An improved crossover operator of genetic algorithm
3
Citation
0
Reference
20
Related Paper
Citation Trend
Abstract:
In order to effectively overcome the disadvantages of traditional Genetic Algorithm which converge slowly and easily run into local extremism,an improved crossover operator of genetic algorithms was proposed.This operator used the autoadaptive crossover probability and entrusted individual having big irrelevance index with a big elected probability to carry on the crossing operation;The two generations competitive selective operator was designed to improve the traditional genetic algorithm based on roulette.In a simulative example of multi-peaks function,the proposed method can reduce useless crossover effectively and thus the convergence speed and the search capability are greatly improved when compared with the elitist reserved genetic algorithm that keeps best strategy.As a result,the average convergence generations and the probability of getting optimal result are superior to the elitist reserved genetic algorithm.Keywords:
Roulette
Operator (biology)
Fitness proportionate selection
Genetic operator
Local optimum
Cite
In order to solve the conflict between algorithm convergence and the best local answer effectively,puts forward an improved genetic algorithm with a modified selection operator and a modified crossover operator.It can increase the probability of the best answer and well avoid approaching the best local solution by using the modified selection operator,it also increased the probability of finding the best answer,and using the modified crossover operator can speed up the convergence rate,thus shortening the time to find the best answer.The experimental result indicates that the two modified operators' combination can converge to the best answer at higher speed,so it can well solve the contradiction between the convergence rate of genetic algorithm and the best local solution.
Operator (biology)
Genetic operator
Local optimum
Cite
Citations (12)
Prematurity and slow convergence are two difficult problems in genetic algorithm, a new crossover is proposed, named hybrid crossover operator based on pattern, to gain a better convergence to the optimal solution. To retain the diversity of population, the approach of pattern is used, together with the behavior of antitone. The new method can be used for those application problems which are wanted to reach their best value quickly. More experiments show that the new crossover can find the global optimal solution and improve convergence ratio obviously.
Operator (biology)
Premature convergence
Cite
Citations (3)
To overcome global situation problem tradition genetic algorithm has very strong robustness in finding the solution,but crossover probability and mutation probability is fixed and invariable,it caused premature convergence and running inefficient to the solution on complicated problem at later evolution process of tradition genetic algorithm.To this problem,an adaptive genetic algorithm is proposed with varying population size based on lifetimes of the chromosomes to realize population size adjust adaptively and crossover probability adjust adaptively and mutation probability adjust adaptively.Experimental results show that the approach proposed is effective in the capability of global optimization and significantly improves the convergence rate.
Robustness
Premature convergence
Feature (linguistics)
Cite
Citations (1)
The fitness value approximation steers the location of progeny,that obtain optimal solutions quickly.However,it was fallen into the local optimal solutions easily.An improved strategy was present,the individual for crossover operator was accertained,the regions with high fitness were explat.Introducing the process to generate new ones at different rate for every individual,it keep population diversity.Experiment results show that the improved method is useful and effective,and promote the convergence rate in genetic algorithm.
Operator (biology)
Genetic operator
Value (mathematics)
Local optimum
Cite
Citations (0)
<p>Route planning is an important part of road network. To select an optimized route several factors such as flow of traffic, speed limits of road. are concerned. Total cost of such a network depends on the number of junctions between the source and the destination. Due to the growth of the nodes in the network it becomes a tough job to determine the exact path using deterministic algorithms so in such cases genetic algorithms (GA) plays a vital role to find the optimized route. Crossover is an important operator ingenetic algorithm. The efficiency of thegenetic algorithmis directlyinfluenced by the time of a crossover operation. In this paper a new crossoveroperator closest-node pairing crossover (CNPC) is recommended which is explicitly designed to improve the performance of the genetic algorithm compared to other well-known crossover operators such as point based crossover and order crossover. The distance aspect of the network problem has been exploited in this crossover operator. This proposed technique gives a better result compared to the other crossover operator with the fitness value of 0.0048. The CNPC operator gives better rate of convergence compared to the other crossover operators.</p>
Operator (biology)
Cite
Citations (6)
Considering the deficiency of the traditional crossover operator of Genetic Algorithms (GAs) in global searching, the partheno-crossover operator is proposed in this paper. The partheno-crossover, a new crossover operator, still uses the traditional method of crossover. The individual of the population doesn't cross over the individual of the same population but the individual of the completely different with it. When these two individuals have many offspring, only the fittest offspring can replace his father's position in the population. So this offspring either inherits the father's good gene fragments or have a better gene fragments. It both ensures that the population develops to excellent direction, and avoids that the population converges to a local optimal solution. Compared to the Standard Genetic Algorithm (SGA), the traditional improved Genetic Algorithm (TIGA) and the Partheno-Genetic Algorithm (PGA), the improved Genetic Algorithm used the partheno-crossover operator (PCGA) allows better stability of global convergence and some comparative results relative to the optimization of test functions has increase of 10 times, even Several decuples.
Operator (biology)
Genetic operator
Cite
Citations (4)
In order to solve the conflict between algorithm convergence and diversity,after analyzing the structure of operator,this paper puts forward an improved genetic algorithm. This algorithm’s core lies on retaining the outstanding individual through parents,offsprings competition and improved crossover operator to guarantee convergence. On the other hand,carrying on the similarity examination to the crossover gene section based on the Haming distance to enhance validity. Finally,using the self - identify high mutation operator based on gene position diverse to improve the population diversity. The experimental results show that the improved operator improves the ability of searching an optimum solution and increases the convergent speed.
Operator (biology)
Similarity (geometry)
Genetic operator
Position (finance)
Cite
Citations (5)
Based on the combination of fitness property selection and elitist model, numerical crossover and uniform mutation, two methods have been put forward to improve the efficiency of real-coded genetic algorithms. The first method is eliminating the inefficient individuals in crossover operation and reserving the parents for participating selection. The second method is inserting chaos serials into the population when a trend of immature convergence appears. Simulation results of function optimization shows that with the presented methods,the searching precision is enhanced,the phenomenon of immature convergence is effectively overcome,and a satisfying optimization result is obtained.
Cite
Citations (5)
The improvement problem of genetic algorithm is studied based on real coding.To the shortcomings that: the search is inefficient and it is easy to premature convergence,the parameter adjusting problem of genetic algorithm is discussed.The adaptive crossover probability and adaptive mutation probability are proposed,considering the influence of every generation to algorithm and the effect of different individual fitness in every generation.Three testing functions are used to validate the algorithm.The results thaw that the ultimate value,the average algebraic sum and the convergence probability all obtain the preferable values.
Premature convergence
Adaptive mutation
Value (mathematics)
Cite
Citations (7)
A new technique called rank based crossover (RBC) to improve the speed of reaching optimal solutions is introduced for genetic algorithms (GAs). In real life, marriages (crossovers) occur between two individuals of similar status in society and/or from neighboring localities. This principle is extended in case of GAs while selecting crossover partners. In the proposed strategy, the probability of crossover is more when their rank in the whole population is close. This probability function changes with advancing generations, so that the effect of RBC is negligible in the beginning and gradually increases. It could easily control fine tuning of the good chromosomes to achieve fast convergence and reach optimum values. Also, the scheme is not centralized like the elitist approach. Different schemes of the probability function are tried and evaluated. The effectiveness of this new method has been demonstrated on the problems of maximizing complex multi-modal functions. The results are compared with standard genetic algorithms (SGA). Another technique called "fitness scaling" is widely used to adaptively scale the objective function to achieve similar goal. We also compared our results with the "linear fitness scaling" strategy. Results using our RBC strategy are found to be superior to those of the fitness scaling method and the SGA in terms of probability of hitting the maximum value as well as speed of finding the maximum.
Rank (graph theory)
Cite
Citations (6)