PDF
Full
Outline
To improve the accuracy of cable insulation status assessment, this paper proposed an assesement model of insulation condition based on Bayesian optimization (BO) algorithm and light gradient boosting machine (LightGBM) algorithm. First, all the features in the dataset were combined to form different feature subsets. By traversing all the feature subsets, the optimal feature combination corresponding to the highest accuracy from five-fold cross-validation was identified to complete the input feature selection. Then, the BO algorithm was used to optimize seven hyperparameters in LightGBM. Finally, the proposed BO-LightGBM algorithm was used to assess the cable insulation status. The results show that the feature subset method proposed in this paper can better improve model performance compared with principal component analysis (PCA) and mutual information-based feature selection methods. After optimization by the BO algorithm, the accuracy of the LightGBM model is further enhanced. Compared with particle swarm optimization (PSO) algorithm and genetic optimization (GA) algorithm, the computational efficiency of BO algorithm increases by approximately 80% and 86.9% at the same accuracy level, respectively. Furthermore, compared with other commonly used machine learning algorithms, the performance metrics of the proposed model are optimal.
PDF
88
34
| 科 Family | 属数 Number of genus | 种数 Number of species | 占总种数比例 Percentage of total species (%) | 属 Genus | 种数 Number of species | 占总种数比例 Percentage of total species (%) |
|---|---|---|---|---|---|---|
| 鹅膏菌科Amanitaceae | 2 | 11 | 5.26 | 鹅膏菌属 Amanita | 10 | 4.78 |
| 小菇科 Mycenaceae | 2 | 12 | 5.74 | 丝盖伞属 Inocybe | 5 | 2.39 |
| 多孔菌科 Polyporaceae | 8 | 14 | 6.70 | 蜡蘑属 Laccaria | 5 | 2.39 |
| 红菇科 Russulaceae | 3 | 23 | 11.00 | 小皮伞属 Marasmius | 6 | 2.87 |
| 小菇属 Mycena | 11 | 5.26 | ||||
| 光柄菇属 Pluteus | 5 | 2.39 | ||||
| 红菇属 Russula | 17 | 8.13 | ||||
| 栓菌属 Trametes | 5 | 2.39 |
关闭全屏
No. 86 Xueyuan South Road, Haidian District, Beijing
010-62199257
qkjq@cast.org.cn
Copyright © 2025 China Association for Science and Technology. All rights reserved. For all open access content, the relevant licensing terms apply.
Sponsored by the Office of the Leading Group for Cybersecurity and Informatization of CAST, and supported by Science and Technology Review Publishing House
