M. Farid Golnaraghi

Prognosis of machine health condition using neuro-fuzzy systems

A reliable machine fault prognostic system can be used to forecast damage propagation trend in rotary machinery and to provide an alarm before a fault reaches critical levels. Currently, there are several techniques available in the literature for time-series prediction. Among the most promising methods are recurrent neural networks (RNNs) and neuro-fuzzy (NF) systems. In this paper, the performance of these two types of predictors is evaluated using two benchmark data sets. Through comparison it is found that if an NF system is properly trained, it performs better than RNNs in both forecasting accuracy and training efficiency. Accordingly, NF system is adopted to develop an on-line machine fault prognostic system. In order to facilitate the automatic monitoring process, reference function approach is proposed here to enhance feature representation. The performance of the developed prognostic system is evaluated by using three test cases including a worn gear, a chipped gear, and a cracked gear, as well as using data sets from previous studies corresponding to a gear pitting damage and a shaft misalignment. From these tests, the NF prognostic system is found to be a very reliable and robust machine health condition predictor. It can capture the system dynamic behaviour quickly and accurately.

A neuro-fuzzy approach to gear system monitoring

The detection of the onset of damage in gear systems is of great importance to industry. In this paper, a new neuro-fuzzy diagnostic system is developed, whereby the strengths of three robust signal processing techniques are integrated. The adopted techniques are: the continuous wavelet transform (amplitude) and beta kurtosis based on the overall residual signal, and the phase modulation by employing the signal average. Three reference functions are proposed as post-processing techniques to enhance the feature characteristics in a way that increases the accuracy of fault detection. Monitoring indexes are derived to facilitate the automatic diagnoses. A constrained-gradient-reliability algorithm is developed to train the fuzzy membership function parameters and rule weights, while the required fuzzy completeness is retained. The system output is set to different monitoring levels by using an optimization procedure to facilitate the decision-making process. The test results demonstrate that the novel neuro-fuzzy system, because of its adaptability and robustness, significantly improves the diagnostic accuracy. It outperforms other related classifiers, such as those based on fuzzy logic and neuro-fuzzy schemes, which adopt different types of rule weights and employ different training algorithms.

An Extended Wavelet Spectrum for Bearing Fault Diagnostics

Rolling-element bearings are widely used in various mechanical and electrical systems. A reliable online bearing fault-diagnostic technique is critically needed to prevent the systems performance degradation and malfunction. In this paper, an extended wavelet spectrum analysis technique is proposed for a more positive assessment of bearing health conditions. Two strategies have been suggested for different wavelet function implementation. Two statistical indexes are proposed to quantify the resulting wavelet (coefficient) functions. Based on the information provided by these indexes, the wavelet functions can be deployed more effectively over the designated frequency bands. An extended Shannon function is proposed to synthesize the wavelet coefficients over selected bandwidths to enhance feature characteristics. An averaged autocorrelation power spectrum is adopted to highlight bearing characteristics. The viability of the developed technique is verified by online experimental tests corresponding to different bearing conditions.

Regulation of flexible structures via nonlinear coupling

In this article, we propose an active/passive vibration controller for a cantilever beam using a sliding mass-spring-dashpot mechanism. The controller is placed at the free end of the beam, introducing Coriolis, inertia, and centripetal nonlinearities into the system, resulting in nonlinear coupling that may be used to quench the transient vibration of the beam. When the natural frequency of the slider is twice the fundamental beam frequency (2:1 internal resonance), the two systems will be coupled through nonlinearities that cause the oscillatory energy to be transferred back and forth between the beam and the slider. Control is achieved once the vibration of the beam is absorbed by the slider and dissipated through the slider damping. Numerical results show that this technique can improve the effective damping ratio of the structure by a factor of 15. This technique is particularly useful for reducing large-amplitude oscillations to levels that may be managed using conventional methods.Due to the nonlinearities in the system, for small or zero controller damping, chaotic transient oscillations can occur depending on the amplitude of the initial disturbance of the beam.

An Enhanced Diagnostic Scheme for Bearing Condition Monitoring

Rolling-element bearings are widely used in various mechanical and electrical facilities; accordingly, a reliable real-time bearing condition-monitoring system is very useful in industries to detect bearing defects at both incipient and advanced levels to prevent machinery performance degradation and malfunctions. The objective of this paper is to develop an enhanced diagnostic (ED) scheme for bearing fault diagnostics. This scheme consists of modules of classification and prediction. A neurofuzzy (NF) classifier is proposed to effectively integrate the strengths of several signal-processing techniques (or resulting representative features) for a more positive assessment of bearing health conditions. A multistep NF predictor is employed to forecast the future states of the bearing health condition to further enhance the diagnostic reliability. The effectiveness of this ED scheme is verified by experimental tests that correspond to different bearing conditions.

Dynamics of a Flexible Cantilever Beam Carrying a Moving Mass

The motion of a flexible cantilever beam carrying a moving spring-mass system is investigated. The beam is assumed to be an Euler–Bernouli beam. The motion of the system is described by a set of two nonlinear coupled partial differential equations where the coupling terms have to be evaluated at the position of the mass. The nonlinearities arise due to the coupling between the mass and the beam. Due to the nonlinearities the system exhibits internal resonance which is investigated in this work. The equations of motion are solved numerically using the Rayleigh–Ritz method and an automatic ODE solver. An approximate solution using the perturbation method of multiple scales is also obtained.

A neural network based fuzzy control approach to improve ride comfort and road handling of heavy vehicles using semi-active dampers

The use of semi-active dampers in suspension systems is increasing rapidly in the automotive industry. This is attributed to the fact that damping can be adjusted to improve ride comfort and road handling. However, accomplishing ride comfort and better road handling concurrently is challenging. In this paper, we propose an intelligent system to optimise the suspension performance in terms of concurrently achieving the above-mentioned objectives. A neural network-based fuzzy logic controller is designed to control the system that embodies this damper. The proposed controller can enhance the vehicle handling and ride comfort concurrently while consuming less energy than existing control methodologies.

A novel fuzzy framework for nonlinear system control

Nonlinear PID and gain scheduling controls have attracted much research interests in recent years. These control strategies can accommodate some nonlinear characteristics by allowing the gains varied or rescheduled as a function of system states. In this paper, a novel fuzzy framework is developed to transfer this type of nonlinear controller to its fuzzy domain representation. It is proved that the resulting fuzzy controller is functionally exactly identical to the original control system. One of the benefits of the suggested approach is that it provides a well-defined prototype for the design of fuzzy control system. The resulting linguistic representation can facilitate investigation in linguistic terms into how the controller operates, whereas expert knowledge can be effectively implemented to improve control performance. The viability of the proposed mapping technique is demonstrated by using simulations corresponding to a flexible-link robot.

Regulation of a lumped parameter cantilever beam via internal resonance using nonlinear coupling enhancement

In this article, we propose a nonlinear control law to enhance the performance of internal resonance (IR) controllers used for the regulation of vibration in flexible structures. Active IR controllers are composed of sliding or rotational actuators attached to the flexible structure, introducing dynamic nonlinearities into the system. The IR control strategy forces a state of internal resonance at which the transfer of oscillatory energy from the structure to the controller mechanism takes place. Once this energy is transferred to the controller it is dissipated via velocity feedback. These controllers are unconventional since they function at a state of resonance. The resonance condition is established upon tuning the controller natural frequency (using position feedback) to twice the fundamental beam frequency which causes the oscillatory energy to be transferred to the controller via the nonlinear coupling terms where it is dissipated. However, the rate of dissipation is limited since the controllers have a limited “critical” damping coefficient.In this article we study the effect of the IR controller on a simplified lumped mass model representing a cantilever beam. The nonlinear enhancement technique proposed in this article improves the performance of this type of regulatory by increasing the nonlinear coupling effect which is responsible for the energy transfer, and hence, speeding the rate of vibration suppression.

Experimental Control of Flexible Structures Using Nonlinear Modal Coupling: Forced and Free Vibration

This paper is an experimental study of free and forced vibration suppression in a piezoceramic actuated flexible beam via the nonlinear Modal Coupling Control (MCC). The method is based on transferring the oscillatory energy from the plant to an auxiliary second order system (controller), coupled to the plant through nonlinear terms. The proposed controller produces an input that can be utilized by unidirectional actuators. Unidirectional actuators do not generate symmetric power during an application. Shape memory alloys, thrusters and cable based actuators are examples of this class of actuators. Existing control methods assume a symmetric actuation and therefore application of unidirectional actuators call for new control techniques. Current control techniques implement a bias in utilizing unidirectional actuators. However, the amount of the bias is variable and depends on the control effort. Moreover, the use of a bias changes the system equilibrium point and introduces a steady state error. The proposed controller is used to control the first mode of a flexible beam. The results show that the method is more effective than conventional control approaches in conjunction with unidirectional actuators.