The harmonic reducer is a key component of the industrial robot. Its reliability has significant influence on the consecutive operation of the industrial robot. However, its failure rate is high due to the complicated internal structure and long-term working conditions of large load and high torque. To identify the fault type of the harmonic reducer under different working conditions, this article proposes the joint wavelet regional correlation threshold denoising (WRCTD) algorithm and convolutional neural network-long short term memory (CNN-LSTM) fault detection method. Firstly, the WRCTD algorithm utilizes the regional correlation of the wavelet decomposition coefficients and the $3\sigma $ criterion to suppress noise in the raw sensor data. Then, the CNN-LSTM model could mine hidden features of the processed sensor data to identify fault correctly. To evaluate the proposed method, the vibration signals of the harmonic reducer under different working conditions are collected by establishing a test rig. Comparative experimental results show that the proposed WRCTD algorithm can significantly improve the accuracy of the fault detection method. The specific values for the CNN-LSTM method and the LSTM method are 9.5% and 12.2%, respectively. In addition, the proposed CNN-LSTM method has better performance than traditional methods. It achieves the fault detection over 94.0%, which is 1.5%, 2.0%, 2.0%, 1.9%, and, 1.9% higher than the highest accuracy values of CNN, LSTM, and SVM methods in five experiments.

Prognostics and Health Management (PHM) is being adopted more and more in the modern engineering systems. As one of the most important technologies in the PHM domain, Remaining Useful Life (RUL) prediction has attracted much attention from the researchers in the scholar and industrial field. Although many methods have been proposed to improve the prediction result, the problem of sensor anomaly detection and data recovery has not been considered together. To achieve this object, the datadriven RUL prediction framework considering sensor anomaly detection and data recovery is proposed, which is expected to improve the performance of RUL prediction caused by sensor anomaly. The selected sensor data are firstly detected to decide whether they are anomalous. If the data of this selected sensor are normal, they are continuously adopted as the input of RUL prediction algorithm. But, if the data are anomalous, they will be recovered by the related algorithm. The recovered data will be utilized as the input of RUL prediction algorithm. In the proposed framework, mutual information, Kernel Principal Component Analysis (KPCA), Least Square - Support Vector Machine (LS-SVM), and Gaussian Process Regression (GPR) are utilized. Both simulation data and practical data are used to evaluate the performance of the proposed method. Compared with abandoning the anomalous sensor data, the recovered data can indeed help to enhance the RUL prediction result.

Planetary gearbox is applied more and more extensively in the modern industry. Its condition directly determines the availability and the operation safety of the machinery system that utilizes it for power transmission. To enhance its reliability, an early fault detection method is proposed in this study, which is based on the acoustic emission (AE) technique and the improved variational mode decomposition (VMD). AE data are expected to contain more valuable information about the early fault than the vibration data. Mutual information is used to determine its layer that can avoid over-decomposition or under-decomposition. Then, energy entropy is used to analyze intrinsic mode functions extracted by the improved VMD. By comparing values of energy entropy, it is easy to distinguish different working conditions. The main contributions of this study include: 1) An improved VMD method for fault detection of planetary gearbox is proposed. 2) This method provides a novel strategy for detecting early fault in different positions of planetary gearbox by utilizing the advantages of AE technique.

The aircraft auxiliary power unit (APU) is mainly used to provide electricity and compressed air to the aircraft. Not only can it help to start main engines, but also it can provide essential thrust for the emergency landing. It is required to be reliable as high as possible when it is installed in the aircraft. However, the reality is that the degradation of APU is a nonlinear process. In this term, only a data-driven method or a physics-based method can hardly make accurate prediction for the remaining useful life (RUL) of the aircraft APU. Therefore, a hybrid RUL prediction method is proposed by fusing an artificial intelligence-based model and a physics-based model in this article. To be specific, Wiener process is used to formulate a physical model for generating degradation data. Then, the generated data and the corrected data are both utilized as the input of long short-term memory (LSTM) to enhance RUL prediction of the aircraft APU. The contributions of this study include: 1) A hybrid method for RUL prediction of the aircraft APU is proposed. 2) The influences of ambient environment and training data dependency of LSTM are solved. 3) The proposed method provides a novel RUL prediction method for the on-wing aircraft APU, which is evaluated by the real data.

Fault detection for sensors of unmanned aerial vehicles is essential for ensuring flight security, in which the flight control system conducts real-time control for the vehicles relying on the sensing information from sensors, and erroneous sensor data will lead to false flight control commands, causing undesirable consequences. However, because of the scarcity of faulty instances, it still remains a challenging issue for flight sensor fault detection. The one-class support vector machine approach is a favorable classifier without negative samples, however, it is sensitive to outliers that deviate from the center and lacks a mechanism for coping with them. The compactness of its decision boundary is influenced, leading to the degradation of detection rate. To deal with this issue, an optimized one-class support vector machine approach regulated by local density is proposed in this paper, which regulates the tolerance extents of its decision boundary to the outliers according to their extent of abnormality indicated by their local densities. The application scope of the local density theory is narrowed to keep the internal instances unchanged and a rule for assigning the outliers continuous density coefficients is raised. Simulation results on a real flight control system model have proved its effectiveness and superiority.

In a complex system, condition monitoring (CM) can collect the system working status. The condition is mainly sensed by the pre-deployed sensors in/on the system. Most existing works study how to utilize the condition information to predict the upcoming anomalies, faults, or failures. There is also some research which focuses on the faults or anomalies of the sensing element (i.e., sensor) to enhance the system reliability. However, existing approaches ignore the correlation between sensor selecting strategy and data anomaly detection, which can also improve the system reliability. To address this issue, we study a new scheme which includes sensor selection strategy and data anomaly detection by utilizing information theory and Gaussian Process Regression (GPR). The sensors that are more appropriate for the system CM are first selected. Then, mutual information is utilized to weight the correlation among different sensors. The anomaly detection is carried out by using the correlation of sensor data. The sensor data sets that are utilized to carry out the evaluation are provided by National Aeronautics and Space Administration (NASA) Ames Research Center and have been used as Prognostics and Health Management (PHM) challenge data in 2008. By comparing the two different sensor selection strategies, the effectiveness of selection method on data anomaly detection is proved.