Sensor Fault Detection Based on Weighted Stochastic Gradient Identification Algorithms

This paper addresses the problem of sensor fault detection for practical systems based on stochastic gradient identification techniques. Three algorithms are constructed by utilizing the hierarchical identification principle. Different from some existing algorithms, a weighted factor was introduced and a combination of the information in both the last step and the current step can be used to update the estimation of the variables. Due to the use of the latest updated information, the proposed algorithms can achieve better convergence performance than some existing algorithms by appropriately choosing the tuning parameter. A numerical example is given to show the superiority of the presented algorithms.

[1]  Yang Jiaben.  Convergence analysis of stochastic gradient algorithms , 1999 .

[2]  A. Wu,et al.  Latest estimation based recursive stochastic gradient identification algorithms for ARX models , 2015, 2015 34th Chinese Control Conference (CCC).

[3]  Li Ya-peng.  Survey of Fault Diagnosis Technology , 2009 .

[4]  Panfeng Huang,et al.  The convergence of stochastic gradient iterative modal identification algorithm and application in modal analysis for space large manipulator , 2014, 2014 IEEE International Conference on Robotics and Biomimetics (ROBIO 2014).

[5]  Liam Maguire,et al.  Fault diagnosis of electronic system using artificial intelligence , 2002 .

[6]  Rui-Qi Dong,et al.  Convergence Analysis of Weighted Stochastic Gradient Identification Algorithms Based on Latest‐Estimation for ARX Models , 2019 .

[7]  Ai-Guo Wu,et al.  Weighted Stochastic Gradient Identification Algorithms for ARX models , 2015 .

[8]  Seungdeog Choi,et al.  Implementation of a Fault-Diagnosis Algorithm for Induction Machines Based on Advanced Digital-Signal-Processing Techniques , 2011, IEEE Transactions on Industrial Electronics.

[9]  Dinghui Wu,et al.  Fault diagnosis of variable pitch for wind turbines based on the multi-innovation forgetting gradient identification algorithm , 2015 .

[10]  B. Nguyen,et al.  Parameter Estimation in Analytical Models of Automotive Vehicles and Fault Diagnosis , 1995 .

[11]  Guo Lili,et al.  Artificial intelligence and learning techniques in intelligent fault diagnosis , 2015, 2015 4th International Conference on Computer Science and Network Technology (ICCSNT).