基于AHP-CRITIC混合加权法与BP-ANN法的加味柴桂方提取工艺优化OA

Objective:To optimize the extraction process of modified Chaigui Formula by analytic hierarchy process-criteria importance through intercriteria correlation(AHP-CRITIC)and back propagation artificial neural network(BP-ANN).Methods:An orthogonal design was adopted,with the content of salvianolic acid B,saikosaponin A,saikosaponin B2,and baicalin and the dry extract yield as evaluation indicators.The factors under investigation included water volume,extraction duration,and number of extraction cycles.The AHP-CRITIC method was employed for weighted scoring of these indicators,and a BP-ANN model was established to optimize the extraction process of modified Chaigui Formula.Results:The AHP-CRITIC hybrid weighting method determined the weight coefficients of salvianolic acid B,saikosaponin A,saikosaponin B2,baicalin,and dry extract yield in modified Chaigui Formula as 0.342 6,0.190 7,0.189 0,0.151 3,and 0.126 4,respectively.The analysis of variance showed that the modified Chaigui Formula prepared under optimal extraction process,determined based on the weight coefficients assigned by the AHP-CRITIC method,had a comprehensive score of 75.39,while that predicted by the BP-ANN had a comprehensive score of 76.04.Considering production costs,the finalized optimal process was determined as follows:12 times the solvent volume,extraction for 1 h,and 3 extraction cycles.The results of the validation experiments indicated that the comprehensive score of modified Chaigui Formula prepared under the optimal process was 75.56,with a relative standard deviation(RSD)of 1.98%.Conclusion:The optimized extraction process is stable and reliable,offering a new approach for optimizing the extraction processes of clinically empirical formulas.