With the improvement of living standards and the evolution of dietary structure and disease spectrum, the nutritional and health functions of foods such as physiological regulation and disease prevention have attracted much attention. As an important dietary source for human beings, the digestive properties and nutritional functions of rice starch, which is the main component of rice, are crucial to human dietary nutrition and organism health. Moreover, people are increasingly demanding for the taste of rice, in which the hardness and viscosity quality of rice are the most easily recognized among many tastes. Starch is the main component of rice. Rice starch has the characteristics of easy digestibility, small particle size, and easy acceptance by consumers. These characteristics have led to the widespread use of rice starch in food and pharmaceutical applications. The properties of starch depend on the size distribution of starch granules, the morphology and molecular structure of starch granules, and other structural properties. Vegetable oils contain substances that break down fats. Moderate intake is beneficial, but this does not mean that it is low in calories. To understand the effect of fatty acids on the thermal properties and textural quality of rice starch, rice starch lipid complexes were made from rice by compounding stearic and oleic acids with rice. Then the effect of different fatty acids on the thermal properties and textural quality of rice starch was investigated and the system was implemented using a group optimization algorithm and a neural network, mainly consisting of a convolutional layer, a fully connected layer and a classifier, where the main structures of the convolutional layer are forward and backward propagation. This paper elucidates the interplay between fatty acids and rice starch heat at the molecular level using deep learning to provide theoretical basis and fundamental data for the creation of cognitively healthy starch-based foods.