Elsevier

Measurement

Volume 55, September 2014, Pages 15-24
Measurement

Fault diagnosis on material handling system using feature selection and data mining techniques

https://doi.org/10.1016/j.measurement.2014.04.037Get rights and content

Highlights

  • We examine sensory signals of material handling systems are nonlinear and have unique characteristics.

  • We prepared fault diagnosis system using feature selection and data mining techniques.

  • Feature selection method will increase classification results.

  • LLE, AutoEncoder and Diffusion map has used for feature selection.

  • GKclsut and k-medoid has used for classification and compared each other.

Abstract

The material handling systems are one of the key components of the most modern manufacturing systems. The sensory signals of material handling systems are nonlinear and have unique characteristics. It is very difficult to encode and classify these signals by using multipurpose methods. In this study, performances of multiple generic methods were studied for the diagnostic of the pneumatic systems of the material handling systems. Diffusion Map (DM), Local Linear Embedding (LLE) and AutoEncoder (AE) algorithms were used for future extraction. Encoded signals were classified by using the Gustafson–Kessel (GK) and k-medoids algorithms. The accuracy of the estimations was better than 90% when the LLE was used with GK algorithm.

Introduction

The modern manufacturing facilities have to detect the problems, identify their sources and fix them very quickly with very limited man power. Researchers have started development of computational diagnostic tools for the industrial applications in early 1970s by considering this need. Although, various diagnostic tools have been developed by research community and successfully used in industrial applications in last two decades [1], [2] still their capabilities are limited. In this study, feasibility of a multipurpose fault detection approach was investigated. The proposed approach used the combinations of the generic dimension reduction methods for feature extraction and classified the encoded data with clustering algorithms.

One of the key components of the automated manufacturing is material handling systems. Pneumatic and hydraulic systems are widely used for material handling. These systems may have hundreds of actuators and sensors. Identification of faulty components and their locations in a very short time is very difficult. Several studies were performed for development of fault diagnostic tools for these systems in the last decade [3]. The studies mainly aimed evaluation of the condition of the cylinders [4] and digitally controlled valves [5]. Other studies focused on detection of leakage of the seals [6], [7], [8], [9], friction increase [4], [10] and malfunctions [11], [12], [13], [14].

Most of the fault diagnostic tools have two components: feature extractor (encoder) and classifier. Some researchers have used the intelligent data analysis techniques for fault diagnostic [15], [16], [17]. Support vector machines [18], self-organizing feature maps (SOM) [19], expert systems, neural networks, rough sets and fuzzy logic have been used for classification of data. The computing complexity of feature extraction and learning process have been the main disadvantages of these approaches.

The data of the material handling systems for the fault diagnosis comes from multiple sensors. The data is high-dimensional and nonlinear. While the large number of data from different sensors provide more information, at the same time feature extraction and classification becomes more complex. The dimension reduction methods compress the data automatically, reduce the noise, may extract features for fault diagnostic and minimize required storage.

Clustering algorithms have been used for classification. Fuzzy c-means (FCM) [20] and its variants Gustafson–Kessel (GK) [21] algorithm are popular pattern classification methods. They have been used for fault detection and isolation [22], [23], [24], k-medoids [25] is a partitional clustering algorithm and may be used for classification purposes. DM method [26], [27], [28], [29] used diffusion semigroups for learning the global characteristics of the data-set. The complex structures were represented at different scales by the help of these semi groups. The eigenfunctions of Markov matrices were effectively used with this purpose. LLE [30] method used an unsupervised learning algorithm to change the dimensions of the data. The algorithm found the neighbors in X space, calculated weights for reconstruction and calculated the embedding coordinates in Y space by using the calculated weights. AutoEncoder (AE) [31] methods use an artificial neural network (ANN) to learn the compact representation of data set. The dimensionality of the data set is reduced by using this ANN. Various multilayer architectures [32], [33] and optimization methods [31], [34] have been proposed to improve the performance of the ANN.

GK and k-medoids algorithms were used to create the desired number of clusters to partition or classify the data after it was compacted. GK algorithm [21] calculates the center and covariance matrix to represent the clusters [35], [36]. They are used during the optimization process. This approach allows identification of ellipsoidal clusters and improves the performance of the method relative to other approaches. k-medoids [37] is another clustering algorithm. The algorithm divides the data into the groups, chooses the data points as medoids. In this study, Clustering and Data Analysis Toolbox [38] was used for classification of the compressed data.

Diffusion map, AutoEncoder, and Local Lineer Embedding techniques were used for dimension reduction process respectively. In classification process two algorithms were used namely; k-medoids and GK. These algorithms were given in detailed below.

Section snippets

k-Medoids

It is a standard clustering algorithm [37] where the update rule always moves the cluster center to the nearest data point in the cluster. k-Medoids is a partitioning technique of clustering that clusters the data set of n objects into k clusters with k known a priori. t could be more robust to noise and outliers as compared to k-means because it minimizes a sum of general pairwise dissimilarities instead of a sum of squared Euclidean distances.

Gustafson–Kessel algorithm (GK)

Gustafson–Kessel algorithm (GK), providing a

Experimental set-up

In this study, servo-pneumatic positioning experimental set-up built by the Festo Didactic Company was used. The experimental setup is presented in Fig. 3. The gripper’s motions along the X and Y axes were controlled with pneumatic dual action rodless cylinders. For the position measurements in the X and Y axes a linear potentiometer and a contactless absolute magnetostrictive linear displacement sensor were used respectively. The pneumatic gripper of the system was installed at the Y axis

Experimental procedure and data collection

In this study, data was collected while the pneumatic system was operated at the normal and additional 10 different faulty conditions. The imposed problems are listed in Table 1. During the experiments, the data were collected from 4 analog and 2 digital sensors for a period of 27 s. The data was collected 3 times at each experimental condition. The pneumatic system’s main pressure; the x and y axis pneumatic cylinders’ pressures; pressure of the gripper’s cylinder; two proximity sensors

Dimension reduction and classification of the experimental data

The two step process for the analysis of the experimental data is presented in Fig. 8. First the dimension of the data was reduced by using the DM, LLE and AE methods. Thus the useful properties, defining the signals adequately, have been obtained from the collected signals for using further classification process. Then the compressed data was classified by using the GK and k-medoids algorithms.

Results

In this study, the data that was taken from the system has been embedded to five dimension. The classification process has been performed with this new feature space. The experimental results are shown in Fig. 9, Fig. 13.

The variation of the cost value with the iterations of the GK algorithm is presented in Fig. 9. The compressed data of the DM, LLE and AE dimension reduction algorithms were used.

The classification performance values for all algorithms were obtained by comparing the accurate

Conclusion

The typical material handling system of automated manufacturing facilities was simulated by using a trainer. The gripper of the trainer picked up objects, moved and piled up. Three pneumatic cylinders moved the gripper along the x and y axis in addition to opened and closed it. The system was operated at the normal and 10 faulty modes to collect data. Four pressure and two digital gripper mode (open/close) signals were monitored in the time domain. The pressures of the entire system and three

Acknowledgements

The authors present their special thanks to the Celal Bayar University Scientific Research Projects Commission for the supports of the study under project number 2012-52.

The authors also thanks to the Marmara University Scientific Research Projects Commission for the supports of the study under project number FEN-A-080410-0081.

References (40)

  • L. Shi, N. Sepehri, Adaptive fuzzy-neural-based multiple models for fault diagnosis of a pneumatic actuator, in:...
  • Ž. Nakutis, P. Kaškonas, An approach to pneumatic cylinder on-line conditions monitoring, Mechanika.-Kaunas:...
  • J. Wang et al.

    Identification of pneumatic cylinder friction parameters using genetic algorithms

    IEEE/ASME Trans. Mech.

    (2004)
  • Z. Nakutis, P. Kaskonas, Pneumatic cylinder diagnostics using classification methods, in: Instrumentation and...
  • Ž. Nakutis et al.

    Application of ANN for pneumatic cylinder leakage diagnostics

    Matavimai

    (2005)
  • M. Sepasi et al.

    On-line fault diagnosis of hydraulic systems using unscented Kalman filter

    Int. J. Control Autom. Syst.

    (2010)
  • T. Nogami et al.

    Failure diagnosis system on pneumatic control valves by neural network

    IEEE Int. Conf. Neural Networks

    (1993)
  • M. Demetgul, S. Taskin, I.N. Tansel, Conditioning Monitoring and Fault Diagnosis for a Servo-Pneumatic System with...
  • F.J. Uppal et al.

    Fault diagnosis of an electro-pneumatic valve actuator using neural networks with fuzzy capabilities

    Proc. Eur. Symp. Art. Neural Networks

    (2002)
  • F.J. Uppal, R.J. Patton, V. Palade, Neuro-fuzzy based fault diagnosis applied to an electro-pneumatic valve, in:...
  • Cited by (28)

    • Analysis of mechanical behaviour and fracture interaction of multi-hole rock mass with DIC measurement

      2022, Measurement: Journal of the International Measurement Confederation
      Citation Excerpt :

      In addition, the axial deformation was recorded by a linear variable differential transformer (LVDT) installed beside the rock specimen. Computer Technology has been widely used for failure detection [29,30]. Digital imaging processing is a powerful method in terms of the determination of the deformation and damage of specimens [31].

    • Local linear embedding algorithm of mutual neighborhood based on multi-information fusion metric

      2021, Measurement: Journal of the International Measurement Confederation
      Citation Excerpt :

      However, due to the complexity of some high-dimensional datasets, the local region cannot be effectively divided, leading to ambiguous geometric structure, which makes the manifold learning algorithm fail. To solve the above problems, a large number of relevant achievements have been proposed [20,21]. He presented a LLE algorithm with homogenized distance (H-LLE), which adjusted by the uniformized distance to make the sample distribution tend to be uniformized, thus effectively dividing the local region and improving the performance of LLE algorithm [22].

    • A BiGRU method for remaining useful life prediction of machinery

      2021, Measurement: Journal of the International Measurement Confederation
      Citation Excerpt :

      Fault prediction and health management include the following four stages: fault detection, fault diagnosis, RUL prediction and health management [2]. When a fault is found, the machine is usually stopped immediately to avoid serious economic losses, and the failure usually occurs at an inconvenient time [3,4]. Therefore, accurate RUL prediction for effective mechanical predictive maintenance is considered to reduce costly unplanned mechanical maintenance.

    • Online defect recognition of narrow overlap weld based on two-stage recognition model combining continuous wavelet transform and convolutional neural network

      2019, Computers in Industry
      Citation Excerpt :

      The DBN-based sequential pattern recognition method can extract features directly from the original time series signals, and does not require large quantities of professional prior knowledge to assist (Shao et al., 2018a; Chen et al., 2015). The AE-based sequential pattern recognition method uses a stacking encoder composed of multiple automatic encoders to extract feature data from time series data and filter it to obtain low-dimensional features (Lu et al., 2017; Demetgul et al., 2014). In recent years, there are a lot of research on CNN-based sequential pattern recognition, which share two main ideas: The first is to use CNN to extract the characteristics of the original signal for defect or fault identification; the second is to convert the original signal into a two-dimensional image, and then use CNN to extract features for pattern recognition of defects or faults.

    • A deviation based assessment methodology for multiple machine health patterns classification and fault detection

      2018, Mechanical Systems and Signal Processing
      Citation Excerpt :

      P1: Health (performance) assessment It is widely stated in the current literature that different failure modes can be clustered in the feature space [12,22,23], as illustrated in Fig. 4. Therefore, health diagnosis can be implemented by assessing the direction of deviation in an online setting.

    View all citing articles on Scopus
    View full text