The validation of sensor measurements has become an integral part of the operation and control of modem industrial equipment. Neural network based models can be used to estimate critical sensor values when neighboring sensor measurements are used as inputs. The discrepancy between the measured and predicted sensor values can then be used as an indicator for sensor health. The proposed winner take all experts network is based on a 'divide and conquer' strategy. It employs a growing fuzzy clustering algorithm to divide a complicated problem into a series of simpler sub-problems and assigns an expert to each of them locally. After the sensor approximation, the outputs from the estimator and the real sensor value are compared both in the time domain and frequency domain. Three fault indicators are used to provide analytical redundancy to detect the sensor failure. In the decision stage, the intersection of three fuzzy sets accomplishes a decision level fusion, which indicates the confidence level of the sensor health. The simulation results show the proposed system is competitive with or even superior to the existing approaches.
[1]
Mario Innocenti,et al.
Sensor validation using hardware-based on-line learning neural networks
,
1998
.
[2]
Sheung Kai Yung,et al.
Signal processing in local sensor validation
,
1992
.
[3]
Vladimir Vapnik,et al.
Chervonenkis: On the uniform convergence of relative frequencies of events to their probabilities
,
1971
.
[4]
G. J. Erickson,et al.
Multisensor data fusion: concepts and principles
,
1991,
[1991] IEEE Pacific Rim Conference on Communications, Computers and Signal Processing Conference Proceedings.
[5]
A. P. Dhawan,et al.
SSME parameter estimation using radial basis function neural networks
,
1994,
Proceedings of 1994 IEEE International Conference on Neural Networks (ICNN'94).