Dual-plane Rotor Balancing Using Artificial Neural Network to Predict the Correction Mass and Phase Angle

Rotating machinery is pervasive in mechanical systems, encompassing components such as motor and engine rotors, machining tools, and industrial turbomachinery. The imbalance of rotating masses around an axis results in rotor unbalance, a critical issue causing excessive vibrations, especially at higher speeds. Centrifugal unbalanced forces can damage bearings and lead to machine destruction, making the resolution of unbalance a fundamental concern in machinery design and operation. Although not to zero, the reduction of vibrations is deemed acceptable by decreasing them below prescribed values for specific machinery quality classes. Balancing the rotor extends bearing life, minimizes vibrations, audible noise, power losses, and enhances product quality. This research delves into the dynamic behavior of a two-plane rotor, meticulously modeled using MATLAB. Intentional mass imbalance, simulating real-world conditions, induces unbalanced vibrations, necessitating correction mass through the plane separation balancing technique. The study innovatively employs an Artificial Neural Network (ANN) to address rotor imbalance, comparing its efficacy with traditional vector balancing methods. The results highlight the remarkable effectiveness of artificial neural networks in mitigating rotor imbalance challenges across various rotating machinery applications. This nuanced investigation into a rotor model, coupled with advanced balancing techniques, signifies a substantial advancement in improving the operational efficiency and reliability of rotating machinery in diverse industrial settings.

Keywords:

Vibration , Rotation , Rigid Rotor Balancing , Artificial Neural Network