原文题目：
作者：
潘俊林，孙博，吴泽豫，易泽辰，冯强，任羿，王自力
原文链接：
期刊信息：
创新点
(1) 所提出的融合贝叶斯深度学习和随机过程的方法能够在无寿命标签的情况下进行 RUL 概率预测，避免额外实验成本。
(2) 基于无模型BNN的方法实现了对认识不确定性和随机不确定性的综合量化，从而为随机过程模型提供了充足的样本输入。
(3)所提出的方法不需要在线检测HI用于随机过程建模，并根据在线参数更新方法进一步实现了端到端的长期RUL预测。
摘要
可信赖的剩余寿命（RUL）预测对于退化系统的长期安全可靠运行至关重要。现有的基于深度学习的RUL预测方法越来越受到关注，但通常面临三大挑战：一是难以获取完整的运行到失效数据，这意味着缺乏寿命标签；二是难以直接测量现场退化系统的健康状态指标（HI）；三是预测模型输点估计值而没有不确定性。为此，本文提出了一种融合贝叶斯深度学习与随机过程的方法，用于无寿命标签下的RUL概率预测。首先，本文构建了一种无模型贝叶斯神经网络，以任意表征形式同时对深度学习中的认识不确定性和不确定性进行量化。基于构建的贝叶斯神经网络，可以预测HI的概率特征。然后，利用非线性维纳过程进行退化建模，推导RUL的概率密度函数。此外，在在线运行过程中通过参数直接更新方法实现模型演化。最后，在CALCE电池退化数据上验证了所提预测方法的有效性和准确性。
Abstract
Trustworthy remaining useful life (RUL) predictions are critical for the long-term safe and reliable operation of degradation systems. The existing deep learning-based methods for RUL prediction are attracting increasing attention but typically face three main challenges. One is the absence of complete run-to-failure data, implying a lack of lifetime labels. Second, it is difficult to directly measure the health indicators (HIs) for field degradation systems. Third, the prediction models output point estimates without uncertainty. To this end, this paper proposes a Bayesian deep learning and stochastic process fusion method for probabilistic RUL prediction without lifetime labels. First, a model-free Bayesian neural network (BNN) is constructed to integrate the quantification of epistemic and aleatoric uncertainties in deep learning. Based on the constructed BNN, it is possible to predict the probability features of HIs. Then, degeneracy modeling is conducted using a nonlinear Wiener process to derive the probability density function of the RUL. Furthermore, model evolution can be achieved through parameter updating during online operations. Finally, the effectiveness and superiority of the proposed prediction method are verified on CALCE battery degradation data.
图1 RUL概率预测框架
图2 CALCE电池数据集的六组实验条件
图3 CALCE电池数据集的容量下降趋势
图4 (a)-(f)分别代表由CS2-33,CS2-34,CS2-35,CS2-36,CS2-37和CS2-38获得的概率预测结果
图5 (a)-(d)分别表示M1、M2、M3和M4的预测结果
图6 (a)不同监测时间的预测结果比较。(b)第650个周期模型演化结果比较
推荐阅读：
用于预测和健康管理的类ChatGPT大型基础模型：综述和路线图
