Wyniki wyszukiwania - BazTech

1

New approach of model based detection of early stages of fuel injector failures

Mączak Jędrzej, Więcławski Krzysztof, Szczurowski Krzysztof

Eksploatacja i Niezawodność

|

2023

|

Vol. 25, no. 1

art. no. 6

EN

The aim of the work was to develop a method of real time diagnosing electromagnetic fuel injectors using the observation of electric current parameters available in the engine control unit. Performing this task required finding a precise criterion for assessing the correct operation of an electromagnetic injector. For this purpose, a mathematical model describing the individual phases of the injector's operation was used, allowing the simulation of the occurrence of typical failures. On its basis, symptoms of particular failures were determined based on the observation of electric current parameters in the control circuit. Observation of voltage and current waveforms allows to locate both electrical and mechanical damages to the injectors and to assess the correctness of the power system components. The presented diagnostic method allows the detection of the described damages in the early stages of their development, which prevents damage to the catalytic converter and other engine systems (valves, piston rings or cylinder surfaces), i.e. damages resulting from an incorrect fuel mixture.

2

A correlation matrix-based tensor decomposition method for early prediction of sepsis from clinical data

Nesaragi Naimahmed, Patidar Shivnarayan, Thangaraj Veerakumar

Biocybernetics and Biomedical Engineering

|

2021

|

Vol. 41, no. 3

1013--1024

EN

Early detection of sepsis can assist in clinical triage and decision-making, resulting in early intervention with improved outcomes. This study aims to develop a machine learning framework to predict the onset of sepsis through EHR data by applying tensor decomposition on correlation matrices of clinical covariates for every record, arranged on an hourly basis for the length of stay (LOS) in intensive care unit. A third-order tensor [...] representing a clinical correlation among selected 24 covariates for a considered time frame of sepsis onset duration of 6 h, with a stride of 1 h is formed for each record. Such a fused tensor with dimensions [...] for every record undergoes Tucker decomposition with an optimal choice of rank. The factor matrices U1; U2; U3 thus obtained after decomposition are excluded and only the core tensor r with a dimension [...] has been retained, and used to provide latent features for prediction of sepsis onset. A five-fold cross-validation scheme is employed wherein the obtained 100 latent features from the reshaped core tensor, are fed to Light Gradient Boosting Machine Learning models (LightGBM) for binary classification, further alleviating the involved class imbalance. The machine-learning framework is designed via Bayesian optimization, yielding an average normalized utility score of 0.4314 on publicly available PhysioNet/Computing in Cardiology Challenge 2019 training data. The proposed tensor decomposition deciphers the higher-order interrelations among the considered clinical covariates for early prediction of sepsis and the results obtained are on par with existing state-of-the-art performances.

3

Open-circuit fault diagnosis in three-phase induction motor using model-based technique

Aswad Raya A. K., Jassim Bassim M. H.

Archives of Electrical Engineering

|

2020

|

Vol. 69, nr 4

815--827

EN

The presence of an open-circuit fault subjects a three-phase induction motor to severely unbalanced voltages that may damage the stator windings consecutively causing total shutdown of systems. Unplanned downtime is very costly. Therefore, fault diagnosis is essential for making a predictive plan for maintenance and saving the required time and cost. This paper presents a model-based diagnosis technique for diagnosing an open-circuit fault in any phase of a three-phase induction motor. The proposed strategy requires only current signals from the faulty machine to compare them with the healthy currents from an induction motor model. Then the errors of comparison are used as an objective function for a genetic algorithm that estimates the parameters of a healthy model, which they employed to identify and localize the fault. The simulation results illustrate the behaviours of basic parameters (stator and rotor resistances, self-inductances, and mutual inductance) and the number of stator winding turn parameters with respect to the location of an open-circuit fault. The results confirm that the number of stator winding turns are the useful parameters and can be utilized as an identifier for an open-circuit fault. The originality of this work is in extracting fault diagnosis features from the variations of the number of stator winding turns.

4

A new approach to multiple fault diagnosis: A combination of diagnostic matrices, graphs, algebraic and rule-based models. The case of two-layer models

Ligęza A., Kościelny J. M.

International Journal of Applied Mathematics and Computer Science

|

2008

|

Vol. 18, no 4

465-476

EN

The diagnosis of multiple faults is significantly more difficult than singular fault diagnosis. However, in realistic industrial systems the possibility of simultaneous occurrence of multiple faults must be taken into account. This paper investigates some of the limitations of the diagnostic model based on the simple binary diagnostic matrix in the case of multiple faults. Several possible interpretations of the diagnostic matrix with rule-based systems are provided and analyzed. A proposal of an extension of the basic, single-level model based on diagnostic matrices to a two-level one, founded on causal analysis and incorporating an OR and an AND matrix is put forward. An approach to the diagnosis of multiple faults based on inconsistency analysis is outlined, and a refinement procedure using a qualitative model of dependencies among system variables is sketched out.

5

Model-basis diagnosis of dynamic systems

Górny B., Ligęza A.

Journal of Applied Computer Science

|

2002

|

Vol. 10, nr 1

23-46

EN

Consistency-based diagnosis falls into the category of model based approaches. In this paper the problem of generation of conflict sets for dynamic systems described with causal graphs is approached. Conflict generation is an important step in model-based diagnosis. The foundations of such a diagnostic approach are constituted by the Reiter's theory concerning consistency-based diagnostic reasoning. This paper is aimed at presenting a discussion of selected problems of conflict generation. A notion of Potential Conflict Structure (PCS) constituting a basic tool for identification of possible conflicts is proposed and its use id discussed. The paper includes elements of algebra of conflict sets.