Improved Fault Detection Model
This chapter details about the improvement in different modules of the fault diagnosis framework. During data acquisition, the data is acquired in cyclic order instead of capturing in one shot. For feature extraction purpose, a new set of 343 features were introduced for detailed analysis. The next step is about to improve the feature selection where we proposed a novel feature selection method based on the graphical analysis. This method is based on how nicely a feature can differentiate between two states based on their feature’s plots for multiple recordings. In the classification phase, each feature votes for a class and the final class is decided based on the majority.
- 3.Verma, N.K., Sevakula, R.K., Goel, S.: Study of transforms for their comparison. In: International Conference on Industrial and Information Systems, Gwalior, India, pp. 1–6 (2014)Google Scholar
- 4.Verma, N.K., Agrawal, A.K., Sevakula, R.K., Prakash, D., Salour, A.: Improved signal preprocessing techniques for machine fault diagnosis. In: 2013 IEEE 8th International Conference on Industrial and Information Systems, pp. 403–408 (2013)Google Scholar