Estimation and forecasting of machine health condition using ARMA/GARCH model

This paper proposes the hybrid model of autoregressive moving average (ARMA) and generalized autoregressive conditional heteroscedasticity (GARCH) to estimate and forecast the machine state based on vibration signal. The main idea in this study is to employ the linear ARMA model and the nonlinear GARCH model to explain the wear and fault condition of machine, respectively. The successful outcomes of the ARMA/GARCH prediction model can give obvious explanation for future states of machine, which enhance the worth of machine condition monitoring as well as condition-based maintenance in practical applications. The advance of the proposed model is verified in empirical results as applying for a real system of a methane compressor in a petrochemical plant.     

Directionally controllable squeeze film damper using liquid crystal

Liquid crystals (LC) are characterized by its phase, which appears as an intermediate state between crystalline solid and isotropic liquid. This intermediate, phase is caused by orientation of molecules, and it can be controlled by an externally applied electric or magnetic field. Subjected to an electric field, the viscosity of the LC varies according to the applied electric field strength, which is called the electroviscous effect. This paper describes an application study of the electroviscous effect of a LC to a controllable squeeze film damper (SFD) for a rotating machine. A prototype controllable SFD using a LC was constructed and its performance was studied. It should be noted that the present SFD can produce anisotropic damping force for a flexible rotor at the supporting position, which enables us to stabilize a flexible rotor in a wide range of its rotating speed.     

Applications of hybrid optimization techniques for model updating of rotor shafts

This paper presents an approach by combining the genetic algorithm (GA) with simulated annealing (SA) algorithm for enhancing finite element (FE) model updating. The proposed algorithm has been applied to two typical rotor shafts to test the superiority of the technique. It also gives a detailed comparison of the natural frequencies and frequency response functions (FRFs) obtained from experimental modal testing, the initial FE model and FE models updated by GA, SA, and combination of GA and SA (GA–SA). The results concluded that the GA, SA, and GA–SA are powerful optimization techniques which can be successfully applied to FE model updating, but the appropriate choice of the updating parameters and objective function is of great importance in the iterative process. Generally, the natural frequencies and FRFs obtained from FE model updated by GA–SA show the best agreement with experiments than those obtained from the initial FE model and FE models updated by GA and SA independently.     

Bearing parameter identification of rotor–bearing system using clustering-based hybrid evolutionary algorithm

A new bearing parameter identification methodology based on global optimization scheme using measured unbalance response of rotor–bearing system is proposed. A new hybrid evolutionary algorithm which is a clustering-based hybrid evolutionary algorithm (CHEA), is proposed for global optimization scheme to improve the convergence speed and global search ability. Clustering of individuals by using a neural network is introduced to evaluate the degree of mature of genetic evolution. After clustering-based genetic algorithm (GA), local search is carried out for each cluster to judge the convexity of each cluster. Finally, random search is adapted for extrasearching to find a potential global candidate, which could be missed in GA and local search. The proposed methodology can identify not only unknown bearing parameters but also unbalance information of disk by simply setting them as unknown parameters. Numerical example and experimental results were used to verify the effectiveness of the proposed methodology.     

Development of integrated evolutionary optimization algorithm and its application to optimum design of ship structures

This paper proposes an integrated evolutionary optimization algorithm (IEOA) which is combined with genetic algorithm (GA), random tabu search method (TS) and response surface methodology (RSM). This algorithm, in order to improve the convergent speed that is thought to be the demerit of GA, uses RSM and the simplex method. Though mutation of GA offers random variety, systematic variety can be secured through the use of tabu-list. Efficiency of this method has been proven by applying traditional test functions and comparing the results to GA. And it is an evidence that the newly suggested algorithm can effectively find the global optimum solution by applying it to minimize the weight of fresh water tank that is placed in the rear of ship designed to avoid resonance. According to the results, GA’s convergent speed in initial phase has been improved by using RSM. An optimized solution was calculated without the evaluation of additional actual objective function. Finally, it can be concluded that IEOA is a very useful global optimization algorithm from the viewpoint of convergent speed and global search ability.     

Data-driven approach to machine condition prognosis using least square regression tree

Machine fault prognosis techniques have been profoundly considered in the recent time due to their substantial profit for reducing unexpected faults or unscheduled maintenance. With those techniques, the working conditions of components, the trending of fault propagation, and the time-to-failure are precisely forecasted before they reach the failure thresholds. In this work, we propose the least square regression tree (LSRT) approach, which is an extension of the classification and regression tree (CART), in association with one-step-ahead prediction of time-series forecasting techniques to predict the future machine condition. In this technique, the number of available observations is first determined by using Cao’s method and LSRT is employed as a prediction model in the next step. The proposed approach is evaluated by real data of a low methane compressor. Furthermore, a comparative study of the predicted results obtained from CART and LSRT are carried out to prove the accuracy. The predicted results show that LSRT offers the potential for machine condition prognosis. This paper was recommended for publication in revised form by Associate Editor Eung-Soo Shin Van Tung Tran is a lecturer at the Hochiminh City University of Technology in Vietnam. He received his B.S. and M.S. degrees in mechanical engineering from Hochiminh City University of Technology, Vietnam, in 1997 and 2003, respectively, and Ph.D. from Pukyong National University, South Korea in 2009. His research interests include machine fault diagnosis and condition prognosis. Bo-Suk Yang is a professor at the Puyong National University in Korea. He received his Ph.D. degree in mechanical engineering from Kobe University, Japan in 1985. His main research fields cover machine dynamics and vibration engineering, intelligent optimum design, and condition monitoring and diagnostics in rotating machinery. He has published well over 190 research papers in the research areas of vibration analysis, intelligent optimum design and diagnosis of rotating machinery. He is listed in Who’s Who in the World, Who’s Who in Science and Engineering, among others.     

Application of grey model for machine degradation prognostics

Predicting machine degradation before final failure occurs is very important. This paper presents a method to predict the future state of machine degradation based on grey model and one-step-ahead forecasting technique. Specifically, the feasibility of grey model as a predictor for machine degradation prognostics system has been investigated. Grey model GM(1,1) has employed to forecast the future state of machine degradation, but the result is not satisfactory. Finally, a modification of GM(1,1) has made to improve the accuracy of prediction. However the model was built by using only four input data, it is able to track closely the sudden change of machine degradation condition. Real trending data of low methane compressor acquired from condition monitoring routine are employed for evaluating the proposed method.     

Optimum design of short journal bearings by enhanced artificial life optimization algorithm

This paper presents an optimum design of high-speed short journal bearing using an enhanced artificial life algorithm (EALA) to compute the solutions of optimization problem. The proposed hybrid EALA algorithm is a synthesis of an artificial life algorithm (ALA) and the random tabu search method (R-tabu method) to solve some demerits of the ALA. The emergence is the most important feature of the artificial life which is the result of dynamic interaction among the individuals consisting of the system and is not found in an individual. The artificial life optimization algorithm is a stochastic searching algorithm using the feature of artificial life. The feature of R-tabu method, which prevents converging to the local minimum, is combined with the ALA. One of the features of the R-tabu method is to divide any given searching region into several sub-steps. As the result of the combination of the two methods, the EALA not only converges faster than the ALA, but also can lead to a more accurate solution. In addition, this algorithm can also find all global optimum solutions. We applied the hybrid algorithm to the optimum design of a short journal bearing. The optimized results were compared with those of ALA and successive quadratic programming, and identified the reliability and usefulness of the hybrid algorithm.     

Optimum shape design of rotating shaft by ESO method

Evolutionary structural optimization (ESO) method is based on a simple idea that the optimal structure can be produced by gradually removing the ineffectively used material from the design domain. ESO seems to have some attractive features in engineering aspects: simple and fast. In this paper, ESO is applied to optimize shaft shape for the rotating machinery by introducing variable size of finite elements in optimization procedure. The goal of this optimization is to reduce total shaft weight and resonance magnification factor (Q factor), and to yield the critical speeds as far from the operating speed as possible. The constraints include restrictions on critical speed, unbalance response and bending stresses. Sensitivity analysis of the system parameters is also investigated. The results show that new ESO method can be efficiently used to optimize the shape of rotor shaft system with frequency and dynamic constraints.     

Intelligent fault diagnosis system of induction motor based on transient current signal

This paper presents a method for induction motor fault diagnosis based on transient signal using component analysis and support vector machine (SVM). The start-up transient current signal is selected as features source for fault diagnosis. Preprocessing of transient current signal is performed using smoothing and discrete wavelet transform to highlight the salient features of faults. In this work, independent component analysis, principal component analysis and their kernel are performed to reduce the dimension of features and to extract the optimal features for classification process. In this work, the influence of the number of component analysis towards diagnosis accuracy is also studied. SVM multi-class classification using one against all strategy is selected for classification tool due to good generalization properties. Performance of the system is validated by applying the system to induction motor faults diagnosis. According to the result, the system has potential to serve an intelligent fault diagnosis system in real application.