Resumen
Water control of grain has always been a crucial link in storage and transportation. The resistance method is considered an effective technique for quickly detecting moisture in grains, making it particularly valuable in practical applications at drying processing sites. In this study, a machine learning method, combining the improved Sparrow Search Algorithm (SSA) and Support Vector Regression (SVR), was adopted for the characteristics of grain resistance. An efficient water content training model was constructed. After comparative validation against three other algorithms, it was found that this model demonstrates superior performance in terms of precision and stability. After a lot of training and taking the average, the correlation coefficient reached 0.987, the coefficient of determination was 0.992, the root mean square error was reduced to 0.64, and the Best accuracy was 0.584. Using the data obtained by the model, the resistance value of grain can be directly measured in the field, and the corresponding moisture value can be found, which can significantly improve the operation efficiency of the grain drying processing site, thereby reducing other interference factors in the detection of grain moisture.