Journal of Intelligent & Fuzzy Systems - Volume 34, issue 6

Authors: Yang, Shuai | Li, Zhaojun | Guo, Jian

Article Type: Research Article

Abstract: Lithium Ion batteries usually degrade to an unacceptable capacity level after hundreds or even thousands of charge and discharge cycles. The continuously observed capacity fade data over time and their internal structure can be informative for constructing capacity fade models. This paper applies a mean-covariance decomposition (MCD) modeling method using data within moving windows to analyze the capacity fade process. The proposed approach directly examines the variances and correlations in data of interest and reparameterize the correlation matrix in hyper-spherical coordinates using angle and trigonometric functions. To improve the interpretation of the prognostics model, the mean function is obtained based …on physics of failure. Non-parametric methods are used to characterize the log variance and correlation through the number of cycles and time lags between capacity measurements, respectively. A numerical example is used to illustrate the superiority of the proposed method in prediction performance. Show more

Keywords: Lithium ion battery, capacity fading, mean-covariance decomposition, correlation matrix reparameterization, physics of failure, non-parametric, moving window

DOI: 10.3233/JIFS-169549

Citation: Journal of Intelligent & Fuzzy Systems, vol. 34, no. 6, pp. 3759-3769, 2018

Authors: Li, Xiaochuan | Duan, Fang | Bennett, Ian | Mba, David

Article Type: Research Article

Abstract: Reciprocating compressors are widely used in oil and gas industry for gas transport, lift and injection. Critical compressors that compress flammable gases and operate at high speeds are high priority equipment on maintenance improvement lists. Identifying the root causes of faults and estimating remaining usable time for reciprocating compressors could potentially reduce downtime and maintenance costs, and improve safety and availability. In this study, Canonical Variate Analysis (CVA), Cox Proportional Hazard (CPHM) and Support Vector Regression (SVR) models are employed to identify fault related variables and predict remaining usable time based on sensory data acquired from an operational industrial reciprocating …compressor. 2-D contribution plots for CVA-based residual and state spaces were developed to identify variables that are closely related to compressor faults. Furthermore, a SVR model was used as a prognostic tool following training with failure rate vectors obtained from the CPHM and health indicators obtained from the CVA model. The trained SVR model was utilized to estimate the failure degradation rate and remaining useful life of the compressor. The results indicate that the proposed method can be effectively used in real industrial processes to perform fault diagnosis and prognosis. Show more

Keywords: Condition monitoring, canonical variate analysis, cox proportional hazard model, support vector regression

DOI: 10.3233/JIFS-169550

Citation: Journal of Intelligent & Fuzzy Systems, vol. 34, no. 6, pp. 3771-3783, 2018

Authors: Pang, Jingyue | Liu, Datong | Peng, Yu | Peng, Xiyuan

Article Type: Research Article

Abstract: Telemetry data, sent by the satellite, is the only basis for ground staffs to monitor on-board equipment status. In addition, the pattern discovery and operating state identification of telemetry data are very essential for automatic anomaly detection and problem diagnosis for satellites. Clustering, as an important data mining method for time series, can realize pattern discovery of satellite telemetry data automatically and intelligently, whereas the large amount of raw data and pseudo-period characteristic make the clustering on raw data inefficient and susceptible to noise interference. Thus, based on the prominent shape features and Time-Spatial specialty, a clustering framework is proposed …for telemetry data mining with physical-based segmentation and improved time series representation. Moreover, different distance measures are introduced to this framework to realize the time series clustering. The experiments are firstly performed on the public data sets which have high similarity with the real satellite telemetry to quantify the clustering accuracy, then a case study on the real satellite telemetry verifies the effectiveness and applicability of the proposed framework. Show more

Keywords: Satellite telemetry time series, clustering, representation, special-points series

DOI: 10.3233/JIFS-169551

Citation: Journal of Intelligent & Fuzzy Systems, vol. 34, no. 6, pp. 3785-3798, 2018

Authors: Cabrera, Diego | Sancho, Fernando | Cerrada, Mariela | Sánchez, René-Vinicio | Tobar, Felipe

Article Type: Research Article

Abstract: Usually, time series acquired from some measurement in a dynamical system are the main source of information about its internal structure and complex behavior. In this situation, trying to predict a future state or to classify internal features in the system becomes a challenging task that requires adequate conceptual and computational tools as well as appropriate datasets. A specially difficult case can be found in the problems framed under one-class learning. In an attempt to sidestep this issue, we present a machine learning methodology based in Reservoir Computing and Variational Inference. In our setting, the dynamical system generating the time …series is modeled by an Echo State Network (ESN), and the parameters of the ESN are defined by an expressive probability distribution which is represented as a Variational Autoencoder. As a proof of its applicability, we show some results obtained in the context of condition-based maintenance in rotating machinery, where vibration signals can be measured from the system, our goal is fault detection in helical gearboxes under realistic operating conditions. The results show that our model is able, after trained only with healthy conditions, to discriminate successfully between healthy and faulty conditions and overcome other classical methodologies. Show more

Keywords: Dynamical system modeling, deep learning, reservoir computing, variational inference

DOI: 10.3233/JIFS-169552

Citation: Journal of Intelligent & Fuzzy Systems, vol. 34, no. 6, pp. 3799-3809, 2018

Authors: Wang, Xiaodan | Yu, Qibing | Yang, Yi

Article Type: Research Article

Abstract: Wind speed forecasting is a prerequisite and a guarantee for the efficient operation of wind farms. Estimation of the non-stationary property and prediction of the future moving average of wind speed are challenging because of various environmental factors. In this paper, a new model is proposed for wind speed forecasting. The proposed model consists of three main stages: preprocessing, regression, and aggregation. In the first stage, the original wind speed time series is decomposed into different subseries by using the variational mode decomposition technique. These subseries are then used to construct training patterns and forecasted outputs. In the second stage, …support vector regression is applied to fit and forecast the wind speed for each subseries. Eventually, the final forecasted wind speed is calculated by summing all the forecasting values of each subseries. The performance of the proposed model is evaluated using real data collected from a wind farm in China, and the experimental results confirm the superiority of the proposed model to the existing models with respect to accurate forecasting and stability. Show more

Keywords: Variational mode decomposition, support vector regression, forecast, short-term wind speed

DOI: 10.3233/JIFS-169553

Citation: Journal of Intelligent & Fuzzy Systems, vol. 34, no. 6, pp. 3811-3820, 2018

Authors: Long, Jianyu | Sun, Zhenzhong | Chen, Haibin | Bai, Yun | Hong, Ying

Article Type: Research Article

Abstract: The integrated determination of the charge batching and casting start time (CBCST) is a combinatorial optimization problem extracted from the production and operations management of steel plants. A hierarchical optimization method based on variable neighborhood search (VNS) is proposed in this paper for integrated determination of CBCST. The number of casts on each continuous caster and the number of charges in each cast are determined in the encoding. The decoding process is decomposed into solving two sub-problems. A mixed integer programming (MIP) model is built for the first sub-problem by considering it as a prize collecting multiple traveling salesmen problem, …and a VNS algorithm is proposed for solving the model. In addition, another MIP model is developed for the second sub-problem, and the model is solved by CPLEX directly. Experimental results on practical production data demonstrate that the proposed algorithm is effective for determining the charge batching and the casting start time simultaneously. Show more

Keywords: Variable neighborhood search, charge batching, casting start time determining, mixed integer programming

DOI: 10.3233/JIFS-169554

Citation: Journal of Intelligent & Fuzzy Systems, vol. 34, no. 6, pp. 3821-3832, 2018

Authors: Li, Yong | Tao, Yan

Article Type: Research Article

Abstract: Air is regarded as one of a fundamental element for the survival of human and other living creatures. Daily PM10 concentration forecasting is a useful measure that is applied to the prevention and control of work in advance. This paper proposes a multiscale fusion support vector regression (MFSVR) method for forecasting daily PM10 concentration. The method uses stationary wavelet transform (SWT) to decompose original time series of daily PM10 concentration into different scales, of which the information represents wavelet coefficients of PM10 concentration. At each scale, wavelet coefficients are used for training a support vector regression …(SVR) model. The estimated coefficients of the SVR outputs for all of the scales applied to the reconstruction of the prediction result by the inverse SWT. To enhance forecasting of the MFSVR, a feature fusion approach that bases on partial least squares is adopted to extract the original features and reduce dimensions for input variables of the SVR model. The experimental confirmation of the proposed method is tested by applying the data of four monitoring stations between 1/1/2015 and 26/12/2015 in Lanzhou, China. The results indicate that the MFSVR approach can precisely forecast daily PM10 concentration on the basis of mean absolute error, mean absolute percentage error, root mean square error and correlation coefficient criteria. This method shows a potential prospect that can be implemented in air quality prediction systems in other areas. Show more

Keywords: PM10 forecasting, multiscale, stationary wavelet transform, partial least squares, support vector regression

DOI: 10.3233/JIFS-169555

Citation: Journal of Intelligent & Fuzzy Systems, vol. 34, no. 6, pp. 3833-3844, 2018

Authors: Shi, Fang | Liu, Yihao | Liu, Zheng | Li, Eric

Article Type: Research Article

Abstract: American Water Works Association has estimated that, by 2050, the total cost of pipeline system management will exceed $1.7 trillion. Thus, it is important to assess the performance of water mains in order to optimize the rehabilitation process. Recently, the use of machine learning methods in pipeline condition prediction has increased. However, existing pipe performance prediction models rely solely on underlying data-generating distributions and do not accommodate different datasets. Hence, a stacking ensemble based method is proposed in this work to overcome the drawbacks of the existing models and improve the predictive power of this mode of analysis. Using soil …property data, both a single-model and an ensemble-model were constructed to forecast the pipe condition, and their prediction performance was compared and contrasted. Finally, the superiority of the proposed ensemble method was verified through its lowest value in the root-mean-square error relative to the individual models. The techniques presented in this work can aid in a reliable decision making in infrastructure management of buried pipeline networks. Show more

Keywords: Stacking ensemble, prediction, regression, cast iron, soil corrosivity

DOI: 10.3233/JIFS-169556

Citation: Journal of Intelligent & Fuzzy Systems, vol. 34, no. 6, pp. 3845-3855, 2018

Authors: Song, Wenlei | Xiang, Jiawei | Zhong, Yongteng

Article Type: Research Article

Abstract: Bearings are essential parts in mechanical transmission systems, and their running states directly affect the reliability and stability of the systems. Therefore, an efficient diagnosis method is necessary to detect faults in bearings. In the present, a simulation model based fault diagnosis method for bears is proposed by combination of finite element method (FEM), wavelet packet transform (WPT) and support vector machine (SVM). In this method, firstly, the agreeable finite element models to simulate faulty bearings are presented to obtain the vibration response signals. Secondly, the vibration signals are decomposed into eight signal components using WPT. Ten time-domain feature parameters …of all the signal components are calculated to generate the training samples to train the SVM. Finally, the eight signal components decomposed by WPT from the measured vibration signal in a bear, which are serve as a test sample into the trained SVM, and the work condition of the bearing can be determined. Experimental investigations are performed to verify the effectiveness of the present method. The classification accuracy rates for four type faults, i.e., inner race fault, rolling body fault, outer race fault and the combination of rolling body and outer race faults, are 79%, 81%, 71% and 76%, respectively. Show more

Keywords: Fault detection, bearing, numerical simulation, wavelet packet transform, support vector machine

DOI: 10.3233/JIFS-169557

Citation: Journal of Intelligent & Fuzzy Systems, vol. 34, no. 6, pp. 3857-3867, 2018

Authors: Guo, Wei | Chen, Cheng | Xiao, Ningcong

Article Type: Research Article

Abstract: Planetary gearbox (PG), especially multi-stage PG, is widely used in various commercial and military applications. A dynamic model of PG is a valuable way to investigate its vibration behaviors in good and faulty statuses for designing a highly reliable PG and corresponding robust monitoring techniques. In this paper, a modified dynamic model is firstly constructed for a two-stage PG with a varying crack. The modifications include the crack propagation path settings as well as the coupled structure of flexible and rigid gear components, which makes a balance between the model accuracy and computational burden. Secondly, using this model, its vibration …responses from a healthy PG and faulty PGs with different cracks are collected for the analyses of its frequency components and statistic features. Experiments are also conducted on a two-stage PG in similar conditions. Both results in simulations and experiments demonstrate the usefulness of the presented dynamic model. Moreover, sidebands and some statistic features are verified to be useful tools to monitor and evaluate the healthy status of PG. Show more

Keywords: Planetary gearbox, tooth crack, dynamic model, vibration response, condition indicator

DOI: 10.3233/JIFS-169558

Citation: Journal of Intelligent & Fuzzy Systems, vol. 34, no. 6, pp. 3869-3880, 2018