| 基于ANFIS的电磁换向阀寿命预测 |
| |
| 引用本文： | 郭锐,何丝丝,叶欣,刘光恒. 基于ANFIS的电磁换向阀寿命预测[J]. 液压与气动, 2023, 0(1): 1-11. DOI: 10.11832/j.issn.1000-4858.2023.01.001 |
| |
| 作者姓名： | 郭锐 何丝丝 叶欣 刘光恒 |
| |
| 作者单位： | 1.航天发射场可靠性技术重点实验室, 海南 海口 571126;2.燕山大学 河北省重型机械流体动力传输与控制重点实验室, 河北 秦皇岛 066004;3.燕山大学 河北省特种运载装备重点实验室, 河北 秦皇岛 066004;4.燕山大学 先进锻压成形技术与科学教育部重点实验室, 河北 秦皇岛 066004 |
| |
| 基金项目： | 国家重点研发计划(2019YFB2005204);国家自然科学基金(52075469,12173054);河北省重点研发计划(19273708D);航天发射场可靠性技术重点实验室开放课题(sys-2021-03-3) |
| |
| 摘 要： | 以电磁换向阀的出入口压降作为失效判据，从流量退化角度深入分析，进行电磁换向阀的寿命预测研究。首先使用改进的集合经验模态分解(Modified Ensemble Empirical Mode Decomposition, MEEMD)方法多尺度分解测试数据，以欧式距离的IMFs分量筛选规则完成噪声信号的降噪重构。随后利用时域分析、频域分析和时频域分析提取特征参数，并用核主元(Kernel Principal Component Analysis, KPCA)方法融合处理，经过3次指数平滑处理，构建了电磁换向阀退化评估指标。融合退化评估指标，训练自适应模糊神经网络(Adaptive Network-based Fuzzy Inference System, ANFIS),通过对未失效样本进行压降趋势预测，实现了电磁换向阀的寿命预测。结果显示，ANFIS的预测指标与实际指标的差异性较小，预测结果准确。
|
| 关 键 词： | 电磁换向阀 ANFIS 寿命预测 KPCA 退化评估指标 |
| 收稿时间： | 2022-05-23 |
Life Prediction of Electromagnetic Directional Valve Based on ANFIS |
| |
| GUO Rui,HE Si-si,YE Xin,LIU Guang-heng. Life Prediction of Electromagnetic Directional Valve Based on ANFIS[J]. Chinese Hydraulics & Pneumatics, 2023, 0(1): 1-11. DOI: 10.11832/j.issn.1000-4858.2023.01.001 |
| |
| Authors: | GUO Rui HE Si-si YE Xin LIU Guang-heng |
| |
| Affiliation: | 1. Key Laboratory of Space Launch Site Reliability Technology, Haikou, Hainan 571126; 2. Hebei Provincial Key Laboratory of Heavy Machinery Fluid Power Transmission and Control, Yanshan University, Qinhuangdao, Hebei 066004; 3. Hebei Key Laboratory of Special Delivery Equipment, Yanshan University, Qinhuangdao, Hebei 066004; 4. Key Laboratory of Advanced Forging & Stamping Technology and Science, Yanshan University, Qinhuangdao, Hebei 066004 |
| |
| Abstract: | The life prediction of electromagnetic directional valve is studied from flow degradation on the failure criterion of pressure drop data. Firstly, decomposing test data with the Modified Ensemble Empirical Mode Decomposition (MEEMD) at multi-scale. Noise signal was denoised and reconstructed on the IMFs component screening rule of Euclidean distance. Then, the characteristic parameters are extracted by time-domain analysis, frequency-domain analysis and time-frequency-domain analysis, fused by Kernel Principal Component Analysis (KPCA). Degradation evaluation indexes of electromagnetic directional valve are constructed after cubic exponential smoothing and fused to train the Adaptive Network-based Fuzzy Inference System (ANFIS). Life of electromagnetic directional valve was predicted by predicting the pressure drop trend of faultless samples. The results show that the difference between the prediction index of ANFIS and the actual index is small, and the prediction result is accurate. |
| |
| Keywords: | electromagnetic directional valve ANFIS life prediction KPCA degradation evaluation index |
|
| 点击此处可从《液压与气动》浏览原始摘要信息 |
|
点击此处可从《液压与气动》下载免费的PDF全文 |
|
