This paper explores the development of a unified hybrid approach for sensor and actuator fault diagnosis in digital twins for remote operations. Central to this approach is the implementation of a robust adaptive Kalman filter algorithm, which forms the backbone of the proposed unified algorithm. The essence of this unified algorithm lies in its capability to effectively filter the sensor measurements. The algorithm is enriched with tuning parameters, offering flexibility in adjusting the convergence rate to suit operational requirements. Noteworthy for its robustness, our approach excels in handling uncertainties and diverse types of faults, including drift, bias, noise, and freeze fault scenarios. Through comprehensive simulation and experimental evaluations conducted on a small surface vessel, the method demonstrates remarkable proficiency in accurately identifying sensor and actuator faults. This precision enables early detection and prompt mitigation of anomalies, contributing to heightened operational resilience.

中文翻译：

---

用于远程操作的数字孪生中的统一传感器和执行器故障诊断


本文探讨了用于远程操作的数字孪生中传感器和执行器故障诊断的统一混合方法的开发。该方法的核心是鲁棒自适应卡尔曼滤波器算法的实现，该算法构成了所提出的统一算法的支柱。这种统一算法的本质在于其能够有效过滤传感器测量结果。该算法具有丰富的调整参数，可以灵活地调整收敛速度以满足操作要求。我们的方法以其稳健性而值得注意，它擅长处理不确定性和不同类型的故障，包括漂移、偏差、噪声和冻结故障场景。通过在小型水面舰艇上进行的综合仿真和实验评估，该方法在准确识别传感器和执行器故障方面表现出卓越的能力。这种精度能够及早发现并迅速缓解异常情况，从而有助于提高运营弹性。