Muthanna Journal of Engineering and Technology
Volume (11), Issue (2), Year (30 December 2023), Pages (01-08)
DOI:10.52113/3/eng/mjet/2023-11-02/01-08
Research Article By:
Auda Raheemah Odhaib, Abbas Swayeh Atiyah and Mohammed Zuhair Azeez
Corresponding author E-mail: auda@mu.edu.iq
ABSTRACT
The maintenance cost is the main challenge of the industrial environment especially that related to the expansive machine. Many faults caused damage to the machine as the oil flow, pressure, vibration, and temperature. The vibration fault of the rotating machine is producing damage if been in the danger zone. In this work, the vibration fault of the induction motor has been detected and classified based on wireless and artificial intelligence techniques. The C++ code was utilized to design and implement the wireless sensor, while the MATLAB code was used for the constructed artificial intelligent part. The results showed that the vibration error can be detected early if the beam length is reduced in the wireless sensor. The system was designed based on utilizing the wireless sensor (sensor of vibration, microprocessor, Zig Bee), while the second part contained the coordinator to collect data from the wireless sensor and the cod for processing and analyzing data within a computer. The method proposed in this work shows that the processing time required to collect and analyze vibration data is 2.63 seconds, which is less than the processing time in other methods. The system can be used with other types of machines based on the training of new neural networks to obtain new information to reduce detection time and classification.
Keywords: Vibration faults detection, faults detection methods, wireless sensor networks, neural network.