AI 中文总结
该研究针对RUL预测中线性标签与振动退化不一致问题，提出阶段感知目标与双尺度预测器，在XJTU-SY和IMS数据集上验证了其对目标拟合及预测精度的提升效果，建立了面向测量的目标有效性框架。
AI 中文摘要
剩余使用寿命（RUL）研究通常将标签视为固定值，但时钟线性标签可能在测量振动几乎稳定时下降，随后在故障附近快速变化。我们将目标设计与预测分离：仅开发的流程构建定向振动健康指标，识别时间上的早期、中期和晚期阶段，并拟合连续线性-二次-指数退化状态目标；紧凑的CNN-LSTM和Transformer从因果特征序列中学习该目标，经验证拟合的有序加权平均（OWA）融合二者输出。在轴承级XJTU-SY留一测试中，所有编号以5结尾的轴承被排除在拟合预处理、训练、早停及融合流程外。融合预测器的均方根误差（RMSE）为0.0608，平均绝对误差（MAE）为0.0392，决定系数（R-squared）为0.9617，其中Transformer贡献了大部分精度。在三个已记录的IMS故障轴承轨迹上独立评估目标形状：与针对同一振动衍生参考的最优锚定线性拟合相比，阶段感知曲线的RMSE降低3.6%-18.2%、MAE降低3.1%-31.1%，平均降幅分别为10.2%和15.0%；保守贝叶斯信息准则（BIC）差值为128.8-368.1，支持阶段感知表示，而移动块自举区间包含零点。因此，阶段依赖目标能更好描述所评估的振动衍生退化状态，但证据仍具描述性，因为仅可获得三次官方IMS运行数据。本研究建立了面向测量的目标有效性框架，而非适用于物理故障时间或鲁棒跨域预测的通用非线性规律。
英文摘要
Remaining useful life (RUL) studies commonly treat the label as fixed, although clock-linear labels may decline while measured vibration remains nearly stable and then changes rapidly near failure. We separate target design from prediction. A development-only pipeline constructs an oriented vibration health indicator, identifies chronological early, middle, and late stages, and fits a continuous linear-quadratic-exponential degradation-state target. A compact CNN-LSTM and Transformer learn the target from causal feature sequences, and validation-fitted Ordered Weighted Averaging combines their outputs. In a bearing-wise XJTU-SY hold-out, all bearings ending in 5 are excluded from fitted preprocessing, training, early stopping, and fusion. The fused predictor obtains an RMSE of 0.0608, an MAE of 0.0392, and an R-squared value of 0.9617, with the Transformer providing most of the accuracy. Target shape is assessed independently on three documented IMS failed-bearing trajectories. Against the best anchored linear fit to the same vibration-derived reference, the stage-aware curve reduces RMSE by 3.6-18.2% and MAE by 3.1-31.1%; the mean reductions are 10.2% and 15.0%, respectively. Conservative BIC differences of 128.8-368.1 favor the stage-aware representation, whereas moving-block bootstrap intervals cross zero. Thus, stage-dependent targets better describe the evaluated vibration-derived degradation states, but the evidence remains descriptive because only three official IMS runs are available. The study establishes a measurement-oriented target-validity framework, not a universal nonlinear law for physical time-to-failure or robust cross-domain prediction.
Comments28 pages, 12 figures, 13 tables
