Neutrosophic Ensemble Classification for Uncertainty-Aware Bearing Fault Detection: Evidence from Laboratory and Variable-Speed Industrial Benchmarks

查看原文 / View Original 编辑 / Edit

Computer Science > Machine Learning arXiv:2610.06880 (cs) [Submitted on 18 Sep 2026] Title: Neutrosophic Ensemble Classification for Uncertainty-Aware Bearing Fault Detection: Evidence from Laboratory and Variable-Speed Industrial Benchmarks Authors: Maikel Leyva-Vazquez, Dayron Rumbaut Rangel, Lorenzo Cevallos-Torres, Alexis Matheu Perez.

计算机科学 > 机器学习 arXiv:2610.06880 (cs) [提交于 2026 年 9 月 18 日] 标题：用于不确定性感知轴承故障检测的中智集成分类：来自实验室和变速工业基准的证据作者：Maikel Leyva-Vazquez, Dayron Rumbaut Rangel, Lorenzo Cevallos-Torres, Alexis Matheu Perez。

Abstract: Machine learning classifiers for bearing fault detection produce scalar confidence scores that conflate confident errors with genuinely ambiguous predictions, and the conventional truth/falsity pair (F = 1 - T) is algebraically redundant by construction. We operationalize a refined neutrosophic decomposition of a Random Forest + XGBoost + Logistic Regression ensemble into four indicators — T-hat (top-class evidence), F-hat (best-competitor evidence), predictive entropy I1-hat, and decision disagreement I2-hat — evaluated on two bearing benchmarks (CWRU and JNU, 600-1000 rpm) under a leave-one-condition-out protocol.

摘要：用于轴承故障检测的机器学习分类器产生的标量置信度分数将自信的错误与真正模糊的预测混为一谈，且传统的真/假对（F = 1 - T）在代数结构上是冗余的。我们将随机森林 + XGBoost + 逻辑回归集成的精炼中智分解操作化为四个指标——T-hat（顶级类别证据）、F-hat（最佳竞争者证据）、预测熵 I1-hat 和决策分歧 I2-hat——并在留一条件协议下对两个轴承基准（CWRU 和 JNU，600-1000 rpm）进行了评估。

On CWRU, after correcting a file-to-class mapping error, the ensemble reaches 100.00 percent accuracy on three of four held-out loads (92.27 percent on the fourth), leaving too few errors for uncertainty analysis. On JNU, holding out 1000 rpm, accuracy collapses to 40.64 percent, below a majority-class baseline; Logistic Regression (57.91 percent) generalizes far better than the tree ensembles.

在 CWRU 上，纠正文件到类别的映射错误后，该集成在四个留出负载中的三个上达到了 100.00% 的准确率（第四个为 92.27%），留下的错误太少，无法进行不确定性分析。在 JNU 上，留出 1000 rpm 时，准确率下降至 40.64%，低于多数类基准；逻辑回归（57.91%）的泛化能力远好于树集成。

I1-hat shows a robust association with error beyond T-hat/F-hat, while I2-hat contributes little; standalone Logistic Regression confidence outperforms the full decomposition, a boundary condition we report honestly. Two further results extend this: fusing a time-domain and a frequency-domain model of the same signal and scoring their Jensen-Shannon divergence beats that model own entropy (AURC 0.29 vs. 0.36 on the standard split; 0.54 vs. 0.73 under a harder single-condition reproduction), the only indicator moving correctly under a CWRU-versus-JNU distributional-shift contrast; and, on CWRU alone, literature-verified bearing fault frequencies, correctly demodulated via the envelope spectrum, separate most fault classes almost perfectly (99.57 percent) using three interpretable features.

I1-hat 显示出与 T-hat/F-hat 之外的错误存在稳健的关联，而 I2-hat 的贡献很小；独立的逻辑回归置信度优于完整分解，这是一个我们如实报告的边界条件。另外两个结果扩展了这一点：融合同一信号的时域和频域模型并对其 Jensen-Shannon 散度进行评分，其效果优于模型自身的熵（在标准划分上 AURC 为 0.29 对 0.36；在更困难的单条件复现下为 0.54 对 0.73），这是在 CWRU 与 JNU 分布偏移对比下唯一正确变化的指标；此外，仅在 CWRU 上，通过包络谱正确解调的文献验证轴承故障频率，使用三个可解释特征几乎完美地（99.57%）分离了大多数故障类别。