A neural network based adaptive fault detection scheme

Sreedhar, R.; Fernandez, B.; Masada, Glenn Y.

doi:10.1109/acc.1995.532205

Cited by 12 publications

(6 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Generally neural networks are used to create a black-box model of the nominal system [34] to capture the possible fault signature(s) in the system [35]. An adaptive neural-network-based fault detection technique in nonlinear systems is presented in [36]. Liu and Scherpen presented a different fault detection technique for nonlinear systems based on probabilistic neural network filtering [13].…”

Section: A2 Neural Network Based Methodsmentioning

confidence: 99%

Review and comparative evaluation of symbolic dynamic filtering for detection of anomaly patterns

Rao

Ray

Sarkar

et al. 2008

SIViP

107

View full text Add to dashboard Cite

Symbolic dynamic filtering (SDF) has been recently reported in literature as a pattern recognition tool for early detection of anomalies (i.e., deviations from the nominal behavior) in complex dynamical systems. This paper presents a review of SDF and its performance evaluation relative to other classes of pattern recognition tools, such as Bayesian Filters and Artificial Neural Networks, from the perspectives of: (i) anomaly detection capability, (ii) decision making for failure mitigation and (iii) computational efficiency. The evaluation is based on analysis of time series data generated from a nonlinear active electronic system.

show abstract

Section: A2 Neural Network Based Methodsmentioning

confidence: 99%

Review and comparative evaluation of symbolic dynamic filtering for detection of anomaly patterns

Rao

Ray

Sarkar

et al. 2008

SIViP

107

View full text Add to dashboard Cite

show abstract

“…By using this method, a good level of control was achieved by decoupling the hotwell and the deaerator. Sreedhar et al (1995a) presented a fault detection scheme based on the adaptive neural network for the non-linear system. Then, this method was validated for the thermal power plant.…”

Section: Review Of Control Of Boilers In a Thermal Power Plantmentioning

confidence: 99%

Modeling, identification, and control of coal-fired thermal power plants

Chandrasekharan

Panda

Swaminathan³

2014

Reviews in Chemical Engineering

View full text Add to dashboard Cite

The major portion of total captive power comes from thermal power from coal, oil, natural gas, etc., of which coal-fired units contribute 41% of global electricity. The overall efficiency of such power plants is 33% due to variations in power plant design, use of wide fuel ranges, different steam turbines, and lack in modeling and control configurations. To improve this efficiency, knowledge on material and energy flows across subsystems, including fuel handling, boiler, turbine, and air management need to be known properly. In addition, there is a widespread lack in measurement and instrumentation systems in power plants across the fleet and operators. This scenario makes the operational adjustments more difficult. Again, the strictness in environmental compliances (sulfur and nitrogen components) is met at the cost of limiting thermal efficiency. Moreover, to comply with faster load change, less fuel consumption and more efficient supercritical boilers have been studied. Hence, to improve the efficiency level, more research starting from the state-ofthe-art review on modeling, identification, and control methods of coal-fired boilers and the integrated plant is aimed in this work. This will also address the issue of global climate change and reduction in emission of greenhouse gases.

show abstract

“…The models, as described by the ANN structure and corresponding weights, can then be used for state estimation, model predictive control, path planning, internal model control, or any other control techniques that may require an accurate, frequently updated model of the plant. Fault detection during operation of the vehicle is also possible using this technique as described in [3], [4], [5].…”

Section: Introductionmentioning

confidence: 99%

An adaptive non-linear state estimator for vehicle lateral dynamics

Broderick

Bevly

Hung

2009

2009 35th Annual Conference of IEEE Industrial Electronics

View full text Add to dashboard Cite

Artificial neural networks are used to estimate side slip angle and yaw rate of a vehicle's lateral dynamics. The networks are adapted to varying operating conditions such as a shift in vehicle weight, a change in road surface, and a radical change in tire characteristics. The structure and characteristics of the networks used are detailed. The methods for both offline and online training are described. Adaptation to the changing conditions is investigated with a high fidelity model and evaluated for ability and accuracy. A method of reducing computational burden while preserving model generalization is described. Model accuracy and generalization are examined to evaluate the networks' ability to describe general vehicle behavior. Improvement in estimate error of 3 to 1 and nearly 300 to 1 for two typical scenarios is demonstrated.

show abstract

A neural network based adaptive fault detection scheme

Cited by 12 publications

References 7 publications

Review and comparative evaluation of symbolic dynamic filtering for detection of anomaly patterns

Review and comparative evaluation of symbolic dynamic filtering for detection of anomaly patterns

Modeling, identification, and control of coal-fired thermal power plants

An adaptive non-linear state estimator for vehicle lateral dynamics

Contact Info

Product

Resources

About