LFT Bond Graph Model-Based Robust Fault Detection and Isolation

2011 ◽  
pp. 105-133 ◽  
Author(s):  
M.A. Djeziri ◽  
B. Ould Bouamama ◽  
G. Dauphin-Tanguy ◽  
R. Merzouki
Keyword(s):  
Fault Detection ◽  
Graph Model ◽  
Bond Graph ◽  
Fault Detection And Isolation ◽  
Model Based ◽  
Robust Fault Detection

Bond graph model-based fault detection using residual sinks

2008 ◽  
Vol 223 (3) ◽  
pp. 337-352 ◽  
Author(s):  
W Borutzky
Keyword(s):  
Fault Detection ◽  
Numerical Computation ◽  
Graph Model ◽  
Numerical Differentiation ◽  
Equation System ◽  
Bond Graph ◽  
Engineering System ◽  
Differential Algebraic Equation ◽  
Symbolic Form ◽  
Model Based

In this paper, residual sinks are used in bond graph model-based quantitative fault detection for the coupling of a model of a faultless process engineering system to a bond graph model of the faulty system. By this way, integral causality can be used as the preferred computational causality in both models. There is no need for numerical differentiation. Furthermore, unknown variables do not need to be eliminated from power continuity equations in order to obtain analytical redundancy relations (ARRs) in symbolic form. Residuals indicating faults are computed numerically as components of a descriptor vector of a differential algebraic equation system derived from the coupled bond graphs. The presented bond graph approach especially aims at models with non-linearities that make it cumbersome or even impossible to derive ARRs from model equations by elimination of unknown variables. For illustration, the approach is applied to a non-controlled as well as to a controlled hydraulic two-tank system. Finally, it is shown that not only the numerical computation of residuals but also the simultaneous numerical computation of their sensitivities with respect to a parameter can be supported by bond graph modelling.

Model-based robust fault detection and isolation of an industrial gas turbine prototype using soft computing techniques

2012 ◽  
Vol 91 ◽  
pp. 29-47 ◽  
Author(s):  
Hasan Abbasi Nozari ◽  
Mahdi Aliyari Shoorehdeli ◽  
Silvio Simani ◽  
Hamed Dehghan Banadaki
Keyword(s):  
Fault Detection ◽  
Gas Turbine ◽  
Soft Computing ◽  
Fault Detection And Isolation ◽  
Model Based ◽  
Robust Fault Detection ◽  
Soft Computing Techniques ◽  
Industrial Gas Turbine

scholarly journals LFT Bond Graph for Online Robust Fault Detection and Isolation of Hybrid Multi-Source System

2021 ◽  
Vol 2065 (1) ◽  
pp. 012010
Author(s):  
Mahdi Boukerdja ◽  
Youness Radi ◽  
Omprakash ◽  
Sumit Sood ◽  
Belkacem Ould-Bouamama ◽  
...  
Keyword(s):  
Fault Detection ◽  
Online Monitoring ◽  
Detection Algorithm ◽  
Bond Graph ◽  
Fault Detection And Isolation ◽  
False Alarms ◽  
Parameter Uncertainties ◽  
Safety Issues ◽  
Robust Fault Detection ◽  
Production Of Hydrogen

Abstract Green hydrogen is undoubtedly the most promising energy vector of the future because it is captured by renewable and inexhaustible sources, such as wind and/or solar energy, and can be stored over the long in high-pressure cylinders, which can be used to feed the fuel cells to produce the electricity without emitting any pollutants. The system incorporated renewable sources and process used to produce the green hydrogen is the hybrid multi-source system (HMS). The production of hydrogen needs a reliable HMS, which always requires online monitoring for real-time Fault Detection and Isolation (FDI) because the risk of accidents in HMS and safety issues increases due to the possibility of faults. However, online monitoring of FDI is challenging due to the multi-physics dynamics of HMS and the inclusion of uncertain parameters and several disturbances. This paper proposes an online robust fault detection algorithm to detect system faults based on the properties of the graphical linear fractional transformation bond graph (LFT-BG) modeling approach. Here, the analytical redundancy relations (ARRs) and their uncertain parts extracted from the LFT-BG model are used to develop an online robust FDI algorithm for HMS. Numerical evaluations of ARRs and their uncertain parts, respectively, generate the residual signals known as “faults indicators” and their uncertain bounds known as “adaptive thresholds.” These thresholds evolve with system variables in the presence of parameter uncertainties for ensuring robust FDI for HMS to minimize false alarms. The validation of this approach is carried out using 20sim software that is familiar with BG modeling.

Optimized fault detection using bond graph in linear fractional transformation form

2021 ◽  
pp. 095965182098561
Author(s):  
Mahdi Ouziala ◽  
Youcef Touati ◽  
Sofiane Berrezouane ◽  
Djamel Benazzouz ◽  
Belkacem Ouldbouamama
Keyword(s):  
Fault Detection ◽  
Detection System ◽  
Main Idea ◽  
Bond Graph ◽  
Fault Detection And Isolation ◽  
Linear Fractional Transformation ◽  
Estimated Parameters ◽  
Robust Fault Detection ◽  
Uncertainty Calculation ◽  
Adaptive Thresholds

This article deals with the optimal robust fault detection problem using the bond graph in its linear fractional transformation form. Generally, this form of the bond graph allows the generation of two perfectly separate analytical redundancy relations, that are used as residual and threshold. However, the uncertainty calculation method gives overestimated thresholds. This may, for instance, lead to undetectable faults. Therefore, enhancing the robustness of fault detection and isolation algorithms is of utmost importance in designing a bond graph–based fault detection system. The main idea of this article is to develop optimized thresholds to ensure an optimal detection, otherwise this article proposes a method to detect tiny magnitude faults concerning parameter’s uncertainties. This work considers the issue of optimal fault detection as an optimization problem of the gap between the residuals and its threshold. New uncertainty values will be calculated in a way that these estimated parameters ensure the desired optimized gap between residuals and thresholds. These estimated uncertainty values will be used to generate optimized adaptive thresholds. Through these thresholds, we increase the sensitivity of the residuals to tiny magnitude faults, and we ensure an optimal and early detection.

Sign in / Sign up

Export Citation Format

Share Document