Integrated model-based and data-driven fault detection and diagnosis approach for an automotive electric power steering system

Hwangbo

International Journal of Distributed Sensor Networks

2018

Connected cars, which are vehicles connected to wireless networks through the convergence of automotive and information technologies, have become an important topic of academic and industrial research on automobiles. In this research, we conducted a field experiment to understand vehicle maintenance mechanisms of a connected car platform. Specifically, we investigated the feasibility of prognostics and health management under different driving circumstances, with varying vehicle models, vehicle conditions, drivers' propensity for speeding, and road conditions. We collected sensor data through a two-stage model of vehicle communication using an on-board diagnostics scanner and data transmission using wireless communication. We found that device defects can be predicted based on driving situations such as the driving mode, mechanical characteristics, and a driver's speeding propensity.

Section: Authorsmentioning

confidence: 99%

An empirical study on real-time data analytics for connected cars: Sensor-based applications for smart cars

Hwangbo

International Journal of Distributed Sensor Networks

2018

IEEE Trans. Comput.-Aided Des. Integr. Circuits Syst.

“…Several methods have been proposed to target the diagnosis of system-level failures [12]- [14]. In [12], the authors proposed a hybrid model-based and data-driven approach to fault detection and diagnosis in an electric power steering system. However, this method needs to conduct timeconsuming fault injection experiments to derive fault-sensor measurement dependences.…”

Section: Introductionmentioning

confidence: 99%

Diagnosis of Board-Level Functional Failures Under Uncertainty Using Dempster–Shafer Theory

Fang

Chakrabarty

Wang

et al. 2012

Despite recent advances in structural test methods, the diagnosis of the root cause of board-level failures for functional tests remains a major challenge. A promising approach to address this problem is to carry out fault diagnosis in two phases-suspect faulty components on the board or modules within components (together referred to as blocks in this paper) are first identified and ranked, and then fine-grained diagnosis is used to target the suspect blocks in a ranked order. We propose a new method based on dataflow analysis and DempsterShafer (DS) theory for ranking faulty blocks in the first phase of diagnosis. The proposed approach transforms the information derived from one functional test failure into multiple-stage failures by partitioning the given functional test into multiple stages. A measure of "belief" is then assigned to each block based on the knowledge of each failing stage, and the DS theory is subsequently used to aggregate the beliefs from multiple failing stages. Blocks with higher beliefs are ranked on the top of the candidate list. Simulations on an industry design for a network interface application as well as on an open source system-ona-chip show that the proposed method can provide accurate ranking for most board-level functional failures. Index Terms-Board-level, dataflow analysis, Dempster-Shafer (DS) theory, diagnosis, functional failure.

Proceeding of the 11th World Congress on Intelligent Control and Automation

“…The physics-modeling approaches employs consistency checks between the sensed measurements and the outputs of a mathematical model [1,2]. The expectation is that inconsistencies are large in the presence of malfunctions and small in the presence of normal disturbances, noise and modeling errors [2,3], and two main methods for generating the residuals are based on observers and parity relations, such as Kalman filters [4,5,6], reduced-order unknown input observers [5], interacting multiple models [6], particle filters [7], etc.…”

Section: Introductionmentioning

confidence: 99%

An incremental FGRA-based fault diagnosis method

Zhang

Liu

et al. 2014

In order to effectively solve the uncertainty smallsamples fault diagnosis problem, a practical data-driven greybased fault detection and diagnosis (FDD) method for complex equipments is investigated. Firstly, an improved fuzzy-grey relational analysis (FGRA) technique is proposed by introducing dynamic identification coefficient and fuzzy relational weight. Compared with the traditional Deng's grey relational analysis (DGRA) technique, the proposed FGRA technique not only can strengthen the veracity and reliability but also can reduce the dependence of uncertain man-made identification and weight coefficient. Secondly, a simple and practical FGRA-based fault diagnosis process is designed. It belongs to a data-driven analytic method which does not need to consider the either statistic assumptions or distributions of diagnosis variables. Finally, the validity and practicability of the proposed FGRA-based method is demonstrated by a example of rotor fault diagnosis, and the results show that the proposed method is more effective than the DGRA-based method. Index Terms -fault detection and diagnosis (FDD); fuzzy-grey relational analysis (FGRA); dynamic identification coefficient; fuzzy relational weight; diesel engine.