As mentioned in Chap. 1, the real-time machine data is collected, which may be corrupt or inconsistent due to the presence of environmental noise. Therefore, the cleaning of data is required to remove unwanted frequencies as well to reduce the size of data for further analysis. This chapter details the second most important step of fault diagnosis framework, i.e., pre-processing of data. Low-quality data leads to misleading results, therefore, to make a better, robust, and more accurate fault classification model, pre-processing is required. The pre-processing involves filtering, clipping, smoothing, and normalization methods. Further, a graphical representation of the acoustic signal has been introduced. The chapter ends by a briefing of the development of pre-processing tool.
- 2.Verma, N.K., Agarwal, A., Sevakula, R.K., Prakash, D., Salour, A.: Improvements in preprocessing for machine fault diagnosis. In: IEEE 8th International Conference on Industrial and Information Systems, Kandy, Sri Lanka, pp. 403–408 (2013)Google Scholar
- 4.Verma, N.K., Singh, S., Gupta, J.K., Sevakula, R.K., Dixit, S., Salour, A.: Smartphone application for fault recognition. In: 6th International Conference on Sensing Technology, Kolkata, India, pp. 1–6 (2012)Google Scholar
- 5.Verma, N.K., Singh, J.V., Gupta, M., Sevakula, R.K., Dixit, S.: Windows mobile and tablet app for acoustic signature machine health monitoring. In: International Conference on Industrial and Information Systems, Gwalior, India, pp. 1–6 (2014)Google Scholar