A New Accident Investigation Approach Based on Data Mining Techniques

Parhizi, Sh.; Shahrabi, Jamal; Pariazar, M.

doi:10.3923/jas.2009.731.737

Cited by 8 publications

(2 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some mathematical techniques including association rule mining, t-weight calculations, and decision tree analysis are used for the outline study [48]. Similarly, accident investigation based on the different conditions is simulated with factor analysis and actions to avoid certain known situations are proposed [49]. A broad review of financial fraud detection [50] and cybersecurity intrusion detection [51] is executed.…”

Section: Model Developmentmentioning

confidence: 99%

Fault Detection and Prognosis of Aerospace Systems Using Long Short-Term Memory Based Recurrent Neural Networks

Dhanagopal¹

2023

Preprint

View full text Add to dashboard Cite

<div>Health monitoring and remaining useful life predictions for the aerospace systems is a challenging and complex task to accomplish. Internal or external complications in these aerospace systems (aircraft and satellites) may lead to extremely hazardous or catastrophic consequences to the entire mission involving human life and budget. Considering the severity and complexity of the problem, this thesis deals in developing a diagnosis and prognosis health management system (DPHM) for the attitude actuator control system that uses reaction wheels in pyramid configuration onboard Kepler spacecraft and for the fleet of air-breathing turbofan engines. The established model is</div><div>comparatively effective and computationally light in managing the objective of fault detection and prognostics.</div><div>An advanced data-driven DPHM scheme with optimization techniques is developed and evaluated. Initially, a recurrent LSTM (Long Short-Term Memory) neural network model is</div><div>established and assessed with the general dataset (Particulate Matter (PM2.5)). Secondly, a statistical-based fault detection method with functional factors of Weibull and mathematical features of frictional parameters showed that reaction wheels 2 and 4 of Kepler spacecraft have an early sign (~2 months) of their respective failures. This statistical method is compared with the proposed LSTM model for validation. Thirdly, the prognostic approach for estimation of remaining useful life (RUL) of the C-MAPPS and PHM08 datasets is successfully achieved. Numerous preprocessing methods such as digital filters (Savitzky-Golay (S-G)), principal component analysis (PCA) are used for standardizing the data. Finally, the optimization tools such as genetic algorithm (GA) and particle swarm optimization (PSO) are merged with LSTM for finetuning the hyper-parameters. Overall, the optimized model performs with better accuracy and can be concluded as a promising algorithm for the health management of complex systems.</div>

show abstract

Section: Model Developmentmentioning

confidence: 99%

Fault Detection and Prognosis of Aerospace Systems Using Long Short-Term Memory Based Recurrent Neural Networks

Dhanagopal¹

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…A number of AI approaches, including data mining (DM), [10][11][12] Bayesian networks (BNs), [13][14][15] artificial neural networks (ANNs), [9][10][11][12][13][14][15][16][17] and support vector machines (SVMs), 18 have been developed to deal with ISHM system FD problems. For instance, Gebraeel and Lawley 16 proposed a neural network-based degradation model that utilized real-time sensory signals to estimate the failure time of partially degraded components.…”

Section: Introductionmentioning

confidence: 99%

Data mining–based intelligent fault diagnostics for integrated system health management to avionics

Sun

2014

Proceedings of the Institution of Mechanical Engineers, Part O:

View full text Add to dashboard Cite

Space avionics are the essential capabilities of a spacecraft that guarantee space flight safety and mission success. One of the most important elements developed to deal with the health of the space avionics is the integrated system health management. Fault diagnostics, a safety-critical process in the integrated system health management, has become more complex as the number of avionics systems within the spacecraft has grown, so failure data are now multidimensional, often incomplete, and have cumulatively acquired uncertainties. Therefore, an accurate fault diagnostics model is needed to handle these types of data and ensure information is adequately adapted and efficiently updated. To date, there has been little research focused on efficient and effective space avionics fault diagnostics. This article presents a novel integrated system health management–oriented intelligent diagnostics methodology based on data mining. A numerical example is provided to illustrate the methodology, which demonstrates the significant benefits of data mining for the efficient processing of massive, incomplete data, and the ability of using a robust diagnostic Bayesian network to identify faults with uncertainty in a dynamic environment. The combined approach shows how some limitations can be overcome with an improved diagnostic performance. For application, sensory information must initially be discretized to Boolean values. Data mining is then used to mine for useful association rules and to learn the dynamic Bayesian network structure. After parameter training, the diagnostics is conducted. This methodology can be applied to systems of varying sizes and is flexible enough to accommodate other efficient diagnostic methods.

show abstract

Information Fusion

2017

Integrated System Health Management

View full text Add to dashboard Cite

A New Accident Investigation Approach Based on Data Mining Techniques

Cited by 8 publications

References 12 publications

Fault Detection and Prognosis of Aerospace Systems Using Long Short-Term Memory Based Recurrent Neural Networks

Fault Detection and Prognosis of Aerospace Systems Using Long Short-Term Memory Based Recurrent Neural Networks

Data mining–based intelligent fault diagnostics for integrated system health management to avionics

Information Fusion

Contact Info

Product

Resources

About