Purposeful Model Parameters Genesis in Simple Genetic Algorithms

Slides:



Advertisements
Similar presentations
Ali Husseinzadeh Kashan Spring 2010
Advertisements

1 An Adaptive GA for Multi Objective Flexible Manufacturing Systems A. Younes, H. Ghenniwa, S. Areibi uoguelph.ca.
COMPARISON BETWEEN A SIMPLE GA AND AN ANT SYSTEM FOR THE CALIBRATION OF A RAINFALL-RUNOFF MODEL NELSON OBREGÓN RAFAEL E. OLARTE 6th International Conference.
Acoustic design by simulated annealing algorithm
Genetic Algorithms Representation of Candidate Solutions GAs on primarily two types of representations: –Binary-Coded –Real-Coded Binary-Coded GAs must.
Non-Linear Problems General approach. Non-linear Optimization Many objective functions, tend to be non-linear. Design problems for which the objective.
1 Lecture 8: Genetic Algorithms Contents : Miming nature The steps of the algorithm –Coosing parents –Reproduction –Mutation Deeper in GA –Stochastic Universal.
INTEGRATED DESIGN OF WASTEWATER TREATMENT PROCESSES USING MODEL PREDICTIVE CONTROL Mario Francisco, Pastora Vega University of Salamanca – Spain European.
A new crossover technique in Genetic Programming Janet Clegg Intelligent Systems Group Electronics Department.
NORM BASED APPROACHES FOR AUTOMATIC TUNING OF MODEL BASED PREDICTIVE CONTROL Pastora Vega, Mario Francisco, Eladio Sanz University of Salamanca – Spain.
A Comparative Study Of Deterministic And Stochastic Optimization Methods For Integrated Design Of Processes Mario Francisco a, Silvana Revollar b, Pastora.
Intro to AI Genetic Algorithm Ruth Bergman Fall 2002.
D Nagesh Kumar, IIScOptimization Methods: M1L4 1 Introduction and Basic Concepts Classical and Advanced Techniques for Optimization.
Chapter 6: Transform and Conquer Genetic Algorithms The Design and Analysis of Algorithms.
Fermentation Kinetics of Yeast Growth and Production
Genetic Algorithm.
Efficient Model Selection for Support Vector Machines
Mining Feature Importance: Applying Evolutionary Algorithms within a Web- based Educational System CITSA 2004, Orlando, July 2004 Behrouz Minaei-Bidgoli,
Optimal n fe Tian-Li Yu & Kai-Chun Fan. n fe n fe = Population Size × Convergence Time n fe is one of the common used metrics to measure the performance.
1 Paper Review for ENGG6140 Memetic Algorithms By: Jin Zeng Shaun Wang School of Engineering University of Guelph Mar. 18, 2002.
SOFT COMPUTING (Optimization Techniques using GA) Dr. N.Uma Maheswari Professor/CSE PSNA CET.
Hierarchical Distributed Genetic Algorithm for Image Segmentation Hanchuan Peng, Fuhui Long*, Zheru Chi, and Wanshi Siu {fhlong, phc,
Improvement of Multi-population Genetic Algorithms Convergence Time Maria Angelova, Tania Pencheva
Zorica Stanimirović Faculty of Mathematics, University of Belgrade
Applying Genetic Algorithm to the Knapsack Problem Qi Su ECE 539 Spring 2001 Course Project.
ACO for Parameter Settings of E.coli Fed-batch Cultivation Model Stefka Fidanova, Olympia Roeva Bulgarian Academy of Sciences.
Evolving Virtual Creatures & Evolving 3D Morphology and Behavior by Competition Papers by Karl Sims Presented by Sarah Waziruddin.
Solution of a Partial Differential Equations using the Method of Lines
Learning by Simulating Evolution Artificial Intelligence CSMC February 21, 2002.
Parameter Optimization of a Bioprocess Model using Tabu Search Algorithm Olympia Roeva, Kalin Kosev Institute of Biophysics and Biomedical Engineering.
Genetic Algorithms What is a GA Terms and definitions Basic algorithm.
1 Effect of Spatial Locality on An Evolutionary Algorithm for Multimodal Optimization EvoNum 2010 Ka-Chun Wong, Kwong-Sak Leung, and Man-Hon Wong Department.
Improving the Genetic Algorithm Performance in Aerial Spray Deposition Management University of Georgia L. Wu, W.D. Potter, K. Rasheed USDA Forest Service.
Institute of Biophysics and Biomedical Engineering - Bulgarian Academy of Sciences OLYMPIA ROEVA 105 Acad. George Bonchev Str Sofia, Bulgaria
CS 8625 High Performance Computing Dr. Hoganson Copyright © 2003, Dr. Ken Hoganson CS8625 Class Will Start Momentarily… CS8625 High Performance.
Alice E. Smith and Mehmet Gulsen Department of Industrial Engineering
Introduction Metaheuristics: increasingly popular in research and industry mimic natural metaphors to solve complex optimization problems efficient and.
Genetic Algorithms based Parameter Identification of Yeast Fed-Batch Cultivation Angelova M., Tzonkov St., Pencheva T.
1 Autonomic Computer Systems Evolutionary Computation Pascal Paysan.
Application of the GA-PSO with the Fuzzy controller to the robot soccer Department of Electrical Engineering, Southern Taiwan University, Tainan, R.O.C.
Tutorial 2, Part 2: Calibration of a damped oscillator.
Sensitivity Analysis for the Purposes of Parameter Identification of a S. cerevisiae Fed-batch Cultivation Sensitivity Analysis for the Purposes of Parameter.
Artificial Intelligence By Mr. Ejaz CIIT Sahiwal Evolutionary Computation.
EVOLUTIONARY SYSTEMS AND GENETIC ALGORITHMS NAME: AKSHITKUMAR PATEL STUDENT ID: GRAD POSITION PAPER.
Genetic Algorithm(GA)
Genetic Algorithm. Outline Motivation Genetic algorithms An illustrative example Hypothesis space search.
Hirophysics.com The Genetic Algorithm vs. Simulated Annealing Charles Barnes PHY 327.
Genetic Algorithm (Knapsack Problem)
Paper Review for ENGG6140 Memetic Algorithms
Using GA’s to Solve Problems
Discrete ABC Based on Similarity for GCP
Bulgarian Academy of Sciences
Balancing of Parallel Two-Sided Assembly Lines via a GA based Approach
THE METHOD OF FIRST INTEGRALS OF DIFFERENTIAL SYSTEM APPLIED TO BATCH ALCHOHOL FERMENTATION WITH IMMOBILIZED CELLS Silvia B. Popova*, Georgi A. Kostov**
A Comparison of Simulated Annealing and Genetic Algorithm Approaches for Cultivation Model Identification Olympia Roeva.
School of Computer Science & Engineering
Presented by: Dr Beatriz de la Iglesia
C.-S. Shieh, EC, KUAS, Taiwan
Mordechai Shacham, Dept. of Chem
GENETIC ALGORITHMS FOR FEED RATE PROFILES DESIGN
Forecasting The Future of Movies
Comparing Genetic Algorithm and Guided Local Search Methods
Predicting Student Performance: An Application of Data Mining Methods with an Educational Web-based System FIE 2003, Boulder, Nov 2003 Behrouz Minaei-Bidgoli,
Behrouz Minaei, William Punch
Aristotelis Charalampakis and Vlasis Koumousis
○ Hisashi Shimosaka (Doshisha University)
EE368 Soft Computing Genetic Algorithms.
Bulgarian Academy of Sciences
Introduction to Genetic Algorithm and Some Experience Sharing
M. Kezunovic (P.I.) S. S. Luo D. Ristanovic Texas A&M University
Presentation transcript:

Purposeful Model Parameters Genesis in Simple Genetic Algorithms Angelova M., K. Atanassov, T. Pencheva maria.angelova@clbme.bas.bg, krat@bas.bg, tania.pencheva@clbme.bas.bg

Fermentation processes Fermentation processes are: characterized as complex, nonlinear, dynamic systems with interdependent and time-varying process variables; described by non-linear models with a very complex structure. That is why an important step for adequate modeling of non-linear models of FP is the choice of a certain optimization procedure for model parameter identification. BIOMATH 2011, June 15-18, Sofia, Bulgaria

Aims of the investigation To improve the algorithm convergence time in standard simple genetic algorithm (SGA). A stepwise procedure for purposeful genesis concerning intervals of variations of the model parameters to be proposed for a standard simple genetic algorithm aiming to improve significantly the algorithm effectiveness. Such stepwise methodology to be applied to a parameter identification of a S. cerevisiae fed-batch cultivation. Developed procedure to be further validated to a modified simple genetic algorithm with changed sequence of main genetic algorithm operators, namely mutation, crossover and selection. BIOMATH 2011, June 15-18, Sofia, Bulgaria

BIOMATH 2011, June 15-18, Sofia, Bulgaria Genetic algorithms Genetic algorithms (GA) : - are a direct random search technique for finding global optimal solution in complex multidimensional search space; - are based on mechanics of natural selection and natural genetics; - have advantages such as hard problems solving, noise tolerance, easy to interface and hybridize; - are proved to be very suitable for the optimization of highly non-linear problems, - are applied in the area of biotechnology, especially for parameter identification of fermentation process models. BIOMATH 2011, June 15-18, Sofia, Bulgaria

Investigated algorithms SGA-SCM and SGA-MCS SGA-SCM is the basic algorithm originally presented by Goldberg; SGA-MCS is elaborated in previous authors’ investigation where the results show that implementation of the main genetic operators in order mutation, crossover and selection significantly improves calculation time of the algorithm without affecting to the model adequacy. According to preliminary investigation of genetic algorithm’s parameters, the generation gap (GGAP - how many new individuals are created) is the most sensitive one towards to convergence time. Up to almost 40% from calculation time can be saved in cases of SGA-MCS application using GGAP = 0.5 instead of 0.9 without loss of model accuracy. BIOMATH 2011, June 15-18, Sofia, Bulgaria

SGA-SCM - matlab code for generational loop % Create population Chrom = crtbp(NIND, NVAR*PRECI); % Counter gen = 0; % Evaluate Initial Population ObjV = objfun(bs2rv(Chrom,FieldDR))'; % Iterate population while gen < MAXGEN, % Fitness assignement to whole population FitnV = ranking(ObjV); % Select individuals from population SelCh = select(‘rws’, Chrom, FitnV, GGAP); % Recombine selected individuals (crossover) SelCh = recombin(‘xovdp’, SelCh, 0.85); % Mutate offspring SelCh = mut(SelCh, MUTR); % Evaluate Current Population ObjVSel =objfun(bs2rv(SelCh, FieldDR)); % Reinsert offspring in population replacing parents [Chrom ObjV] = reins(Chrom, SelCh, 1, 1, ObjV, ObjVSel'); % Increment counter gen = gen+1 end BIOMATH 2011, June 15-18, Sofia, Bulgaria

SGA-MCS - matlab code for generational loop % Create population Chrom = crtbp(NIND, NVAR*PRECI); % Counter gen = 0; % Evaluate Initial Population ObjV = objfun(bs2rv(Chrom,FieldDR))'; % Iterate population while gen < MAXGEN, % Mutate offspring NewCh = mut(Chrom); % Recombine individuals (crossover) NewCh = recombin(‘xovdp’, NewCh, 0.85); % Fitness assignement to whole population FitnV = ranking(NewCH); % Select individuals from population SelCh = select(‘rws’, NewCh, FitnV, GGAP); % Evaluate Current Population ObjVSel =objfun(bs2rv(SelCh, FieldDR)); % Reinsert offspring in population replacing parents [Chrom ObjV] = reins(Chrom, SelCh, 1, 1, ObjV, ObjVSel'); % Increment counter gen = gen+1 end BIOMATH 2011, June 15-18, Sofia, Bulgaria

Mathematical model of S. cerevisiae fed-batch cultivation (1) – (5) where X, S, E, O2 and O2* are concentrations of biomass, substrate (glucose), ethanol, [g.l-1], oxygen and dissolved oxygen saturation, [%]; F – feeding rate, [l.h-1]; V – volume of bioreactor, [l]; – volumetric oxygen transfer coefficient,[h-1]; Sin – glucose concentration in the feeding solution, [g.l-1]; , qS, qE and are respectively specific rates of growth, substrate utilization, ethanol production and dissolved oxygen consumption, [h-1]. BIOMATH 2011, June 15-18, Sofia, Bulgaria

Specific rates according to functional state modeling approach (6) where – maximum growth rates of substrate and ethanol, [h-1]; kS, kE – saturation constants of substrate and ethanol, [g.l-1]; Yij – yield coefficients, [g.g-1]. Optimization criterion: (7) where Y is the experimental data, Y* – model predicted data BIOMATH 2011, June 15-18, Sofia, Bulgaria

A procedure for purposeful genesis 1. Performing N runs of genetic algorithms 2. Determination of minimum and maximum of the objective function 3. Determination of top, middle and low level of performance with low boundary (LB) and up boundary (UB) following the scheme: Determination of discrimination number: Determination of top, middle and low level of performance with lower bound (LB) and upper bounds (UB) as follows Top_level: lower bound (TL_LB) min J upper bound (TL_UB) min J + ∆-ε Middle_level: lower bound (ML_LB) min J + ∆ upper bound (ML_UB) min J + 2∆-ε Low_level: lower bound (LL_LB) min J + 2∆ upper boundary (LL_UB) max J 4. Determination of parameters average values in each level 5. Based on averaged values, determination of new intervals of variations of model parameters 6. Run of genetic algorithm with new intervals BIOMATH 2011, June 15-18, Sofia, Bulgaria

BIOMATH 2011, June 15-18, Sofia, Bulgaria SGA-SCM results BIOMATH 2011, June 15-18, Sofia, Bulgaria

BIOMATH 2011, June 15-18, Sofia, Bulgaria SGA-SCM analysis BIOMATH 2011, June 15-18, Sofia, Bulgaria

BIOMATH 2011, June 15-18, Sofia, Bulgaria SGA-SCM analysis BIOMATH 2011, June 15-18, Sofia, Bulgaria

BIOMATH 2011, June 15-18, Sofia, Bulgaria SGA-MCS results BIOMATH 2011, June 15-18, Sofia, Bulgaria

BIOMATH 2011, June 15-18, Sofia, Bulgaria SGA-MCS analysis BIOMATH 2011, June 15-18, Sofia, Bulgaria

BIOMATH 2011, June 15-18, Sofia, Bulgaria SGA-MCS analysis BIOMATH 2011, June 15-18, Sofia, Bulgaria

Results after model parameters purposeful genesis BIOMATH 2011, June 15-18, Sofia, Bulgaria

BIOMATH 2011, June 15-18, Sofia, Bulgaria Experimental and model data for biomass and ethanol obtained with SGA-MCS BIOMATH 2011, June 15-18, Sofia, Bulgaria

BIOMATH 2011, June 15-18, Sofia, Bulgaria Experimental and model data for substrate and dissolved oxygen obtained with SGA-MCS BIOMATH 2011, June 15-18, Sofia, Bulgaria

Analysis and conclusions Two kinds of simple genetic algorithms have been investigated aiming to improve the algorithms convergence time. A stepwise procedure for purposeful genesis concerning intervals of variations of model parameters is proposed for the algorithms. When proposed stepwise parameter identification procedure has been applied the GA’s performance has been improved. - SGA-SCM has become with 38% faster; - SGA-MCS has achieved a solution 29% faster. Results obtained with both applications SGA-SCM and SGA-MCS show significantly improvement of algorithm convergence time saving the model accuracy. BIOMATH 2011, June 15-18, Sofia, Bulgaria

BIOMATH 2011, June 15-18, Sofia, Bulgaria Acknowledgements This work is partially supported by the European Social Fund and Bulgarian Ministry of Education, Youth and Science under Operative Program “Human Resources Development”, grant BG051PO001-3.3.04/40 and National Scientific Fund of Bulgaria, grant number DID 02-29. BIOMATH 2011, June 15-18, Sofia, Bulgaria

BIOMATH 2011, June 15-18, Sofia, Bulgaria Thank you for your attention BIOMATH 2011, June 15-18, Sofia, Bulgaria