This paper proposes a method for analog fault diagnosis using neural networks. The primary focus of the paper is to provide robust diagnosis using a mechanism to deal with the problem of component tolerances and reduce testing time. The proposed approach is based on the k-fault diagnosis method and artificial backward propagation neural network. Simulation results show that the method is robust...

Pattern classifications have become important tools for fault diagnosis in nuclear power plants (NPP). However, it is often difficult to obtain training data under fault conditions to train a supervised classification model. By contrast, normal plant operating data can be easily made available through increased deployment of supervisory, control, and data acquisition systems. Such data can also...

Fault diagnosis depends largely on feature analysis of vibration signals. However, feature extraction for fault diagnosis is difficult because the vibration signals often contain a strong noise component. Noises stronger than the actual fault signal may interfere with diagnosis and ultimately cause misdiagnosis. In order to extract the feature from a fault signal highly contaminated by the nois...

DGA (Dissolved Gas Analysis) is the traditional transformer fault diagnosis method, but it mainly depends on the experience of operators. In order to solve the limitations of traditional method, this paper introduces intelligent method for fault diagnosis of transformer. The intelligent method made fusion of various data, including SCADA data, oil dissolved gas sensor data, related electrical t...

In this paper we present three new schemes for sensor and actuator fault diagnosis for linear time-invariant systems: multirate parity-space techniques, multirate observers, and multirate detection filters. We show that these approaches offer distinct advantages over conventional fault diagnosis schemes in terms of ease of failure detection, the speed and the class of systems for which detectio...

Historical data collected from processes are readily available. This paper looks at recent advances in the use of data-driven models built from such historical data for monitoring, fault diagnosis, optimization and control. Latent variable models are used because they provide reduced dimensional models for high dimensional processes. They also provide unique, interpretable and causal models, al...

The quality of surface mount technology (SMT) product solder joint was key factor for SMT products. In this paper, for chip components, a fuzzy fault diagnosis principle of SMT products was analyzed. And the total framework of fault diagnosis system was built based on solder joint shape theory. Geometrical shape parameters which reflected chip components solder joint quality were analyzed and d...

Aiming at the PVC production process is complex, the large critical devices polymerizer running need to constantly monitor the characteristics, performance monitoring and fault diagnosis polymerizer for the large PVC batch production process. First of all, for the lack of standard LMBP algorithm, the LMBP neural network algorithm is improved; Secondly, based on the genetic algorithm (GA) and im...

The monitoring of turbine in operation condition and fault diagnosis system in power plant is the key to guarantee the units long-term security and economic operation. The turbine faults were determined by a variety of factors, which brought many difficulties to the fault diagnosis. In general condition, using the RBF neural network can make the turbine fault diagnosis right. However, in some t...