Analysis and Design of Observer-Based Fault Detection Systems

This paper presents a new approach to the design of robust fault detection systems via a genetic algorithm. To achieve robustness, a number of performance indices are introduced, which are expressed in the frequency domain to account for the frequency distributions of incipient faults, noise and modelling uncertainty. All objectives are then reformulated into a set of inequality constraints on performance indices. A genetic algorithm is thus used to search an optimal solution to satisfy these inequality constraints. The approach developed is applied to a flight control system example and results show that incipient sensor faults can be detected reliably in the presence of modelling uncertainty.

Download Full-text

Time domain optimization of fault detection systems

Proceedings of 35th IEEE Conference on Decision and Control ◽

10.1109/cdc.1996.574392 ◽

2002 ◽

Cited By ~ 1

Author(s):

X. Ding ◽

L. Guo

Keyword(s):

Fault Detection ◽

Time Domain ◽

Domain Optimization ◽

Detection Systems

Download Full-text

Metrics for Evaluating the Accuracy of Diagnostic Fault Detection Systems

Volume 2: Turbo Expo 2004 ◽

10.1115/gt2004-54144 ◽

2004 ◽

Cited By ~ 5

Author(s):

Hans DePold ◽

Jason Siegel ◽

Jon Hull

Keyword(s):

Decision Support ◽

Fault Detection ◽

Decision Support System ◽

Support System ◽

Diagnostic Tests ◽

Design Sensitivity ◽

Classifier Systems ◽

Operating Characteristics ◽

Gas Generator ◽

Detection Systems

This paper presents a method for providing metrics to evaluate the accuracy and cost effectiveness of diagnostic decision support systems. One intention of engine health management (EHM) fault detection systems is to have engines identified for removal and refurbishment as soon as there is evidence of an adverse gas generator trend shift. The benefits of EHM diagnostics and prognostics tests are derived from the resulting improved safety, the reduced operating costs, and most importantly, the good will and trust of the customer. The method presented in this paper is a generalized way of evaluating the performance of some of the tests that are used to make inspection, removal, and maintenance decisions [Ref 1,2]. The detection of faults from shifts in classification data is the first step in EHM systems that use diagnostics and prognostics [Ref 3,4,5]. The minimum parameter shift required to trigger a fault indication is called the threshold. Typically, it is a predetermined multiple of the standard deviation of the parameter measurements. Root cause isolation is usually invoked following these detection tests for the gas path parameter shifts. This paper shows how the achievable accuracy of diagnostic and prognostic system tests can be determined from the signal to noise ratio (SNR), and the system’s design (sensitivity and specificity). From these tests we extract two features, true positives (TP) and false positives (FP) that can be used to compare the accuracy of any simple or complex decision support system. This method is conducive to efficiently handling large amounts of data from multiple sensor tests because it avoids explicit correlation among individual diagnostic tests, and focuses instead on the net results. Each piece of classification information is used to reduce ambiguity. In this approach, the individual diagnostic tests and any data fusion weighting factors can be parametrically varied to optimize the accuracy of the decisions. The resulting plot of TP versus FP is then directly compared to the results of simple idealized classifier systems having known SNRs. This paper applies the receiver operating characteristics (ROC) process to evaluate the potential accuracy of EHM decisions. The paper also shows that the actual accuracy depends on how thresholds are set, and on the local shape of the ROC in the regions where it is used. The method presented can be applied to test the relative accuracy of each phase of the EHM decision-making process. The effects of test accuracies, event probabilities, and consequential event costs on the value of the decision support system are also presented.

Download Full-text

Towards to an Embedded Edge AI Implementation for Longitudinal Rip Detection in Conveyor Belt

10.5753/sbesc_estendido.2020.13096 ◽

2020 ◽

Author(s):

Emerson Klippel ◽

Ricardo Oliveira ◽

Dmitry Maslov ◽

Andrea Bianchi ◽

Saul Emanuel Silva ◽

...

Keyword(s):

Deep Learning ◽

Fault Detection ◽

Conveyor Belt ◽

Complex Problem ◽

Operational Conditions ◽

Detection Systems ◽

Conveyor Belts ◽

Maintenance Personnel ◽

Deep Learning Model ◽

Financial Losses

The use of deep learning on edge AI to detect failures in conveyor belts solves a complex problem of iron ore beneficiation plants. Losses in the order of thousands of dollars are caused by failures in these assets. The existing fault detection systems currently do not have the necessary efficiency and complete loss of belts is common. Correct fault detection is necessary to reduce financial losses and unnecessary risk exposure by maintenance personnel. This problem is addressed by the present work with the training of a deep learning model for detecting images of failures of the conveyor belt. The resulting model is converted and executed in an edge device located near the conveyor belt to stop it in case a failure is detected. The results obtained in the development and tests carried out to date show the feasibility of using Edge AI to solve complex problems in a mining environment such as detecting longitudinal rips and stimulate the continuity of the work considering new scenarios and operational conditions in the search for a robust and replicable solution.

Download Full-text