Intelligent fault diagnosis and control reconfiguration

1994 ◽  
Vol 14 (3) ◽  
pp. 6-12 ◽  
2012 ◽  
Vol 45 (11) ◽  
pp. 306-311
Author(s):  
Gianni Bertoni ◽  
Paolo Castaldi ◽  
Maria E. Penati

Author(s):  
Chunjian Wang ◽  
Beshah Ayalew ◽  
Zoran Filipi

A Digital-Displacement Pump/Motor (DDPM) has recently been proposed as an attractive candidate for hydraulic powertrain applications. A DDPM uses solenoid-controlled valves for each cylinder. This provision offers flexibility of control that can be exploited to boost system efficiency by matching individual cylinder operations with load conditions. However, the added complexity from individual cylinder control necessitates mechanisms for fault diagnosis and control reconfiguration to ensure reliable operation of the DDPM. Furthermore, available measurements are often limited to supply and return line pressures, shaft angle and speed. In this paper, it is shown that, with only these measurements, individual cylinder faults are structurally unobservable and un-isolable by the use of a system model relating the cylinder faults to the shaft dynamics. To overcome this difficulty, the phase angles at which possible individual cylinder faults can begin to affect the shaft dynamics are tabulated for each cylinder, and a fault indicator that is akin to a shaft acceleration fault is modeled and estimated via a fast sliding mode observer. Simultaneous detection and isolation of individual cylinder faults can be achieved using this fault indicator and a table of fault begin angles. Illustrative examples are included from simulations of a 5 cylinder DDPM to demonstrate this diagnosis process.


Author(s):  
Chun Cheng ◽  
Wei Zou ◽  
Weiping Wang ◽  
Michael Pecht

Deep neural networks (DNNs) have shown potential in intelligent fault diagnosis of rotating machinery. However, traditional DNNs such as the back-propagation neural network are highly sensitive to the initial weights and easily fall into the local optimum, which restricts the feature learning capability and diagnostic performance. To overcome the above problems, a deep sparse filtering network (DSFN) constructed by stacked sparse filtering is developed in this paper and applied to fault diagnosis. The developed DSFN is pre-trained by sparse filtering in an unsupervised way. The back-propagation algorithm is employed to optimize the DSFN after pre-training. Then, the DSFN-based intelligent fault diagnosis method is validated using two experiments. The results show that pre-training with sparse filtering and fine-tuning can help the DSFN search for the optimal network parameters, and the DSFN can learn discriminative features adaptively from rotating machinery datasets. Compared with classical methods, the developed diagnostic method can diagnose rotating machinery faults with higher accuracy using fewer training samples.


Sign in / Sign up

Export Citation Format

Share Document