Calculation Strategy of Improved Genetic Algorithm Based on Growth Mechanism

To improve the optimization characteristics of genetic algorithm, this paper presents a growth-mechanism calculation strategy, that is, in each iteration of genetic population, a group of new individuals are generated in the neighborhood of a set of optimal values to replace the poor individuals in the population, so as to update the population. On the one hand, the individuals generated in the neighborhood of excellent individuals have the relevant attributes of excellent individuals, which can ensure the correctness of the optimization direction of the population. On the other hand, the neighborhood search also expands the range of optimization of the population and diversifies the population, so as to avoid falling into local optimization. The growth-mechanism calculation strategy involves three factors, including how many outstanding individuals are selected for inheritance, the size of the neighborhood, and how many individuals are eliminated from the population. In this paper, genetic factor, carrier probability and evolutionary weight are used to define the above three factors. The article verifies that the growth-mechanism genetic algorithm is superior to a single genetic algorithm by taking the example of finding the optimal value of a quadratic function. By designing different calculation cases, the influence of the genetic factor, carrier probability, evolution weight, population size and number of calculations on the calculation of the growth-mechanism genetic algorithm is discussed. The study finds that the growth-mechanism calculation strategy can greatly enhance the sensitivity of the population to the optimal solution and can effectively improve the optimization efficiency and convergence speed of the genetic algorithm. The three influencing factors have a great influence on the optimization calculation results. Setting the values of the three influencing factors reasonably can greatly improve the optimization efficiency. Although the smaller the value of the carrier probability is, the higher the optimization efficiency, this value is affected by the range of variables, and there is an optimal genetic factor and an optimal evolution weight in the calculation of the growth-mechanism genetic algorithm. In this paper, when calculating the optimal value of a quadratic function, it is found that the optimal result is obtained when the genetic factor is 0.8 and the evolution weight is 0.5. In addition, as the population size and number of calculations increase, the optimization calculation results improve, however, after reaching a certain value, the optimization result peaks, and increasing the population size and number of calculations will waste calculation time. Therefore, a reasonable setting of the population size and number of calculations will help improve the efficiency of optimization calculations.