Cultural genetic algorithm-based resident medium-and-long-term power consumption prediction method

Abstract

The invention discloses a cultural genetic algorithm-based resident medium-and-long-term power consumption prediction method, which is specifically implemented according to the following steps of: step 1, selecting a data sample: counting a plurality of related data in a certain region within a certain period of time, and taking the data as an influence factor; step 2, data preprocessing: carryingout data cleaning on the data selected in the step 1; step 3, determining a resident medium and long-term electricity consumption prediction model: establishing a multivariate linear regression model-based medium and long-term load prediction model, and taking the data obtained in the step 2 as a training sample to obtain an influence coefficient of a data sample so as to obtain the medium and long-term resident electricity consumption prediction model; and step 4, parameter optimization based on a cultural genetic algorithm: optimizing the parameters in the medium-and-long-term residential electricity consumption prediction model obtained in the step 3 by using the cultural genetic algorithm, obtaining the optimized parameters through the cultural genetic algorithm, and constructing a residential medium-and-long-term electricity consumption prediction model for prediction.

Description

Technical field [0001] The invention belongs to the technical field of genetic algorithms, and specifically relates to a method for predicting mid- and long-term electricity consumption of residents based on cultural genetic algorithms. Background technique [0002] With the continuous increase in the total amount of electricity used by residents, electricity consumption has become an important indicator of people's quality of life. For example, in 2017, the total electricity in Shaanxi reached 129.303.4 billion kW·h, the permanent population was 38.36 million, and the urbanization rate was as high as 57.1%. Among them, residential electricity consumption reached 21.97174 billion kW·h, an increase of 10.85% over the previous year, accounting for 16.99% of the province's total electricity consumption. With the implementation of the coal reduction substitution policy and the further advancement of the "coal to electricity" project, the demand for electricity in the whole society i...

AI Technical Summary

Problems solved by technology

[0005] In 1975, Professor Holland first proposed the Genetic Algorithm (GA), which is a random parallel search algorithm based on the principles of natural selection and genetics [10] , compared with the traditional algorithm, it has many advantages such as global optimization and fast convergence speed, but it is found through research that the defect of the genetic algorithm is that its population diversity gradually disappears when the algorithm is in progress, which will fall into premature convergence

A single introduction of random population can improve this defect, but it will affect the efficiency of the algorithm

Images