Research of synthesized fault diagnose for flat wheel detecting system

Fault diagnosis (FD) has recently received considerable attention in industry and academia. In order to make a more accurate and more complete estimation and decision for the problem, an approach that is capable of overcoming the disadvantage of single FD methodology, and eliminating the system uncertainty is presented. This paper takes Flat Wheel Detecting System as an example. First, the proposed method diagnoses the example system with model based parameter estimation and frequency spectrum analysis respectively according to groups of data, and the estimated parameters and power spectrum are taken as the index to make decision. Then, it makes final decision according to data fusion and statistic reasoning methods. The results show that the methodology is capable of making the right diagnosis according to experimental data.