| There are a lot of multi-objective optimize problems in the research of medicine and pharmacy, for example, the optimum value of diagnostic test, the optimum allocation of public health resources and the optimal extraction of medicine effective component. Multi-objective optimization problem is to find a group of decision variables values not only satisfying the constraints also make the overall objective function to optimize. Multi-objective optimization is to find a set of selectable and non-controllable optimal solution, it is so called'Pareto Set', by corresponding alternative operations on each sub-objectives.Direct approach, contour method, the rapid dropping law, enumerating law and so on, they have large subjective or local optimazation in the solutions. Decision-makers want to hold more than one solution in practical applications, and in traditional ways, multi-objectives are converted into a single objective or a series of objectives which also have larger subjectivity. It is often the best at a certain goal, but in another target may be the worst. There is no guarantee that the optimal solution for all goals, and they can only provide the only solution. This is a problem which tackle the operations research.It need to be solve.Genetic algorithm (or GA) is a search technique used in computing to find true or approximate solutions to optimization and search problems. Genetic algorithms are robust, efficient, demand for lead-free, highly parallet, simple in principle easy to use and so on. To Multi-dimensional search, the traditional genetic algorithm requires a large initial population size and more generational evolution in order to reach or approach the optimal solution. This must spend a lot of computing resources and computing time. Micro-GA can overcome these problems.micro-GAs'theory was introduced in this research, and then a simulation tool of SGALAB was applied into the optimal anemia decision-making criterion analysis and application in Medicine and Pharmacy. SGALAB is a GAs toolbox for MATLAB by Yi Chen who is now doing his research in University of Glasgow, UK. There are 4 parts of this dissertation:Part 1:Micro-GAs'theory was introduced.Part2:Evaluate the effects of micro-GA and test program. Test micro-GAs by simple two-objective optimization test function, complex two-objective optimization test function and three-objective optimization test functions. The 95% confidence interval of Pareto optimal solution set of simple two-objective optimization test function using micro-GAs contained the crossing point, Pareto optimal front showed a smooth curve of distribution; The Pareto optimal front of complex two-objective optimization test function was zonal distribution; The Pareto optimal front of three-objective optimization test functions was a Non-linear, non-symmetrical surface. micro-GAs can give reasonable Pareto optimal solution set, the results are satisfactory, the program was reliable and it can also be used in actual analysis of the problem.Part 3:Using micro-GA to optimize the conditions of drug extraction. Compare its results with the results of traditional methods.This chapter using micro-GA studies the data of Fructus Schisandrae Chinensis themicrowave-extraction, which including three mutually competed objectives-rate of Extractum extracted (%), Schizandrin armor content (%), total Schisandrae fat element content (%). The results showed that:multi-objectives micro-GAs have achieved more than 88% than in single objective functional values both in Schizandrin armor content and total Schisandrae fat element content. The optimal extraction conditions were higher than any test of uniform design. The results of VEGA and micro-GA are better than the results of uniform design. VEGA used a population size of 30, while the micro-GA used a population size of 6, saving computing resources and computing time. It is better for using micro-GA a number of competing optimization problem. Three goals Micro-GA search for the optimal extraction conditions are 50 grams Schisandrae crushed 76 item,11 times of 86% alcohol joined and 11 minutes under 192W microwave, show that the rate of Extractum extracted is 23.59%, Schizandrin armor content is 4.88% and total Schisandrae fat element content is 10.22%.Using micro-GA optimazes the extraction technology of anti-tumor in Lysimchia clethroides, which including two contradictory objectives-the content of total flavones (%),the content of total saponin (%). The results showed that:multi-objectives micro-GAs have achieved more than 99% than in single objective functional values both in total flavones content (%) and total saponin content (%), and the optimal extraction conditions were higher than any test of orthogonal design. The results of VEGA and micro-GA are better than the results of orthogonal design. VEGA used a population size of 30, while the micro-GA used a population size of 6, saving computing resources and computing time. It is better for using micro-GA a number of competing optimization problem. Two goals micro-GA search for the optimal extraction conditions are adding 8.77 times the 79.90% ethanol, extraction 1.02 times 11 minutes, extract yield 10.44 flavones% total saponin 25.61%.Part 4 Optimize the extraction conditions of Trollius based on micro-GA. This chapter studies the data of Trollius extraction, which including three mutually competed objectives-rate of Extractum extracted (%),Total flavone content(%), and the results show micro-GAs have always achieved above single objective functional 99% values both in rate of Extractum extracted (%) and Total flavone (%) in Water Extracts, and optimized the condition without making others worse off. The best extraction technique including two objectives of micro-GA, was adding 13.73-fold water, soaking for 0.52 hours boil out 3 times and for 1.84 hours, shows that the rate of Extractum extracted is 43.16%, with Total flavonoid content is 6.94%. The results of micro-GA have achieved above single objective functional 99% values both in rate of Extractum extracted (%) and Total flavone (%) in Alcohol Extracts, and optimized the condition without making others worse off. The best extraction technique including two objectives of Micro-GA, was adding 11.95 times 73.11% ethanol, extraction 3 times 1.49 minutes, shows that the rate of Extractum extracted is 42.84%, with Total flavonoid content is 12.36%.According to the result of micro-GA and the convenience of testing, adjust the water extraction condition:14 times of water, soak 0.5 hours, boiled for 2 hours, decoction 3 times; the alcohol extract conditon adjusted:70% ethanol concentration, extraction 1.5 hours, extract 3 times,12 times the amount of solvent. Repeated two tests, the percentage of the extract in water extraction were 45.247% and 45.069%, total flavonoids contents were 7.518% and 7.745%; the percentage of the extract in ethanol extraction were 44.247% and 44.325%, total flavonoids content was 15.032% and 14.090%. The results provided by the micro-GA in water extraction and ethanol conditions for the return test results better than any one of the orthogonal test.It is known that Micro-genetic algorithm is reliable in theory and the procedure is feasible by testing functions. The exploratory research shows that micro-genetic algorithm can obtain reasonable Pareto non-inferior solution set-optimal extraction conditions in application. The verification test with the optimal extraction condition shows that the optimal extraction conditions obtained by micro-genetic algorithm is reliable, the results are satisfactory.
|
PDF Full Text Request