| 776 | 1 | 250 |
| 下载次数 | 被引频次 | 阅读次数 |
针对多分类问题中样本类间不均衡引起的电缆状态分类准确性不高的问题，提出一种基于贝叶斯优化极端梯度提升树的电缆状态分类方法。首先，利用贝叶斯优化对极端梯度提升树算法里面的超参数进行训练，获取最优超参数配置。其次，将最优超参数配置应用于极端梯度提升树算法中，得到Bo-XGBoost分类模型。最后，通过实例验证该分类方法相较于SVM、TabNet、LightGBM等方法有更高的准确性，可为电缆状态分类提供一种新方向。
Abstract：Addressing the issue of low accuracy in cable condition classification due to imbalanced sample classes in multiclass classification problems, a cable condition classification method based on Bayesian-optimized extreme gradient boosting was proposed. Firstly, Bayesian optimization was employed to train the hyperparameters within the XGBoost algorithm, with the aim of acquiring the optimal hyperparameter configuration. Then, this optimal hyperparameter configuration was applied to the XGBoost algorithm, which resulted in the Bo-XGBoost classification model. Finally, the verification through case studies demonstrated that this classification method achieved higher accuracy compared to methods such as SVM, TabNet, and LightGBM, thereby providing a new direction for cable condition classification.
[1] 江秀臣，盛戈皞.电力设备状态大数据分析的研究和应用[J].高电压技术，2018,44(4):1041-1050.JIANG X C,SHENG G H.Research and application of big data analysis of power equipment condition[J].High voltage engineering,2018,44(4):1041-1050.
[2] 江秀臣，罗林根，余钟民，等.区块链在电力设备泛在物联网应用的关键技术及方案[J].高电压技术，2019,45(11):3393-3400.JIANG X C,LUO L G,YU Z M,et al.Technologies and solutions of blockchain application in power equipment ubiquitous Internet of Things[J].High voltage engineering,2019,45(11):3393-3400.
[3] 鞠平，周孝信，陈维江，等.“智能电网+”研究综述[J].电力自动化设备，2018,38(5):2-11.JU P,ZHOU X X,CHEN W J,et al."Smart Grid Plus" research overview[J].Electric power automation equipment,2018,38(5):2-11.
[4] LIU H,HUANG F P,LI H,et al.A big data framework for electric power data quality assessment[C]//The 14th Web Information Systems and Applications Conference.Piscataway:IEEE Press,2017:289-292.
[5] 李玉凌，李二霞，樊勇华，等.一种基于概率神经网络算法的电缆状态评估方法：CN106251059-B[P].2016-12-21.LI Y L,LI E X,FAN Y H,et al.A cable condition evaluation method based on probabilistic neural network algorithm:CN106251059-B[P].2016-12-21.
[6] 周路遥，曹俊平，王少华，等.基于多状态量特征及变化规律的高压电缆状态综合评估[J].高电压技术，2019,45(12):3954-3963.ZHOU L Y,CAO J P,WANG S H,et al.Comprehensive state evaluation of high voltage cable based on multi-state variables characteristics and variation law[J].High voltage engineering,2019,45(12):3954-3963.
[7] 马骁，马弢，王晓峰，等.基于PDC多特征量权重的电缆绝缘老化状态评估方法研究[J].绝缘材料，2022,55(9):80-88.MA X,MA T,WANG X F,et al.Evaluation method of cable insulation ageing state based on PDC multi feature weight[J].Insulating materials,2022,55(9):80-88.
[8] 董芝春，冯宝，周慧彬，等.一种基于长短时记忆网络的配网电缆状态评估方法及系统：CN117171642-A[P].2023-12-05.DONG Z C,FENG B,ZHOU H L,et al.A distribution network cable condition evaluation method and system based on long short-term memory network:CN117171642-A[P].2023-12-05.
[9] 李占山，刘兆赓.基于XGBoost的特征选择算法[J].通信学报，2019,40(10):101-108.LI Z S,LIU Z G.Feature selection algorithm based on XGBoost[J].Journal on communications,2019,40(10):101-108.
[10] DE BOER P T,KROESE D P,MANNOR S,et al.A tutorial on the cross-entropy method[J].Annals of operations research,2005,134(1):19-67.
[11] 崔佳旭，杨博.贝叶斯优化方法和应用综述[J].软件学报，2018,29(10):3068-3090.CUI J X,YANG B.Survey on Bayesian optimization methodology and applications[J].Journal of software,2018,29(10):3068-3090.
[12] 李亚茹，张宇来，王佳晨.面向超参数估计的贝叶斯优化方法综述[J].计算机科学，2022,49(S1):86-92.LI Y R,ZHANG Y L,WANG J C.Overview of Bayesian optimization methods for hyperparameter estimation[J].Computer science,2022,49(S1):86-92.
[13] 王方成，刘玉敏，崔庆安.基于高斯过程回归的混合型参数建模及优化[J].统计与决策，2023,39(1):34-39.WANG F C,LIU Y M,CUI Q A.Hybrid parameter modeling and optimization based on Gaussian process regression[J].Statistics & decision,2023,39(1),34-39.
[14] 谭笑，陈杰，刘洋，等.基于朴素贝叶斯的电缆老化状态评估方法和装置：CN111610407B[P].2022-09-01.TAN X,CHEN J,LIU Y,et al.Cable aging state assessment method and apparatus based on naive Bayes:CN111610407-A[P].2020-09-01.
[15] SZEGEDY C,LIU W,JIA Y Q,et al.Going deeper with convolutions[C]//IEEE Conference on Computer Vision and Pattern Recognition.Piscataway:IEEE Press,2015:1-9.
[16] ARIK S ?,PFISTER T.TabNet:attentive interpretable tabular learning[J].Proceedings of the AAAI conference on artificial intelligence,2021,35(8):6679-6687.
[17] TIAN Z H,ZHANG R,HOU X Y,et al.FederBoost:private federated learning for GBDT[J].IEEE transactions on dependable and secure computing,2024,21(3):1274-1285.
[18] ZHAO G S,WANG Y,WANG J.Intrusion detection model of Internet of Things based on LightGBM[J].IEICE transactions on communications,2023,106(8):622-634.
[19] 张淑芬，董燕灵，徐精诚，等.基于目标扰动的AdaBoost 算法[J].通信学报，2023,44(2):198-209.ZHANG S F,DONG Y L,XU J C,et al.AdaBoost algorithm based on target perturbation[J].Journal on Communications,2023,44(2):198-209.
基本信息:
DOI：10.13705/j.issn.1671-6841.2024079
中图分类号:TP18;TM75
引用信息:
[1]佘维,王欣,陈斌,等.基于贝叶斯优化极端梯度提升树的电缆状态分类研究[J].郑州大学学报(理学版),2025,57(06):1-7.DOI:10.13705/j.issn.1671-6841.2024079.
基金信息:
嵩山实验室预研项目(YYYY022022003); 河南省重点研发与推广专项(科技攻关)(212102310039)