Scribd
Upload
26 views12 pages

Genetic Algorithm Is A Multiple Point Probabilistic Search

The document describes using a genetic algorithm to solve an optimization problem involving unit commitment for power generation. It discusses: 1) The basic genetic algorithm operators of reproduction, crossover and mutation which are used to probabilistically search for optimal solutions. 2) Encoding the unit commitment problem as strings representing the on/off schedule and real values for the power generated at each time step. 3) Modifications made to the genetic algorithm including elitism to preserve the best solution, a repair operator to handle infeasible solutions, and mutation applied to low fitness strings. 4) The application of the modified genetic algorithm to solve a profit-based unit commitment problem by maximizing profit while satisfying constraints.

Uploaded by

Raghu Gokavarapu
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
26 views12 pages

Genetic Algorithm Is A Multiple Point Probabilistic Search

The document describes using a genetic algorithm to solve an optimization problem involving unit commitment for power generation. It discusses: 1) The basic genetic algorithm operators of reproduction, crossover and mutation which are used to probabilistically search for optimal solutions. 2) Encoding the unit commitment problem as strings representing the on/off schedule and real values for the power generated at each time step. 3) Modifications made to the genetic algorithm including elitism to preserve the best solution, a repair operator to handle infeasible solutions, and mutation applied to low fitness strings. 4) The application of the modified genetic algorithm to solve a profit-based unit commitment problem by maximizing profit while satisfying constraints.

Uploaded by

Raghu Gokavarapu
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 12

INTRODUCTION

Genetic algorithm is a multiple point probabilistic search technique and is characterized by the mechanism of natural selection and natural genetics. Genetic Algorithm consists of three basic operators, namely, reproduction, crossover, and mutation The length of a genetic algorithm string is measured by a genetic string called its fitness value

CONT..

Based on the fitness values of the population strings, two parent strings are selected probabilistically in the process of reproduction. Two child strings are then generated from the parent string by using the mechanism of crossover, where one half of the first parent string is combined with the other half of the second parent. Mutation is then applied on the child string by complementing the child string at selected bit positions, thus introducing variety in the child population.

SOLUTION ENCODING
A solution of the unit commitment is determined by both the on-off schedule (0-1 values) of the units at each hour and the power, Pi (real values), generated by each individual unit every hour. The on-off schedule of the units is stored as a string and the power generated as a real-matrix P where N and T are the number of units and time horizon respectively with dimension N*T.

The following four modifications have been proposed


Modification in parent selection Improvement in child string Elitism Repair operator

parent selection
The first parent in each reproduction is the string having the best fitness value. The second parent is selected from the ordered population using normal selection technique. At the ith reproduction, first parent is the ith string of the population arranged in the order of fitness values. Second parent is selected from the ordered population using normal selection technique.

By Mutation
To improve the quality of child strings, the strings having fitness values less than average fitness values of the previous generation are identified first. The quality is improved by selecting a number of bits at random with probability Pm1 and complimenting.

Elitism
The best individual at generation k (the parent) is maintained in the next generation k+1 if its child a performance inferior than its parent. Without elitism, the best results can be lost during the selection, mutation and crossover operations. Hence the best solution of every generation is copied to the next.

Repair Operator
This operator repairs the solutions that are infeasible regarding the load balance constraints. When the generation is less than load demand, this operator turns on the units, which are off. This is done with a probability of Pr. After the application of this operator, considerable improvement in results has been observed.

GA FOR PROFIT-BASED UC
The modified genetic algorithm as explained above can be applied to the profit based Unit commitment problem, where the main objective is to maximize the profit. The algorithm presented here is for solving the new UC problem is a modification of genetic based UC algorithm

The algorithm proceeds as given below:


1. It first reads in the contract demand prices, and forecast of demand and forecasted spot prices. 2. During the initialization step, a population of UC schedules is initialized as done in the traditional UC 3. For each member of the population, ELD is run to set the level of generation of each unit. 4. The cost of each schedule is calculated from the generator data read in at the beginning of the program. 5. Now it is checked whether constraints are violated or not. 6. If a constraint 7. is violated a penalty is added to the corresponding string. CONT

8. Next the fitness/profit is calculated. 9. Then the algorithm proceeds to reproduction process. 10. The first step of reproduction is to select parents from the population. 11. After selecting parents, candidate children are created using single point crossover. 12. Following crossover, standard mutation is applied. 13. Then child strings having fitness values less than that of average fitness of previous generation are selected and mutation is performed again on these set of strings. 14. This makes the GA to converge fast. 15. After the optimum schedule is obtained exhaustive search is done in order to determine the level of power generation of the generators to get maximum profit.

Flow Chart For Solving The Profit-based UC Problem Using Modified GA

You might also like