Efficiency Of Fault Diagnosis Research Articles

Comparison of SOM and PCA-SOM in Fault Diagnosis of Ground-testing Bed

In order to compare the fault diagnosis reliability and fault diagnosis efficiency (time consumption property) of PCA-SOM and SOM in liquid propellant rocket engine ground-testing bed, two types of ground-testing bed fault data are used. One is generated by the mechanism model of ground-testing bed. The other is generated according to the expert's experience and the statistical model of fault mode. The comparison results using these two types of fault data both indicate that the fault diagnosis reliability and fault diagnosis efficiency of PCA-SOM are better than SOM.

Transformer Fault Dignosis Based on Feature Selection and Parameter Optimization

Failure of transformer is very complex, dissolved Gas in Oil Analysis (DGA) is presently the easier and simpler way for fault diagnosis of oil-immersed transformers.The correct selection of features of dissolved gas data can improve efficiency of transformer fault diagnosis. SVM is more effective than traditional methematic model to discribe the type of fault of transformer.As for the problem of difficulty of determining parameters in SVM applications, genetic algorithm (GA) was used to select SVM parameters. The test results show that this GA-SVM model is effective to detect failure of transformer.

Batch Process Monitoring and Fault Diagnosis Based on Multi-Time-Scale Dynamic PCA Models

Dynamics are inherent characteristics of batch processes, which can be divided into short time-scale dynamics within a batch duration and long time-scale dynamics across several batches. The interactions between process variables make different types of dynamics confounded. Under such situations, it is difficult to perform efficient fault diagnosis. In this paper, a batch process monitoring scheme is proposed to separate different types of process variations for modeling and perform monitoring and fault diagnosis with multi-time-scale dynamic principal component analysis (PCA) models. Simulation results show that the fault diagnosis efficiency is enhanced.

Research of Information System of Monitoring Remote on Distributed Micro-turbine

Aiming at the problems existing in fault diagnosis of Distributed micro gas turbine, a kind of new information system frame of fault Diagnosis on Distributed micro gas turbine is proposed in this paper. The consumers and the manufactures contact with each other through Internet, share the corresponding data of the Distributed Micro-turbine, and diagnose and solve the faults together. Through the study of collecting method of the micro gas turbine generating sets’ vibration signals, the technique of gathering reliable vibration signals is designed. The traditional method is revised by this new system, and this can greatly improve the efficiency of Fault Diagnosis, reduce the cost of equipment maintenance, and be in favor of the accumulation and updating of techniques in this field.

Generalized robust H∞ fault diagnosis filtering based on conditional stochastic distributions of system outputs

The efficiency of fault detection and diagnosis by using output probability density functions (PDFs) for stochastic time-delayed systems has been shown in practical processes. Neural network modelling has been applied to characterize the output PDFs and to the dynamical weighting system. In this paper, the system perturbations and disturbance are considered and the robust fault diagnosis design is studied for the general stochastic system in the presence of time delays. The main objective is to design linear matrix inequality (LMI)-based fault diagnostic filtering (FDF) to estimate the fault and attenuate the disturbances. The modelling errors and system uncertainties resulting from both B-spline expansion and the weighting system are merged into system disturbance. It can be seen that the resulting weighting system comprises non-linearities, uncertainties, disturbances, and time delays, and includes the non-zero initial condition. The generalized H∞ optimization is presented and applied to the fault diagnosis problem of the weighting system with the non-zero initial condition and truncated norms. Simulations are given to demonstrate the efficiency of the proposed approach.

TWO-DIMENSIONAL DYNAMIC PCA WITH AUTO-SELECTED SUPPORT REGION

Support region selection is a key step of two-dimensional dynamic PCA modeling for batch process dynamics, as it can affect the accuracy of the model and the efficiency of monitoring and fault diagnosis. In this paper, an automatic method for support region selection is developed. This data-based method can be applied universally on different batch processes without any prior process knowledge. Simulation shows that developed method has good application potentials for both monitoring and fault diagnosis.