Accurate prediction of aero-engine remaining useful life (RUL) is essential for providing reliable maintenance or alarm decisions. The extraction of degraded features has a great impact on the accuracy of RUL predicti...
Accurate prediction of aero-engine remaining useful life (RUL) is essential for providing reliable maintenance or alarm decisions. The extraction of degraded features has a great impact on the accuracy of RUL prediction. This paper proposes a feature fusion framework that relies on a multi-dimensional convolutional neural network (MD-CNN). First, the data of each working condition is normalized separately to extract degradation features more effectively. Subsequently, the temporal features are extracted using the one-dimensional convolutional neural network (1D-CNN), while the spatial local features are captured through the utilization of the two-dimensional convolutional neural network (2D-CNN). Finally, the long short-term memory network (LSTM) is utilized as the prediction model to establish the mapping relationship from fusion features to RUL. The result shows that the performance in predicting using the proposed method surpasses that of the existing methods.
In the future, large-scale integration of new energy into the grid is an inevitable trend. In order to effectively analyze the trend and random characteristics of wind power output, a wind power scenario simulation me...
详细信息
Network technology has become an inseparable part of daily life, which is easily attacked by some criminals. Therefore, ensuring the security of network becomes an urgent problem. In the past few years, various deep l...
Network technology has become an inseparable part of daily life, which is easily attacked by some criminals. Therefore, ensuring the security of network becomes an urgent problem. In the past few years, various deep learning techniques have been put forward for identifying network intrusions. However, the majority of these approaches overlook the non-Euclidean nature of relationships within network intrusion data. In this paper, graph neural network (GNN) is used to learn the relationship between data. The sample balancing module is to address the significant disparity between the number of intrusive samples and normal samples. The results obtained from the experiments indicate that the suggested approach can enhance the effectiveness of the intrusion detection system.
To address the problems of poor physical interpretability and huge sample size requirement when using neural networks to fit nonlinear control system models for state prediction, this paper proposes a model predictive...
To address the problems of poor physical interpretability and huge sample size requirement when using neural networks to fit nonlinear control system models for state prediction, this paper proposes a model predictive control algorithm based on a physics-informed long short-term memory(LSTM) network. Firstly, the neural network incorporating physical information is extended to model the ordinary differential equations with variable initial states and external control quantities, which makes the network adaptable to the control task and makes the training model physically interpretable. Secondly, a network structure with a mixture of fully connected layers and LSTM layers is built by using the good learning ability of LSTM for time-series data, and the loss function is designed according to the system characteristics and prediction requirements. The trained neural network model is then used as an internal prediction model to construct a nonlinear model predictive control algorithm. Finally, taking the continuous stirring reactor system as an example, the method is verified to be able to fit the system model highly and reduce the time to reach the steady state with a small number of samples.
Artificial neural networks (ANN) have been shown to be flexible and effective function estimators for identification of nonlinear state-space models. However, if the resulting models are used directly for nonlinear mo...
详细信息
With the rapid development of space technology, the demand for satellite reliability is getting higher and higher. Momentum wheel is a key component of satellite attitude control system, and its reliability is an impo...
详细信息
With the rapid development of space technology, the demand for satellite reliability is getting higher and higher. Momentum wheel is a key component of satellite attitude control system, and its reliability is an important factor affecting the life of satellite. The condition monitoring of momentum wheel bearing (MWB) is of great significance to ensure the long life and high reliable operation of the satellite. In this paper, a new monitoring method based on multivariate statistics and canonical variable analysis (CVA) is proposed, and a new health degree function is defined from both dynamic and static aspects. First, the time-frequency domain analysis technique is used to extract the features of MWB in time domain, frequency domain and time-frequency domain, and the multi-domain high-dimensional health condition feature set is constructed. Then, in order to reduce the complexity of the problem, feature reduction is realized based on CVA. On the basis of considering steady-state error and sliding interval variance (SIV), the health degree (HD) characterizing the performance condition of MWB is defined. Finally, the experimental results based on the bearing test-bed show that the proposed method is feasible and effective, and the data-driven health condition monitoring of satellite momentum wheel bearing is realized.
A challenge in Terahertz (THz) thickness measurement of thermal barrier coatings (TBCs) is the change of refractive index due to the uneven microstructure of topcoat, which may degrade the accuracy of the time-of-flig...
详细信息
In order to solve the problem of wind and light abandonment in new energy power generation system, a scheduling strategy of wind, light and fire combined power generation system including concentrating solar power (CS...
详细信息
In this paper, a wind-solar combined power generation system is proposed in order to solve the absorption problem of new energy power generation. Based on the existing installed capacity of local wind power, a concent...
详细信息
We introduce a method for measuring three degrees of freedom using low-coherence light and temporal phase-shifting to discern position and orientation from the coherence envelope shape, enabling direct measurement wit...
详细信息
暂无评论