A Novel Sensor Fault Detection Method

Chen Yongqi (Ningbo University)
Zhao Yiming (Ningbo University)

Indonesian Journal of Electrical Engineering and Computer Science
Vol 11, No 12: December 2013

To detect the sensor faults of the plant with min-disturbance, a weighted least squares interval regression model is proposed. The output of the proposed model is an interval band which can resist the disturbance influence and give correct sensor fault alarm. Additionally, the time complexity of this model is low because only a set of linear equations can determine the parameters. The experiments of fault instance demonstrate the feasibility and effectiveness of the interval regression model. DOI: http://dx.doi.org/10.11591/telkomnika.v11i12.3684