Optimization of High-Pressure Ultrasound-Assisted Enzymatic Extraction of Collagen from Sheep Skin Using Genetic Algorithm-Back Propagation Neural Network
1. Tianjin Key Laboratory of Food Biotechnology, School of Biotechnology and Food Science, Tianjin University of Commerce, Tianjin 300134; 2. Key Laboratory of Agricultural Product Quality, Safety, Storage, Transportation and Control, Ministry of Agriculture and Rural Affairs, Institute of Agricultural Product Processing, Chinese Academy of Agricultural Sciences, Beijing 100193
Abstract：This study compared the effectiveness of genetic algorithm-back propagation neural network (GA-BPNN) and response surface methodology (RSM) in optimizing the high-pressure ultrasound-assisted enzymatic extraction of collagen from sheep skin to determine the optimal process parameters. The results showed that GA-BPNN had superior performance in model fitting and prediction compared to RSM. The optimal extraction parameters were as follows: high pressure holding time of 23 min, ultrasound time of 22 min, enzyme dosage of 3.2%, and hydrolysis time of 222 min. Under these conditions, the extraction rate of collagen from sheep skin was (80.5 ± 1.6)%, which is 40% higher than that of the traditional papain method. The results of ultraviolet-visible (UV-Vis) spectroscopy and Fourier transform infrared (FTIR) spectroscopy demonstrated that the structure of the extract collagen was complete.
2024, Vol. 38 
