Abstract

Smart industrial processes are impossible without the use of machinery. Majority of these machines comprise rotating machines. As supporting tools for developing maintenance strategies, a Data-driven approach for multifault diagnosis of Rotating Machines is currently used in smart industry. This Chapter presents Multisensor Data Fusion and its application in Predictive maintenance methodology for multifault diagnosis of Industrial Rotating Machines. The main objective of Multisensor Data Fusion, rather than the different sensors being applied independently, are optimally fused to take advantage of their respective strengths, by combining data from multiple sensors and related information to achieve specific inferences than could be achieved by a single sensor. This chapter presents a methodology using data extraction features domain parameters for multifault diagnosis of Rotating Machines. The analysis shows that Root Mean Square Amplitude increases as fault develops, Crest Factor is more for healthy bearing and less for faulty bearing, these can provide early warning of faults