Dissertation > Excellent graduate degree dissertation topics show

Multi-Objective Optimize the Extraction Medicine Effective Component Based on NSGA-Ⅱ

Author: ZhangXiaoLi
Tutor: ChouLiXia
School: Shanxi Medical
Course: Epidemiology and Biostatistics,
Keywords: Elitist Non-dominated Sorting Genetic Algorithm multi-objective optimization Pareto non-inferior solution Optimal Extraction Condition
CLC: TP18
Type: Master's thesis
Year: 2011
Downloads: 124
Quote: 2
Read: Download Dissertation

Abstract

Mult-objective optimization problem is a hotspot in the current studies.With the deepening of theoretical research,the range of its applications is wide increasingly,and has been related to process control、aerospace、artificial intelligence、computer science and other fields.There are also a large mumber of multi-objective optimization problem in the field of medicine, such as the optimal extraction of drug active ingredient,the optimal decision value of diagnostic tests, the optimum test conditions of molecular biology, the optimal treatment of diseases, the optimal allocation of public health resources,and so on. Classical multi-objective evolutionary algorithm to multi-objective problem is converted to one or a series of single objective optimization problem, which has been proven by single objective optimization method to be resolved, but these methods has significant drawbacks.Solving multi-objective problem is a challenge scientists face in solvingpractical problems.Elitist Non-dominated Sorting Genetic Algorithm (NSGA-Ⅱ) is a relatively new multi-objective optimization algorithm,and it has been researched deeply in foreign and applied a lot of practical multi-objective optimization problems which have acheved excellent results.But the study in the domestic research and application are more limited, especially in the medical field,so the study of NSGA-Ⅱhas important significance.The subject using Matlab plug SGALAB toolbox exploited by the original software engineer of Glasgow University of Brithsh test the reliability of the process of NSGA-Ⅱ, and study its application of medicine in the multi-objectice optimization of exraction conditions.The main content is:The first part The overview of NSGA-Ⅱtheory. NSGA-Ⅱuses fast non-dominated sorting method to reduce the complexity of the algorithm; proposed congestion and congestion comparison operator to maintain the diversity of population; introduction of elite strategy infavor of the good parent individuals into the next generation.The second part evaluation and program testing. Using three standard test function test the procedures, and the results showed that:within the range of variables, NSGA-Ⅱsearch for the objective function value that has better approximation with the solution of function. So the process is feasible and the results are satisfactory.The third part Using NSGA-Ⅱto optimize microwave assisted extraction of effective components of Polygonum extract condictions.The three active ingredients of Polygonum is extract yield, total anthraquinone content and stilbene glucoside, which are competitive relationship among the three objectives. When the three targets were larger,the extraction conditions searching by NSGA-Ⅱare microware power 714W, irradiation time 36min, soak time 2h,94% ethanol,5 times the amount of solvent in the extract obtained. Under these conditions, the yield is 10.06%, the total anthraquinone content is 0.417%, stilbene glucoside is 10.71%. NSGA-Ⅱmulti-objective genetic algorithm is a compromise deal, the largest possible access to the solution of the sub-goals, the main goal of total anthraquinones and stilbene glucoside content reached single target of 93.05% and 67.74%,which is better than any results of a program.The forth part Using NSGA-Ⅱto determine the optimal extraction of active ingredients Trollius conditions, and return by verificate test. The results showed that:water extracting conditions is:12.18 times the amount of water, decoction 3 times, then the cream was 43.52%, the total content of flavonoids was 6.72%, NSGA-Ⅱmulti-objective genetic algorithm in the two goals were achieved more than 95% of single target, the effect is satisfactory. Conditions of NSGA-Ⅱsearch for the optimal extraction conditions are:68.08% times the concentration of ethanol, extraction time 1.49h, extract 3 times,11.67 times the amount of solvent, when the paste was 42.14%,12.11% of total flavonoid content. NSGA-Ⅱmulti-objective genetic algorithm reached the target in both the single goal of more than 99%.Test the conditions searched by the NSGA-ⅡLotus water extraction and ethanol conditions, the results are better than any one orthogonal test. NSGA-Ⅱsearch of the Golden Lotus extraction and obtained similar results by micro-genetic algorithm.In summary, multi-objective optimization results by NSGA-Ⅱmulti-objective genetic

Related Dissertations

CLC: > Industrial Technology > Automation technology,computer technology > Automated basic theory > Artificial intelligence theory
© 2012 www.DissertationTopic.Net Mobile