Article

Fault detection and diagnosis of an industrial steam turbine using fusion of SVM (support vector machine) and ANFIS (adaptive neuro-fuzzy inference system) classifiers

Department of Instrumentation and Automation, Petroleum University of Technology, Tehran, Iran; Department of Control Engineering, Islamic Azad University South Tehran branch, Iran
Energy 01/2010; DOI: 10.1016/j.energy.2010.06.001

ABSTRACT The subject of FDD (fault detection and diagnosis) has gained widespread industrial interest in machine condition monitoring applications. This is mainly due to the potential advantage to be achieved from reduced maintenance costs, improved productivity and increased machine availability. This paper presents a new FDD scheme for condition machinery of an industrial steam turbine using a data fusion methodology. Fusion of a SVM (support vector machine) classifier with an ANFIS (adaptive neuro-fuzzy inference system) classifier, integrated into a common framework, is utilized to enhance the fault detection and diagnostic tasks. For this purpose, a multi-attribute data is fused into aggregated values of a single attribute by OWA (ordered weighted averaging) operators. The simulation studies indicate that the resulting fusion-based scheme outperforms the individual SVM and ANFIS systems to detect and diagnose incipient steam turbine faults.

0 Bookmarks
 · 
226 Views
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Resiliency and improved state-awareness of modern critical infrastructures, such as energy production and industrial systems, is becoming increasingly important. As control systems become increasingly complex, the number of inputs and outputs increase. Therefore, in order to maintain sufficient levels of state-awareness, a robust system state monitoring must be implemented that correctly identifies system behavior even when one or more sensors are faulty. Furthermore, as intelligent cyber adversaries become more capable, incorrect values may be fed to the operators. To address these needs, this paper proposes a fuzzy-neural data fusion engine (FN-DFE) for resilient state-awareness of control systems. The designed FN-DFE is composed of a three-layered system consisting of: 1) traditional threshold based alarms; 2) anomalous behavior detector using self-organizing fuzzy logic system; and 3) artificial neural network-based system modeling and prediction. The improved control system state-awareness is achieved via fusing input data from multiple sources and combining them into robust anomaly indicators. In addition, the neural network-based signal predictions are used to augment the resiliency of the system and provide coherent state-awareness despite temporary unavailability of sensory data. The proposed system was integrated and tested with a model of the Idaho National Laboratory's hybrid energy system facility known as HYTEST. Experiment results demonstrate that the proposed FN-DFE provides timely plant performance monitoring and anomaly detection capabilities. It was shown that the system is capable of identifying intrusive behavior significantly earlier than conventional threshold-based alarm systems.
    IEEE transactions on cybernetics. 10/2014; 44(11):2065-2075.
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Fault detection and diagnosis have an effective role for the safe operation and long life of systems. Condition monitoring is an appropriate way of the maintenance technique that is applicable in the fault diagnosis of rotating machinery faults. A unique flexible algorithm is proposed for classifying the condition of centrifugal pump based on support vector machine hyper-parameters optimization and artificial neural networks (ANNs) which are composed of eight distinct steps. Artificial neural networks (ANNs), support vector classification with genetic algorithm (SVC-GA) and support vector classification with particle swarm optimization (SVC-PSO) algorithm have been considered in a flexible algorithm to perform accurate classification in the manufacturing area. SVC-GA, SVC-PSO and ANN have been used together due to their importance and capabilities in classifying domain. Also, the superiority of the proposed hybrid algorithm (SVC with GA and PSO) is shown by comparing its results with SVC performance. Two types of faults through six features, flow, temperature, suction pressure, discharge pressure, velocity, and vibration, have been classified with proposed integrated algorithm. To test the robustness of the efficiency results of the proposed method, the ability of proposed flexible algorithm in dealing with noisy and corrupted data is analyzed.
    Applied Soft Computing. 03/2013; 13(3):1478–1485.
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: A new fault tolerant control (FTC) problem via the output probability density functions (PDFs) for non-Gaussian stochastic distribution control systems (SDC) is investigated. The PDFs can be approximated by the radial basis functions (RBFs) of neural networks. Differently from the conventional FTC problems, the measured information is in the form of probability distributions of the system output rather than the actual output values. The control objective is to use the output PDFs to design control algorithm that can compensate the faults and attenuate the disturbances. As a result, the concerned FTC problem subject to dynamic relation between the input and output PDFs can be transformed into a nonlinear FTC problem subject to dynamic relation between the control input and the weights of the RBFs neural networks. Feasible criteria to compensate the faults and attenuate the disturbances are provided in terms of linear matrix inequality (LMI) techniques. In order to improve FTC performances, H ∞ optimization techniques are applied to the FTC design problem to assure that the faults can be compensated and the disturbances can be attenuated. At last, an illustrated example is given to demonstrate the efficiency of the proposed algorithm, and the satisfactory results have been obtained.
    Circuits Systems and Signal Processing 32(1). · 0.98 Impact Factor