Intelligent control of ginger far-infrared radiation and hot-air drying based on multi-sensor fusion technology

Abstract

In this study, a far infrared radiation and hot-air drying (FIR-HAD) system with multi-sensor detection was developed. Flavor and image information of ginger slices were online monitored with a PEN3 electronic nose and computer vision throughout the whole drying process, respectively. Firstly, ginger slices were dried at four constant temperatures. The response signals of the W1W and W2W sensors (the two most responsive sensors in PEN3 electronic nose system), as well as ΔE (color difference), BI/BI₀ (browning index of dried samples/ browning index of fresh samples), were determined to reflect flavor and color changes based on results from the multi-sensor results. The effects of various constant drying temperatures on the main chemical properties of ginger slices were also examined. The results indicate that the dried samples subjected to FIR-HAD at 60 °C had the highest content of 6-gingerol and 8-gingerol compared to other constant drying temperature. In addition, the antioxidant activity and total flavonoid content were also maintained at satisfactory levels under this condition. Then, a fuzzy logic control (FLC) strategy was developed to continuously adjust the output power for optimizing the FIR-HAD process. To determine fuzzy input parameters, the correlation analysis between the online detection signals and the main chemical properties was conducted based on the moisture ratio of samples. The FLC strategy was developed in two stages based on the results of the correlation analysis, with various control rules applied at each stage. The organoleptic quality and chemical quality of dried samples using FLC drying were significantly improved compared to the constant temperature drying. In addition, the FLC drying strategy achieved the highest gray relational degree value, a metric used to evaluate the overall performance of various drying strategies based on multiple indicators, in the assessment of drying time, color properties, and main qualities. This indicates that combining the FLC strategy with multi-sensor detection technology provides an advantageous optimal treatment for FIR-HAD.