Related Experiment Video
Updated: Sep 27, 2026

Construction of Models for Nondestructive Prediction of Ingredient Contents in Blueberries by Near-infrared Spectroscopy Based on HPLC Measurements
Published on: June 28, 2016
Optimization of Sugar-Substituted Strawberry Preserve Formulation via Back-Propagation Neural Network
Qing Gao1, Nanxin Wang1, Yutian Wang1
1College of Food Science and Technology, Yunnan Agricultural University, Kunming 650201, China.
Abstract:
Single-factor experiments were conducted with sensory score set as the evaluation index to investigate individual effects of sugar substitution dosage, mass ratio of xylitol (Xyl) to isomaltooligosaccharide (IMO), and osmotic soaking duration on the comprehensive quality of low-sugar strawberry preserves. Box-Behnken design (BBD)-based response surface methodology (RSM) and back-propagation neural network (BPNN) models were subsequently established for parallel process optimization and comparative modeling analysis based on single-factor experimental datasets. Physicochemical indicators including total sugar, titratable acidity, total flavonoids, total phenolics, chromatic aberration, moisture content and water activity of low-sugar strawberry preserves manufactured under optimized technological parameters were determined and systematically compared with those of sucrose-controlled counterparts. The RSM and BPNN models predicted maximum sensory scores of 81.62 and 82.62, respectively. Experimental verification conducted under the practically adjusted optimal conditions derived from the RSM and BPNN models yielded sensory scores of 81.63 and 82.72, respectively, with the BPNN-derived condition achieving a higher verified sensory score than the RSM-derived condition. Based on these results, the optimal processing parameters were determined as a 40% (w/w) addition amount of compound sugar substitute, a Xyl/IMO mass ratio of 1.5, and a soaking time of 5.5 h. Lower total sugar and titratable acidity, yet higher contents of total phenolics and flavonoids, were detected in sugar-substituted strawberry preserves in comparison with sucrose-preserved samples. Meanwhile, stronger red-yellow chromaticity was observed in sucrose-based preserves, while sugar-substituted products delivered higher lightness. Superior predictive capacity of the BPNN model to RSM was validated in the formulation optimization of strawberry preserves. Our experimental findings show that partially replacing sucrose with functional sweeteners improves the nutritional attributes of fruit preserves without impairing their sensory acceptability.
