LCC prediction model of a transformer substation in a high-latitude severe cold area of an LSSVM

Abstract

The invention belongs to the technical field of electric power, and particularly relates to an SSA-based device. The invention discloses an LCC prediction model of a transformer substation in a high-latitude severe cold area of an LSSVM. The method comprises the following steps: setting parameters; Initializing a population; setting a fitness function; starting optimization; finishing optimization. According to the method, the global optimal solution of the LCC prediction model parameters is searched by adopting the SSA algorithm, and the obtained optimal parameter combination is used as the LS-SVM paraamters. Therefore, precision of the prediction model is enhanced. The method comprises the following steps of establishing a transformer substation LCC prediction model based on an SSA algorithm. The SSA optimization LS-is verified by performing comparative analysis on a calculation example and prediction results of other prediction models. Due to the superiority of the performance of the SVM prediction model, rapid and high-precision LCC prediction of the transformer substation can be realized when the transformer substation is newly built, and economic and technical assessment of the transformer substation in a high-latitude severe cold region is facilitated.

Description

technical field [0001] The invention belongs to the field of electric power technology, and in particular relates to an SSA-LSSVM-based LCC prediction model of substations in high-latitude severe cold regions. Background technique [0002] With the construction of large-scale long-distance transmission networks in my country, the construction of substations in high-latitude cold areas is increasing. Due to the unique environmental and climatic conditions such as low temperature and frozen soil, the initial investment cost of substation construction in high-latitude cold areas is significantly higher than that in other areas. Affecting the post-operation and maintenance costs, it is urgent to establish a substation LCC prediction model suitable for high-latitude severe cold regions to improve the efficiency of power grid asset management. [0003] China's substation investment and construction are mostly based on the whole process management, that is, only pay attention to the...

AI Technical Summary

Problems solved by technology

However, due to the lack of historical data, the LCC modeling of substations conforms to the characteristics of small samples. In the case of small samples, the neural network is easy to fall into overfitting and local optimum, resulting in a greatly reduced prediction accuracy.

[0016] Common algorithm tools such as genetic algorithm and particle swarm algorithm are commonly used to optimize the parameters of the least squares support vector machine. value

Images