet the requirements of engineering detection.