| 引用本文: | 谭贵生,赵 波,张桂莲,等.多策略改进DBO算法与KELM的变压器故障辨识[J].电力系统保护与控制,2025,53(14):111-122.[点击复制] |
| TAN Guisheng,ZHAO Bo,ZHANG Guilian,et al.Transformer fault identification based on multi-strategy improved DBO algorithm and KELM[J].Power System Protection and Control,2025,53(14):111-122[点击复制] |
|
| 摘要: |
| 针对油浸式变压器故障诊断中因样本存在冗余特征，导致故障诊断精度低的问题，提出一种新的多策略改进蜣螂算法(multi-strategy improved dung beetle optimizer, MSIDBO)优化核极限学习机(kernel extreme learning machine, KELM)的变压器故障辨别模型。首先，利用随机森林(random forest, RF)与核主成分析法(kernel principal component analysis, KPCA)对变压器原始数据进行特征提取，合理降低特征量的维度。其次，引入改进的Circle混沌映射、变螺旋搜索机制、非线性控制因子、融合正余弦算法和融合多种群差分进化算法的变异策略对蜣螂算法(dung beetle optimizer, DBO)进行改进，提高全局搜索能力和收敛精度。最后，利用MSIDBO对KELM中的核参数和正则化参数进行优化，构建KPCA-MSIDBO-KELM的变压器故障诊断模型。实验表明，其诊断准确率为94.07%。与DBO-KELM、WOA-KELM、HHO-KELM、GWO-KELM和PSO-KELM故障模型进行对比分析，准确率分别提高了2.54%、3.39%、5.93%、7.63%和13.56%。相比其他模型，所提方法能够有效提高变压器故障诊断的准确率。 |
| 关键词: 变压器 故障诊断 多策略改进蜣螂算法 核极限学习机 核主成分分析 |
| DOI：10.19783/j.cnki.pspc.241245 |
| 投稿时间：2024-09-13修订日期：2025-02-19 |
| 基金项目:国家自然科学基金项目资助(61901101)；丽江市科技局项目资助(2022LJSHFZ013)；丽江文化旅游学院项目资助(2023xshb04)；新疆政法学院项目资助(XZZK2024002) |
|
| Transformer fault identification based on multi-strategy improved DBO algorithm and KELM |
| TAN Guisheng1,ZHAO Bo2,ZHANG Guilian1,LIU Dandan3,SHI Yijin4 |
| (1.Lijiang Culture and Tourism College, Lijiang 674199, China;
2. School of Automation Engineering, Northeast Electric Power University, Jilin 132012, China;
3. Lijiang Power Supply Bureau, Yunnan Power Grid Co., Ltd., Lijiang 674199, China;
4. Xinjiang University of Political Science and Law, Tumushuke 843900, China) |
| Abstract: |
| Aiming at the low accuracy of fault diagnosis in oil-immersed transformers caused by redundant features in the sample data, this paper proposes a new transformer fault identification model based on a multi-strategy improved dung beetle optimizer (MSIDBO) and kernel extreme learning machine (KELM) is proposed. First, random forest (RF) and kernel principal component analysis (KPCA) are used to extract the features from the raw transformer data, effectively reducing dimensionality. Next, the dung beetle optimizer (DBO) is comprehensively improved through multiple strategies, including improved Circle chaotic mapping, variable spiral search mechanism, nonlinear control factor, fusion with the sine-cosine algorithm, and a hybrid multi-population differential evolution mutation strategy. These enhancements significantly improve the algorithm’s global search ability and convergence accuracy. Finally, the MSIDBO algorithm is applied to optimize the kernel and regularization parameters of KELM, constructing the KPCA-MSIDBO-KELM transformer fault diagnosis model. Experiments show that the constructed transformer fault diagnosis model achieves an accuracy of 94.07%, which is 2.54%, 3.39%, 5.93%, 7.63% and 13.56% higher than DBO-KELM, WOA-KELM, HHO-KELM, GWO-KELM, and PSO-KELM. Overall, the proposed method significantly improves the accuracy of transformer fault diagnosis. |
| Key words: transformer fault diagnosis multi-strategy improved dung beetle optimizer kernel extreme learning machine kernel principal component analysis |
