Optimize The Extraction Medicine Effective Component Analysis Based On Mendelian Multiobjective Simple Genetic Algorithms

There are a lot of multi-objective optimization problems in the field of medical research. So, Multi-objective optimization is a very important research. The simultaneous optimization in Multi-objective may be competing with each other, the reason why the objective function deviated from the single optimization function is that it rarely allows a good single solution, but often allows a group to be the alternative solutions. The main aim of the Multi-objective optimization technic is to find one or more acceptable solution for Pareto concentration. The traditional method would often transfer multi-objective problems into one or a series of single-objective optimization problem to complete, objective programming method, multiplication and division, linear weighted combination method and effectiveness coefficient of the optimal solution, often the best in a certain target, while the worst in another objective, that can not guarantee all objectives exist the optimal solution, and they can only provide a unique solution, which to be a research question for Operational Research.Genetic algorithm is a simulation of natural evolutionary processes to search the optimal solution, and has made a successful application and has been widespread concern in solving complex problems. However, the optimization technology of genetic algorithm has much work needed to enrich and to improve in the future development. Thus, using genetic research as objects to solve practical problems and to find a more efficient algorithm is of great significance.This study is introducing the principles and methods of Mendelian Multi-objective Simple Genetic Algorithm, and then using the United Kingdom Glasgow University's SGALAB toolbox of Matlab pulg-in which is developped by Chen Yi software engineer, to study several optimation problems of policy-making choice in medical research. The main research contents of this study are:Part I Evaluation and procedures tests of Mendelian Multi-objective Simple Genetic Algorithm Optimized separately to the standard test functions. The results showed that:three test functions are in the given context, and there is a better degree of approximation between the objective function value and the unique solution of function. Tip:single-objective, multi-objective optimization results reach the level of the theoretical value of the test function, the used Toolbox SGALAB of Matlab2009a plug-in beta5008 program is feasible; due to the random of genetic algorithms, it can be applied to run over for several times, select the maximum level of objective function value as the optimal solution.Part II The applied research of Mendelian Multi-objective Simple Genetic Algorithm in the optimal extraction conditions of optimization process of micro-extraction Schisandra and active ingredient of bupleurum. This part using the microwave-assisted extraction information of Schisandra, research extract yield (%), schisandrin content (%), Schisandra the total lignan content (%) the three competing multi-objective optimization results showed that:Schisandrin, the total Schisandra lignans of this MMOSGA method have reached the objective function of the single-objective maximum value of 95% or more, to determine the effect of optimal extraction conditions is higher than any of the program in the trial. In terms of three goals MMOSGA 15, 50 grams of Schisandra Pieces crushed 77 mesh,6.84-fold by adding 89.3% ethanol, extraction for 8.83 minutes in the 417W microwave, extract yield of 20.07%, Schisandrin of 4.95%, Gomi Sub-total lignan of 11.16%. On studying extraction of active ingredient of bupleurum, with MMOSGA as two-objective optimization, if both absorbance of volatile oil and content of saikosaponin a in distilled bupleurum fluid have achieved maximum function value of more than 88% in relation to single objective, then the ideal No.6 program can be regarded as the optimum one. That is to dip under 78℃, collect distilled fluid equaling to 2 times amount of the medicinal materials, extract it for one hour,8 times with 5% ammonia of 95% ethanol extraction, then absorbance of volatile oil can reach 72.49% and content of saikosaponin a up to 32.71%.PartⅢBased on MMOSGA to determine Trollius's active ingredients of the optimal extraction conditions. The information of Trollius's active ingredients used in this part, researched ointment rate (%), total flavonoids content (%), are two competing multi-objective optimization, the results showed that:the process of water extraction nasturtium, MMOSGA can search the rate of the paste, the total flavonoid content have reached the largest single-objective genetic algorithm for the objective function value of more than 95%, to determine the effect of optimal extraction conditions is higher than any of the program in the trial, if taking program 2 of two target genetic algorithm, with 13.31-fold water, soaking 0.61h, decoction 1.49h, decoction three times, a paste could reach 41.79%, with a total flavonoid content could reach 6.73%; alcohol extraction nasturtium process, MMOSGA search out the rate of a paste, with a total flavonoid content has reached the largest single-objective genetic algorithm for the objective function value of more than 95%, to determine the effect of optimal extraction conditions is higher than any of the program in the trial, if taking program 8 of two target genetic Algorithm, with 12-fold the 66% of the ethanol extract of 1.50h, extracted three times, a paste could reach 40.72%, with the total flavonoids can reach 11.81%.In summary, MMOSGA multi-objective optimization results are satisfactory, the program is feasible when applied to solving practical problems has been the desired results.