Research on spatial-temporal synergistic sensor fault diagnosis method for top-blowing furnace.

Top-blowing furnace systems, characterized by a large number of sensors and harsh working environments, are prone to sensor failures due to factors like component aging and external interference. These failures can significantly impact the system's safe and reliable operation. However, traditional sensor fault diagnosis methods often neglect the exploration of spatial-temporal characteristics and focus solely on learning temporal relationships between sensors, failing to effectively consider their spatial relationships. In this study, we propose a spatial correlation model based on the maximal information-based graph convolutional network (MI-GCN) by constructing a sensor network knowledge graph using maximal mutual information. The MI-GCN leverages the graph convolution mechanism to extract multi-scale spatial features and capture the spatial relationships between sensors. Additionally, we develop a spatial-temporal graph-level prediction model, known as the spatial-temporal graph transformer (STGT), to extract temporal features. By combining the spatial features extracted by the MI-GCN with the temporal features captured by the STGT, accurate predictions can be achieved. Sensor fault diagnosis is conducted by analysing the normalized residuals between the predicted values and the ground truth. Finally, the feasibility and effectiveness of the proposed method are validated using test data from a top-blowing furnace system in the nickel smelting process.

Research on Mechanical Equipment Fault Diagnosis Method Based on Deep Learning and Information Fusion.

Solving the problem of the transmission of mechanical equipment is complicated, and the interconnection between equipment components in a complex industrial environment can easily lead to faults. A multi-scale-sensor information fusion method is proposed, overcoming the shortcomings of fault diagnosis methods based on the analysis of one signal, in terms of diagnosis accuracy and efficiency. First, different sizes of convolution kernels are applied to extract multi-scale features from original signals using a multi-scale one-dimensional convolutional neural network (1DCNN); this not only improves the learning ability of the features but also enables the fine characterization of the features. Then, using Dempster-Shafer (DS) evidence theory, improved by multi-sensor information fusion strategy, the feature signals extracted by the multi-scale 1DCNN are fused to realize the fault detection and location. Finally, the experimental results of fault detection on a flash furnace show that the accuracy of the proposed method is more than 99.65% and has better fault diagnosis, which proves the feasibility and effectiveness of the proposed method.

Incipient fault diagnosis and amplitude estimation based on K-L divergence with a Gaussian mixture model.

In this paper, a method of incipient fault diagnosis and amplitude estimation based on Kullback-Leibler (K-L) divergence is proposed. An incipient fault is usually regarded as the precursor of a significant system fault, but due to a low amplitude and non-obvious characteristics, it is easy for such a fault to be hidden by disturbance and noise. Based on this and considering the sensitivity of the K-L divergence method in data feature extraction, a method of diagnosing incipient faults is designed. In order to consider the safety performance and lay a foundation for the fault tolerance of the system, an amplitude estimation method for incipient faults is also proposed. By mapping the characteristic change in the residual data to the numerical change in the K-L divergence, the amplitude of the incipient fault can be measured with high sensitivity. Considering the generality of the method, a Gaussian mixture model is used to model the residual data in order to increase the accuracy of fault amplitude estimation. Finally, the effectiveness of the proposed method for incipient fault diagnosis and amplitude estimation is verified by experiment.

Permissible Area Analyses of Measurement Errors with Required Fault Diagnosability Performance.

Fault diagnosability is the basis of fault diagnosis. Fault diagnosability evaluation refers to whether there is enough measurable information in the system to support the rapid and reliable detection of a fault. However, due to unavoidable measurement errors in a system, a quantitative evaluation index of system fault diagnosability is inadequate. In order to overcome the adverse effects of measurement errors, improve the accuracy of the quantitative evaluation of fault diagnosability, and improve the safety level of the system, a method for a permissible area analysis of measurement errors for a quantitative evaluation of fault diagnosability is proposed in this paper. Firstly, in order for the residuals obey normal distribution, a design method of the permissible area of measurement errors based on the Kullback-Leibler divergence (KLD) is given. Secondly, two key problems in calculating the KLD are solved by sparse kernel density estimation and the Monte Carlo method. Finally, the feasibility and validity of the method are analyzed through a case study.