To address the problem of dataset imbalance caused by the scarcity of fault samples in the fault diagnosis of bearings in helicopter tail drive systems, we propose the Integrable Physical Information Generative Adversarial Network (IE-PIGAN) model. On the basis of retaining the original adversarial loss and network structure of Generative Adversarial Network (GAN), this model embeds the physical prior knowledge of tail drive fault bearings into the generator loss in the form of integral equations, and combines the data loss of physics-informed neural networks to impose physical constraints on the generation process, thus generating fault samples that take random noise as input and conform to the physical characteristics of bearings. Experimental results show that the signals generated by IE-PIGAN are highly similar to real signals in both time domain and frequency domain, and its percent root mean square error, root mean square error, and frequency-domain correlation coefficient of maximum mean discrepancy are all lower than those of GAN. When the generated data are added to the training set, the diagnostic accuracy of the convolutional neural network reaches 91.7%, which is 8.4% higher than that without generated samples, and the accuracy outperforms five mainstream models such as denoising autoencoders and variational autoencoders. The volume of generated data exhibits a trend of first increasing and then decreasing with respect to the enhancement effect. Compared with other generative models, IE-PIGAN has the slowest accuracy decline rate after the sample volume exceeds the inflection point. This model effectively alleviates the imbalance problem of fault samples for helicopter tail drive bearings, and improves the accuracy and robustness of tail drive bearing fault diagnosis.
[1]陈雪峰, 王诗彬, 程礼.航空发动机快变信号的匹配同步压缩变换研究[J].机械工程学报, 2019, 55(13):13-22
[2]陈国旺, 唐倩, 李恒, 等.直升机尾传动系统响应预测建模与试验研究[J].航空动力学报, 2025, 40(07):67-79
[3]康玉祥, 陈果, 尉询楷, 等.深度残差对冲网络及其在滚 动轴承故障诊断中的应用[J].航空学报, 2022, 43(8):57-68
[4]Wang, T. Zhang, Y. Qin.Qian, Y. Wang, T. Zhang, et al.Wang, T. Zhang, Y. Qin.Qian, Y. Wang, T. Zhang, et al, Maximum mean square discrepancy: a new discrepancy representation metric for mechanical fault transfer diagnosis, Knowl. Based. Syst. 276 (2023) 110748.[J].Knowledge-Based Systems, 2023, 276:110748-[J].Knowledge-Based Systems, 2023, 276:110748-
[5]樊红卫, 严杨, 张旭辉, 等.滚动轴承优选与-故障诊断振动.测试与诊断[J].振动．测试与诊断, 2023, 43(3):593-602
[6]韩淞宇，邵海东，姜洪开，等.基于提升卷积神经网络的航空发动机高速轴承智能故障诊断[J].航空学报, 2022, 43(9):150-163
[7]X. Wang, Y. Qin, Y. Wang, et al.X. Wang, Y. Qin, Y. Wang, et al, ReLTanh: an activation function with vanishing gradient resistance for SAE-based DNNs and its application to rotating machinery fault diagnosis, Neurocomputing. 363 (2019) 88–98.[J].Neurocomputing, 2019, 363:88-98
[8]陈泳益, 蓝积炎, 杨喜, 等.数据不均衡条件下数据增强辅助的自动调制识别[J].华侨大学学报自然科学版, 2026, 47(01):104-111
[9]黄炟, 黄诗浩, 王周润, 等.扩散模型辅助的FPN小样本轴承故障诊断方法[J/OL].机电工程技术, 1-7[2026-01-21].https://link.cnki.net/urlid/44.1522.th.20251121.1652.007.HUANG D, HUANG S H, WANG Z R, et al.Diffusion Model-Assisted FPN Small Sample Bearing Fault Diagnosis Method[J/OL]. MECHANICAL & ELECTRICAL ENGINEERING TECHNOLOGY, 1-7[2026-01-21]. https://link.cnki.net/urlid/44.1522.th. 20251121.1652.007 (in Chinese).
[10]Yi Qin, Hongyu Liu, Yi Wang, et al.Inverse physics–informed neural networks for digital twin–based bearing fault diagnosis under imbalanced samples[J].Knowledge-Based Systems, 2024, 292, DOI:10.1016/j.knosys.2024.111641.[J].Knowledge-Based Systems, 2024, 292:111641-
[11]Sun Mingyue, Wu Fan, Ji Xin, et al.Systematic GAN Parameter Selection for Fault Data Generation using Particle Swarm Optimization[C]//2024 IEEE International Conference on Energy Internet. 2024:580-585.
[12]王英杰, 朱景建, 龚智强, 等.基于--的滚动轴承故障诊断方法[J].机械设计, 2024, 41(11):123-129
[13]张钊光, 蒋庆磊, 詹瑜滨, 等.基于-数据增强算法的小样本滚动轴承故障分类方法[J].原子能科学技术, 2023, 57(S1):228-237
[14]金毓林, 陈长征, 罗园庆, 等.WGAN-SPCNN在滚动轴承故障诊断中的应用[J].机械设计与制造, 2025, (06):96-101.DOI:10.19356/j.cnki.1001-3997.2025.06.006.JIN Y L, CHEN C Z, LUO Y Q, et al.Application of WGAN-SPCNN in Rolling Bearing Fault Diagnosis[J]. Machinery Design & Manufacture, 2025, (06):96-101.DOI:10.19356/j.cnki.1001-3997.2025.06.006 (in Chinese).[J].机械设计与制造, 2025, 412(6):96-101
[15]张秋生, 吴志刚, 张文亮, 等.基于-的轴承碰摩故障识别方法研究[J].信息化研究, 2025, 51(01):48-56
[16]Li, Chengcheng, Qin, Yi, Wang, Yi, et al.Vibration Analysis of Deep Groove Ball Bearings With Local Defect Using a New Displacement Excitation Function[J].Journal of Tribology, 2020, 142,(12):121202-121202-13
[17]赵筛筛, 陈剑飞, 何怡刚, 等.基于轻量化CNN与PINN的四电平变流器开路故障在线诊断方法[J/OL].高电压技术, 1-16[2026-01-21].https://doi.org/10.13336/j.1003-6520.hve.20250269.ZHAO S S, CHEN J F, HE Y G, et al.Online Diagnosis Method of Open-circuit Fault for Four-level Converter Based on Lightweight CNN and PINN[J/OL]. High Voltage Engineering, 1-16[2026-01-21].https://doi.org/10.13336/j.1003-6520.hve.20250269 (in Chinese).[J].高电压技术, 2025, :1-16
[18]Shi X, Yu Y.Design and application of art creation education system based on generative adversarial network[J].Discover Artificial Intelligence, 2025, 6(1):24-24
[19]Yang C, CHEN H Y, Shan M, et al.Improved physics-informed neural networks incorporating lattice Boltzmann method optimized by tanh robust weight initialization[J].Chinese Physics B, 2025, 34(11):110701-110701
[20]Huang D, Wang P, Li W, et al.Modeling,response and BPNN-PID control of symmetric multistable galloping energy harvesters[J].Chaos, Solitons and Fractals: the interdisciplinary journal of Nonlinear Science, and Nonequilibrium and Complex Phenomena, 2025, 199(P2):116736-116736
[21]Yu D, Min Y, Yang J, et al.Dense synergistic attention network: An effective CNN model for communication facilities image classification[J].Engineering Applications of Artificial Intelligence, 2026, 165(PA):113452-113452
[22]刘强强, 谷艳玲, 张品杨.基于及退化趋势相似性分析的轴承剩余使用寿命预测[J].机电工程, 2024, 41(05):853-861
[23]易怀胜, 李少义, 陈汉新, 等.一种多特征融合的-齿轮箱故障诊断方法[J].噪声与振动控制, 2025, 45(06):162-168
[24]方绵绵.基于变分模态分解的风电齿轮箱轴承故障诊断方法[J].工程与试验, 2021, 61(01):4-11
CJA