genetic algorithms d nagesh kumar, iisc water resources planning and management: m9l2 advanced...
DESCRIPTION
Introduction D Nagesh Kumar, IISc Water Resources Planning and Management: M9L2 3 Real world optimization problems mostly involve complexities like discrete- continuous or mixed variables, multiple conflicting objectives, non-linearity, discontinuity and non-convex region Global optimum cannot be found in a reasonable time due to the large search space For such problems, existing linear or nonlinear methods may not be efficient Various stochastic search methods like simulated annealing, evolutionary algorithms (EA), hill climbing can be used in such situationsTRANSCRIPT
Genetic Algorithms
D Nagesh Kumar, IIScWater Resources Planning and Management: M9L2
Advanced Topics
Objectives
D Nagesh Kumar, IIScWater Resources Planning and Management: M9L22
To introduce the concept of genetic algorithms (GA)
To explain the basic steps in a simple GA Parent Selection
Cross over
Mutation
To introduce Multiobjective GA’s (MOGA)
To apply MOGA in reservoir operation
Introduction
D Nagesh Kumar, IIScWater Resources Planning and Management: M9L23
Real world optimization problems mostly involve complexities like discrete-
continuous or mixed variables, multiple conflicting objectives, non-linearity,
discontinuity and non-convex region
Global optimum cannot be found in a reasonable time due to the large search space
For such problems, existing linear or nonlinear methods may not be efficient
Various stochastic search methods like
simulated annealing,
evolutionary algorithms (EA),
hill climbing can be used in such situations
Introduction
D Nagesh Kumar, IIScWater Resources Planning and Management: M9L24
Among these techniques, the special advantage of EA’s are
Can be applied to any combination of complexities (multi-objective, non-
linearity etc) and also
Can be combined with any existing local search or other methods
Various techniques which make use of EA approach
Genetic Algorithms (GA), Evolutionary Programming, Evolution Strategy,
Learning Classifier System etc.
EA techniques operate mainly on a population search basis.
Basic Concept of EA
D Nagesh Kumar, IIScWater Resources Planning and Management: M9L25
EAs start from a population of possible solutions (called individuals) and move
towards the optimal one by applying the principle of Darwinian evolution theory i.e.,
survival of the fittest
Objects forming possible solution sets to the original problem is called phenotype
The encoding (representation) or the individuals in the EA is called genotype
The mapping of phenotype to genotype differs in each EA technique
In GA, variables are represented as strings of numbers (normally binary)
Let the number of design variables be n
Let each design variable is given by a string of length ‘l
Then the design vector will have a total string length ‘nl’
Basic Concept of EA…
D Nagesh Kumar, IIScWater Resources Planning and Management: M9L26
For example, if the string length be 4 for each design variable and there are 3 design
variables,
Then the chromosome length is 12 as shown
An individual consists of a genotype and a fitness function
Fitness Function: Represents the quality of the solution
Forms the basis for selecting the individuals and thereby facilitates
improvements
17,4 321 xandxx
Basic Concept of EA…
D Nagesh Kumar, IIScWater Resources Planning and Management: M9L27
i = 0
Initialize population P0
Evaluate initial population
while ( ! termination condition)
{
i = i+1
Perform competitive selection
Create population Pi from Pi-1 by recombination and mutation
Evaluate population Pi
}
Pseudo code for a simple EA
Basic Concept of EA…
D Nagesh Kumar, IIScWater Resources Planning and Management: M9L28
Flowchart indicating the steps of a simple genetic
algorithm
Generate Initial Population
Start
Encode Generated Population
Evaluate Fitness Functions
Meets Optimization Criteria?
Best Individuals
Yes
Stop
Selection (select parents)
Mutation (mutate offsprings)
Crossover (selected parents)
No
REGENERATION
Basic Concept of EA…
D Nagesh Kumar, IIScWater Resources Planning and Management: M9L29
Initial population is randomly generated
Individuals with better fitness functions from generation ‘i' are taken to generate
individuals of ‘i+1’th generation
New population (offspring) is created by applying recombination and mutation to
the selected individuals (parents)
Finally, the new population is evaluated and the process is repeated
The termination condition may be a desired fitness function, maximum number of
generations etc
Parent Selection
D Nagesh Kumar, IIScWater Resources Planning and Management: M9L210
Individuals are distinguished based on their fitness function value
According to Darwin's evolution theory the best ones should survive and
create new offspring for the next generation
Different methods are available to select the best chromosomes
Roulette wheel selection,
Rank selection,
Boltzman selection,
Tournament selection,
Steady state selection
Parent Selection : Roulette wheel selection
D Nagesh Kumar, IIScWater Resources Planning and Management: M9L211
Each individual is selected with a probability proportional to its fitness value
In other words, an individual is selected depending on the percentage contribution to
the total population fitness
Thus, weak solutions are eliminated and strong solutions survive to form the next
generation Consider a population containing four strings shown
Candidate Fitness value Percentage of total fitness
1011 0110 1101 1001 109 28.09
0101 0011 1110 1101 76 19.59
0001 0001 1111 1011 50 12.89
1011 1111 1011 1100 153 39.43
Total 388 100
Parent Selection : Roulette wheel selection…
D Nagesh Kumar, IIScWater Resources Planning and Management: M9L212
Each string is formed by concatenating four substrings representing variables a, b, c
and d. Length of each string is taken as four bits
First column represents the possible solution in binary form
Second column gives the fitness values of the decoded strings
Third column gives the percentage contribution of each string to the total fitness of
the population
Thus, the probability of selection of candidate 1, as a parent of the next generation is
28.09%
Probabilities of other candidates 2, 3, 4 are 19.59, 12.89 and 39.43 respectively
These probabilities are represented on a pie chart
Then four numbers are randomly generated between 1 and 100
Parent Selection : Roulette wheel selection…
D Nagesh Kumar, IIScWater Resources Planning and Management: M9L213
The likeliness of these numbers falling in the region of candidate 2 might be once, whereas for candidate 4 it might be twice and candidate 1 more than once and for candidate 3 it may not fall at all
Thus, the strings are chosen to form the parents of the next generation
The main disadvantage of this method is when the fitnesses differ very much
For example, if the best chromosome fitness is 90% of the entire roulette wheel then the other chromosomes will have very few chances to be selected
Parent Selection: Rank Selection
D Nagesh Kumar, IIScWater Resources Planning and Management: M9L214
Population is ranked first
Every chromosome will be allotted with one fitness corresponding to this ranking
The worst will have fitness 1, second worst 2 etc. and the best will have fitness N
(number of chromosomes in population)
By doing this, all the chromosomes have a chance to be selected
But this method can lead to slower convergence, because the best chromosomes
may not differ much from the others
Parent Selection
D Nagesh Kumar, IIScWater Resources Planning and Management: M9L215
Through selection new individuals cannot get introduced into the
population
Selection cannot find new points in the search space
New individuals are generated by genetically-inspired operators known are
crossover and mutation.
Crossover
D Nagesh Kumar, IIScWater Resources Planning and Management: M9L216
Crossover can be of either one-point or two-point scheme
One point crossover: Selected pair of strings is cut at some random position and their
segments are swapped to form new pair of strings
Consider two 8-bit strings given by '10011101' and '10101011'
Choose a random crossover point after 3 bits from left
100 | 11101
101 | 01011
Segments are swapped and the resulting strings are
10001011
10111101
Crossover…
D Nagesh Kumar, IIScWater Resources Planning and Management: M9L217
Two point crossover: There will be two break points in the strings that are randomly
chosen
At the break-point the segments of the two strings are swapped so that new set of
strings are formed
If two crossover points are selected as
100 | 11 | 101
101 | 01 | 011
After swapping both the extreme segments, the resulting strings formed are
10001101
10111011
Mutation
D Nagesh Kumar, IIScWater Resources Planning and Management: M9L218
Mutation is applied to each child individually after crossover
Randomly alters each gene with a small probability (generally not greater than 0.01)
Injects a new genetic character into the chromosome by changing at random a bit in
a string depending on the probability of mutation
Example: 10111011 is mutated as
10111111
Sixth bit '0' is changed to '1‘
In mutation process, bits are changed from '1' to '0' or '0' to '1' at the randomly
chosen position of randomly selected strings
Real-coded GAs
D Nagesh Kumar, IIScWater Resources Planning and Management: M9L219
GAs work with a coding of variables i.e., with a discrete search space
GAs have also been developed to work directly with continuous variables
In these cases, binary strings are not used
Instead, the variables are directly used
After the creation of population of random variables, a reproduction
operator can be used to select good strings in the population.
Areas of Application in Water Resources
D Nagesh Kumar, IIScWater Resources Planning and Management: M9L220
Water distribution systems
Hydrological modeling
Watershed Management
Groundwater modeling
Reservoir Operation
Advantages and Disadvantages of EA
D Nagesh Kumar, IIScWater Resources Planning and Management: M9L221
Advantages EA can be efficiently used for highly complex problems with multi-objectivity, non-
linearity etc Provides not only a single best solution, but the 2nd best, 3rd best and so on as
required. Gives quick approximate solutions Can incorporate with other local search algorithms
Disadvantages Optimal solution cannot be ensured on using EA methods Convergence of EA techniques are problem oriented Sensitivity analysis should be carried out to find out the range in which the model is
efficient Implementation requires good programming skill
Multi-objective Evolutionary Algorithms
D Nagesh Kumar, IIScWater Resources Planning and Management: M9L222
Genetic algorithms are efficient in solving multiobjective problems Considerations in Multi-Objective Evolutionary Algorithms (MOEAs) implementation
are
1. Preserve non-dominated points – elitism
2. Progress towards points on Pareto front
3. Maintain diversity of points on Pareto Front (phenotype) and/or Pareto Optimal solutions
(genotype)
4. Provide decision maker a limited number of Pareto Front (PF) points.
Non-dominated solutions are always better than 1st-level dominated solutions, which
are always better than 2nd-level dominated solutions, etc. Within the same level of dominance, solutions which are isolated are better than
solutions that are clumped together.
Flow chart of Multi-objective Genetic Algorithm with Elitism
D Nagesh Kumar, IIScWater Resources Planning and Management: M9L223
Multi-objective reservoir system optimization –Multi-objective reservoir system optimization –Case StudyCase Study
D Nagesh Kumar, IIScWater Resources Planning and Management: M9L224
Bhadra Reservoir –Multipurpose reservoirLocation - Chickmangalur Dt, Karnataka state, India;
75o38’20’’ E longitude and 13o42’ N latitude
Schematic diagram of Bhadra reservoir Project
Left bank canal
Irrigated area6, 367 ha
Irrigated area 87, 512 ha
PH1PH2
PH3
Bhadra river
Reservoir
Left turbine capacity=2,000 kW
Bed turbine capacity=24,000 kW
Right turbine capacity=13,200 kW
Right bank canal
River Water Quality
Multi Objective Reservoir Operation Model
D Nagesh Kumar, IIScWater Resources Planning and Management: M9L225
Subject to constraints on
System dynamics
Storage bounds
Maximum power production limits
Irrigation demands and
Water quality requirements
1. Minimize
Irrigation deficit (f1):
2. Maximize
Hydropower production (f2):
Multi Objective Reservoir Operation Model: Constraints
D Nagesh Kumar, IIScWater Resources Planning and Management: M9L226
Reservoir storage continuity constraints
for all t=1, 2,…,12
)(S ,3,2,11t ttttttt OERRRIS
Storage bound constraintsSmin ≤ St ≤ Smax
Turbine capacity Limitsp R1,t H1,t ≤ E1,max
p Rr,t H2,t ≤ E2,max
p R3,t H3,t ≤ E3,max
Canal capacity limits R1,t ≤ C1,max R2,t ≤ C2,max
for all t=1, 2,…,NT
Irrigation demands D1min, t ≤ R1,t ≤ D1max, t D2min, t ≤ R2,t ≤ D2max, t
for all t=1, 2,…,NTWater Quality Requirements
R3,t ≥ MDTt
for all t=1, 2,…,12
for all t=1, 2,…,12
Pareto optimal solution for reservoir Pareto optimal solution for reservoir operationoperation
D Nagesh Kumar, IIScWater Resources Planning and Management: M9L227
Improvement in Pareto optimal front over the iterations. f1 is annual squared irrigation deficit; f2 is hydropower generated MkWh
0 2 4 6 8 10 12
x 104
150
160
170
180
190
200
210
220
230
gen=50gen=200gen=500
f1
f2
Model Application
D Nagesh Kumar, IIScWater Resources Planning and Management: M9L228
MOGA model is solved for three different inflow scenarios into the reservoir
Scenario 1: Mean monthly inflows – 0.5 * SD
Scenario 2: Mean monthly inflows
Scenario 3: Mean monthly inflows + 0.5 * SD
where SD is the standard deviation of monthly flows
Pareto optimal front
D Nagesh Kumar, IIScWater Resources Planning and Management: M9L229
Pareto optimal front, showing the trade-off between irrigation ( f1) and hydropower ( f2) for different inflow scenarios.
f1 = sum of squared irrigation deficits, (Mm3)2; f2 = hydropower generated, (MkWh)
Reservoir operating policies
D Nagesh Kumar, IIScWater Resources Planning and Management: M9L230
Reservoir operating policies for different
inflow scenarios, showing the initial
storages for different situations,
viz., equal priority case, irrigation only
priority case and hydropower only
priority case.
Optimal release policy
D Nagesh Kumar, IIScWater Resources Planning and Management: M9L231
Optimal release policy obtained for equal priority case, showing releases in Mm3 for
Left bank canal (R1), Right bank canal (R2) and River bed (R3) for different inflow
scenarios.
Advantages of MOEAs
D Nagesh Kumar, IIScWater Resources Planning and Management: M9L232
MOEAs are easy to adopt and can provide efficient solutions for multi-objective
problems
MOEAs are capable of handling nonlinear objectives/ constraints, disconnected
Pareto-fronts, non-convex decision space
MOEAs can find solutions to extremely complex and high dimensional real-world
applications in reasonable computation time
MOEAs have high potential for multi-objective optimization of hydrological &
water resources problems
D Nagesh Kumar, IIScWater Resources Planning and Management: M9L2
Thank You