Abstract:Abstract：To optimize the extraction process of Zingiber officinale Roscoe flavonoids，Box-Behnken experiment design was adopted to arrange the tests based on single factor experiments and factorial experiments，BP neural network model was established using the data of Box-Behnken experiments，the back propagation （BP）neural network model was used to simulate the nonlinear relationship between the significant factors and the yield of Zingiber officinale Roscoe flavonoids，BP neural network combined with genetic algorithm were used to optimize the extraction process of Zingiber officinale Roscoe flavonoids.The results showed that the optimal extraction process conditions of Zingiber officinale Roscoe flavonoids were ethanol concentration 59%，extracting temperature 63℃，solid-liquid ratio 1∶37（g/mL），extracting time 90 min.The yield of Zingiber officinale Roscoe flavonoids could reach 14.685 2 mg/g under these conditions.The results could also provide a reference for further exploitation research of Zingiber officinale Roscoe flavonoids.