Oil paper insulation damp state quantitative evaluation method based on Stacking model fusion

Abstract

According to the oil paper insulation damp state quantitative evaluation method based on Stacking model fusion provided by the invention, an oil paper insulation damp state comprehensive evaluation model is constructed by utilizing a Stacking model fusion algorithm, so that damp state quantitative evaluation on to-be-tested sample data under the condition of multiple characteristic quantities is realized; and the accuracy and reliability of the quantitative evaluation result of the oil paper insulation damp state are improved. According to the method, learning principle differences of different algorithms and data characteristics of FDS are fully considered, support vector machine regression, random forest, a K-nearest neighbor method, XGBoost and other algorithms are selected as a primary evaluation model, a kernel ridge regression algorithm is selected as a meta evaluation model, and thus a stacked two-layer Stacking evaluation model is constructed. Experimental results show that compared with a traditional single algorithm evaluation model, the Stacking evaluation model can better process complex sample data, evaluation errors are obviously reduced, and a new method basis can be provided for application research of the artificial intelligence technology in the aspect of oil paper insulation state evaluation.

Description

technical field [0001] The invention belongs to the technical field of oil-paper insulation and the quantitative evaluation of the damp state of oil-paper insulation, in particular to a method for quantitative evaluation of the damp state of oil-paper insulation based on Stacking model fusion. Background technique [0002] Oil-immersed transformers are key equipment for power transmission and distribution. Studies have shown that each time the moisture content in insulating paper is doubled, its mechanical life is reduced by half, so insulation moisture is considered to be the "second biggest killer" that leads to the overall deterioration of the insulation system. Therefore, how to effectively, accurately and timely quantitatively evaluate the damp condition of the transformer oil-paper insulation system is the top priority to ensure the safe and stable operation of the power system. Traditional methods for assessing moisture conditions mainly include coulomb method, chrom...

AI Technical Summary

Problems solved by technology

However, this method only uses a single algorithm, which is easily limited by the randomness of samples, and it is difficult to guarantee the accuracy and robustness of the model.

Images