To enhance the drying quality of peony flowers, a recent study developed an integrated intelligent control and monitoring system.
The system incorporates computer vision technology to enable real‐time continuous monitoring and analysis of the total color change (ΔE) and shrinkage rate (SR) of the material. Additionally, by integrating drying time and temperature data, a hybrid neural network model combining convolutional neural networks, long short‐term memory, and attention mechanisms (CNN‐LSTM‐Attention) was employed to accurately predict the moisture ratio (MR) of peony flowers. The predictive model achieved a coefficient of determination (R²) of 0.9962, a mean absolute error (MAE) of 0.6870, and a root mean square error (RMSE) of 0.7634, demonstrating high accuracy in predicting moisture content during the drying process. Furthermore, the system utilized a fuzzy controller to dynamically regulate the drying parameters.
The fuzzy control strategy was used to shorten the drying time by approximately 1 h, improve the drying efficiency by roughly 12%, and significantly maintain the quality of peony flowers. These findings underscore the potential of the system to enhance drying efficiency and product quality.
Wang, Dong & Wang, Yong & Niu, Yao & Zhang, Weipeng & Li, Cunliang & Li, Pei & Zhang, Xuyang & Zhao, Yifan & Yuan, Yuejin. (2025). Development of an intelligent control system of high efficiency and online monitoring for hot air drying of peony flowers. Journal of Food Science. 90. 10.1111/1750-3841.17652.
Source: Research Gate