Fault Diagnosis Improvement Using Dynamic Fault Model in Optimal Sensor Placement: A Case Study of Steam Turbine

Validity of sensor data is a challenge in system monitoring due to stochastic nature of failure occurrence. The quantity and location of sensors affect the system health information, while sensor malfunction causes misleading results about the system condition. Occurred economic losses are irrecoverable expenses in respect to the monitoring system as well as the system failure. In this study, a dual index approach is proposed for the determination of sensor placement scenarios based on two criteria: (1) uncertainty of sensor information and (2) risk of sensor failure. With about variation of environmental factors conditions (e.g. temperature) and their failure threshold characterization, system failure model is developed and analyzed by a proposed efficient Monte Carlo simulation. Statistical variance of sensor information about estimating of system state as the quantitative uncertainty measure of choice in this research is estimated according to the information value that each possible sensor placement scenario provides through sensor information. In the next phase, risk index is determined based on sensor malfunction and corresponding quantifiable losses through both failure costs and maintenance expenditure. All feasible combinations of sensor failures are considered in the risk model. Finally, a combinatorial criterion is determined through information entropy calculation, which considers both indexes proposed above simultaneously. Sensor placement scenarios are comparatively prioritized based on this criterion. Accordingly, the technical directions are provided for suitability of the criteria for prioritizing sensor arrangement in various systems with different reliability-based characteristics. As a case study, determination of sensor placement is demonstrated on a typical steam turbine. According to the low variation of both information uncertainty and risk indexes, it is concluded that the combinatorial index is the proper criterion for sensor placement determination in health monitoring process of the steam turbine.

show abstract

“…In this step, all combinations of failure-causes are developed to form the SVs. 16,17 The i th SV is written as equation (1)…”

Section: Uncertainty Of Sensor Informationmentioning

confidence: 99%

Sensor placement determination in system health monitoring process based on dual information risk and uncertainty criteria

Salehpour-Oskouei

Pourgol-Mohammad

2017

Proceedings of the Institution of Mechanical Engineers, Part O:

Self Cite

View full text Add to dashboard Cite

show abstract

“…The scenario that sensors malfunction is also considered in his research to analyze the effect of sensor failure on system condition monitoring. A dynamic gate, called the priority AND (PAND) gate, is then introduced to evaluate the sequence of sensor failure and component failure [11]. In order to deal with the epistemic uncertainty, a dual index approach is proposed by taking advantage of statistical variance of sensor readings.…”

Section: Introductionmentioning

confidence: 99%

Locating Sensors in Complex Engineering Systems for Fault Isolation Using Population-Based Incremental Learning

2020

View full text Add to dashboard Cite

Fault diagnostics aims to locate the origin of an abnormity if it presents and therefore maximize the system performance during its full life-cycle. Many studies have been devoted to the feature extraction and isolation mechanisms of various faults. However, limited efforts have been spent on the optimization of sensor location in a complex engineering system, which is expected to be a critical step for the successful application of fault diagnostics. In this paper, a novel sensor location approach is proposed for the purpose of fault isolation using population-based incremental learning (PBIL). A directed graph is used to model the fault propagation of a complex engineering system. The multidimensional causal relationships of faults and symptoms were obtained via traversing the directed path in the directed graph. To locate the minimal quantity of sensors for desired fault isolatability, the problem of sensor location was firstly formulated as an optimization problem and then handled using PBIL. Two classical cases, including a diesel engine and a fluid catalytic cracking unit (FCCU), were taken as examples to demonstrate the effectiveness of the proposed approach. Results show that the proposed method can minimize the quantity of sensors while keeping the capacity of fault isolation unchanged.

show abstract

“…21 presented an optimal sensor placement method for power systems health monitoring, and dynamic Bayesian belief network was used to model the dynamic failure behaviors. A dynamic fault model 22 was constructed to evaluate the influence of time dependencies between sensors and components on sensor placement. For the CCF problem, many researchers proposed some effective models such as the α -factor model, 23 the β -factor model, 24 the multi Greek letter (MGL) model, 25 and the multiple error shock model (MESH).…”

Section: Introductionmentioning

confidence: 99%

“…Based on this research, Farzin et al. 22 used a priority AND gate (PAND) to build a sensor diagnostic model considering the effect of time dependencies between sensors and components on the optimal sensor placement. On this basis, Duan et al.…”

Section: Introductionmentioning

confidence: 99%

Optimal sensor placement based on multiattribute decision-making considering the common cause failure

Feng

Duan

Lin

et al. 2019

Proceedings of the Institution of Mechanical Engineers, Part C:

View full text Add to dashboard Cite

A new optimal sensor placement is developed to improve the efficiency of fault diagnosis based on multiattribute decision-making considering the common cause failure. The optimal placement scheme is selected based on the reliability of the top event on condition that the number of sensors is preset. Specifically, a β-factor model is introduced to deal with the common cause failure, and dynamic fault tree is used to describe the dynamic failure behaviors. Besides, a dynamic fault tree is converted into a dynamic Bayesian network to calculate the reliability parameters, which construct the decision matrix. Furthermore, an efficient TOPSIS algorithm is adopted to determine the potential locations of sensors. In addition, a diagnostic sensor model is developed to take into account the failure sequence between a sensor and a component using a priority AND gate, and the failure probability of the top event for all sensor placement scenarios is calculated to determine the optimal sensor placement. Finally, a case is provided to prove that the common cause failure has made a considerable impact on the sensor placement.

show abstract

Fault Diagnosis Improvement Using Dynamic Fault Model in Optimal Sensor Placement: A Case Study of Steam Turbine

Cited by 19 publications

References 30 publications

Sensor placement determination in system health monitoring process based on dual information risk and uncertainty criteria

Sensor placement determination in system health monitoring process based on dual information risk and uncertainty criteria

Locating Sensors in Complex Engineering Systems for Fault Isolation Using Population-Based Incremental Learning

Optimal sensor placement based on multiattribute decision-making considering the common cause failure

Contact Info

Product

Resources

About