Balancing of a Highly Flexible Rotor by Using Artificial Neural Networks

2007 ◽  
Author(s):  
Manuel Villafan˜e Saldarriaga ◽  
Jarir Mahfoud ◽  
Valder Steffen ◽  
Johan Der Hagopian
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Influence Coefficient ◽  
Flexible Rotor ◽  
Active Balancing ◽  
Influence Coefficient Method ◽  
Alternative Methodology ◽  
Artificial Neural ◽  
Coefficient Method

The present work is an alternative methodology in order to balance an unsymmetrical damped highly flexible rotor by using neural networks. This procedure was developed aiming at improving the performance of classical balancing methods, which are not well adapted to these situations. The approach developed is based on the influence coefficient method and is adequate to integrate an active balancing system. The methodology is tested successfully experimentally.

Further Experiments on Balancing of a High-Speed Flexible Rotor

1974 ◽  
Vol 96 (2) ◽  
pp. 431-440 ◽  
Author(s):  
J. Tonnesen
Keyword(s):  
Computer Program ◽  
Least Squares ◽  
High Speed ◽  
Influence Coefficient ◽  
Flexible Rotor ◽  
Minimum Level ◽  
Absolute Level ◽  
Quality Class ◽  
Influence Coefficient Method ◽  
Coefficient Method

The accuracy of the influence coefficient method is experimentally investigated. The influence of the number of measurement transducers, their location, and type is demonstrated on a flexible rotor where simultaneous balancing is performed in up to five planes and passing through three critical speeds. The correction weights are calculated by means of a computer program, based on a least-squares minimizing procedure. The method itself is shown to be accurate and uses only a minimum of balancing runs to reduce the vibrations to a true minimum level. The overall accuracy in determining the unbalance weights is found to be 4.5 percent. The method’s effectiveness is demonstrated on a rotor with four balancing planes and with unbalance distributed at random in six and seven planes. The absolute level of residual vibrations is found to be in the ISO 0.4 quality class [5].

A Generalized Minmax Influence Coefficient Method for Flexible Rotor Balancing

2005 ◽  
Author(s):  
Costin Untaroiu ◽  
Paul Allaire
Keyword(s):  
High Speed ◽  
Optimization Problem ◽  
Current Method ◽  
Linear Matrix ◽  
Influence Coefficient ◽  
Flexible Rotor ◽  
Influence Coefficient Method ◽  
Rotor Balancing ◽  
Practical Constraints ◽  
Coefficient Method

Rotor balancing is a requirement for the smooth operation of high-speed rotating machinery. In field balancing, minimization of the residual vibrations at important locations/speeds under practical constraints is usually a challenging task. In this paper, the generalized minmax coefficient influence method is formulated as an optimization problem with flexible objective functions and constraints. The optimization problem is cast in a Linear Matrix Inequality (LMI) form and a balancing code is developed to solve it. Two balancing examples are run to verify the efficiency and flexibility of the proposed method. Over the existing methods, current method is more flexible for the various requirements encountered in field balancing and can be solved accurate with current mathematical software.

Flexible Rotor Balancing by the Exact Point-Speed Influence Coefficient Method

1972 ◽  
Vol 94 (1) ◽  
pp. 148-158 ◽  
Author(s):  
J. M. Tessarzik ◽  
R. H. Badgley ◽  
W. J. Anderson
Keyword(s):  
Influence Coefficient ◽  
Flexible Rotor ◽  
Critical Speeds ◽  
Angle Measurements ◽  
Flexible Rotors ◽  
Speed Range ◽  
Influence Coefficient Method ◽  
Rotor Balancing ◽  
Exact Point ◽  
Coefficient Method

A test program was conducted to confirm experimentally the validity of the exact point-speed influence coefficient method for balancing rotating machinery, and to assess the practical aspects of applying the method to flexible rotors. Testing was performed with a machine having a 41-in. long, 126-lb rotor. The rotor was operated over a speed range encompassing three rotor-bearing system critical speeds: two “rigid-body” criticals and one flexural critical. Rotor damping at the flexural critical was very low due to the journal bearings being located at the nodal points of the shaft. The balancing method was evaluated for three different conditions of initial rotor unbalance. The method was found to be effective and practical. Safe passage through all the critical speeds was obtained after a reasonable number of balancing runs. Success of the balancing method was, in large part, due to the accuracy of the instrumentation system used to obtain phase-angle measurements during the balancing procedure.

Extended Influence Coefficient Method for Rotor Active Balancing During Acceleration

2004 ◽  
Vol 126 (1) ◽  
pp. 219-223 ◽  
Author(s):  
Shiyu Zhou ◽  
Stephen W. Dyer ◽  
Kwang-keun Shin ◽  
Jianjun Shi ◽  
Jun Ni
Keyword(s):  
Fatigue Life ◽  
Gain Scheduling ◽  
Rotor System ◽  
Influence Coefficient ◽  
Resonant Vibration ◽  
Active Balancing ◽  
Look Up Table ◽  
Influence Coefficients ◽  
Influence Coefficient Method ◽  
Coefficient Method

Imbalance-induced vibration of rotating machineries is an important factor limiting the performance and fatigue life of a rotor system. Particularly, the severe resonant vibration of a rotor when it passes through its critical speeds could damage the rotor system. To avoid this peak vibration, this paper presents an active balancing method to offset the imbalance of the rotor system during acceleration by using an electromagnetic balancer. In this method, “instantaneous” influence coefficients at different speeds are obtained and stored in a look-up table. Then, a gain scheduling strategy is adopted to suppress the imbalance-induced vibration during acceleration based on the “instantaneous” influence coefficient table. A comprehensive testbed is built to validate this scheme, and the validation results are presented.

scholarly journals On a programme for the balancing calculation of flexible rotors with the influence coefficient method

2000 ◽  
Vol 22 (4) ◽  
pp. 235-247
Author(s):  
Nguyen Van Khang ◽  
Tran Van Luong
Keyword(s):  
Computer Software ◽  
Rotor System ◽  
Influence Coefficient ◽  
Flexible Rotor ◽  
C Language ◽  
Flexible Rotors ◽  
University Of Technology ◽  
Influence Coefficient Method ◽  
Coefficient Method

This paper presents the influence coefficient method of determining the locations of unbalances on a flexible rotor system and the correction weights. A computer software for calculating the at-the-site balancing of a flexible rotor system was created using C++ language at the Hanoi University of Technology. This software can be used by balancing flexible rotors in Vietnam.

Determination of S-N curves with the application of artificial neural networks

1999 ◽  
Vol 22 (8) ◽  
pp. 723-728 ◽  
Author(s):  
Artymiak ◽  
Bukowski ◽  
Feliks ◽  
Narberhaus ◽  
Zenner
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Artificial Neural
Sign in / Sign up

Export Citation Format

Share Document