Research on Optimization of Power Grid Infrastructure Investment Model Based on Improved Genetic Algorithm

The optimization of infrastructure investment model belongs to the category of project portfolio optimization model, which not only has the general characteristics of project portfolio optimization model, that is, it is necessary to consider the relationship and influence among the projects in the portfolio, and determine the optimal portfolio under the condition of capital constraints. At present, there is little research on the optimal portfolio model of power grid. In this paper, the optimization research of power grid infrastructure investment model based on improved GA (genetic algorithm) is carried out. The optimization of power grid infrastructure investment model and improved GA are proposed and used to solve the model. The improved niche GA divides the objective function into two types: peak height and unequal height. If the function is equal peak value, it can be used directly. The accuracy of the algorithm optimization process is improved. The experimental results show that under the same change range, the change of annual operating cost has the greatest impact on the net present value, followed by investment cost, life span and operating income. The relative prediction error before the improvement is too large, and the relative prediction error after the improvement is reduced. It shows that the combined prediction method of this algorithm can reduce the relative error and make its value more balanced.