Title: Research on carbon emission accounting of SF6 electrical equipment based on improved random forest algorithm
Authors: Wenwei Zhu; Baichong Pan; Weixian Che; Chenghao Xu
Addresses: Research Centre of Grid Planning, Guangdong Power Grid Co., Ltd., Guangzhou, Guangdong, 510080, China ' Research Centre of Grid Planning, Guangdong Power Grid Co., Ltd., Guangzhou, Guangdong, 510080, China ' Research Centre of Grid Planning, Guangdong Power Grid Co., Ltd., Guangzhou, Guangdong, 510080, China ' Research Centre of Grid Planning, Guangdong Power Grid Co., Ltd., Guangzhou, Guangdong, 510080, China
Abstract: Due to the large convergence error and high interference coefficient of carbon emissions accounting, research on carbon emission accounting of SF6 electrical equipment based on improved random forest algorithm is proposed. Firstly, the arc extinguishing characteristics and insulation performance of SF6 electrical equipment are determined. Then, the differences in the decomposition of substances in SF6 electrical equipment under various conditions are analysed, and differential optical absorption spectroscopy is used to determine the carbon emission equivalent of the equipment. Finally, the OOB error estimation algorithm is introduced to build an improved random forest algorithm model, and the nonlinear activation function is used to determine the synapse strength, and the information function is used to adjust the convergence value of accounting error to complete the carbon emissions accounting. The results indicate that the proposed method can reduce the convergence value of accounting errors and the interference coefficient of accounting results.
Keywords: SF6 electrical equipment; carbon emissions; accounting; arc extinguishing characteristics; OOB error estimation; non-linear activation function.
DOI: 10.1504/IJETP.2024.138544
International Journal of Energy Technology and Policy, 2024 Vol.19 No.1/2, pp.135 - 153
Received: 08 Jun 2023
Accepted: 14 Nov 2023
Published online: 10 May 2024 *