Using PCA in Acoustic Emission Condition Monitoring to Detect Faults in an Automobile Engine (original) (raw)
Related papers
Automobile engine condition monitoring using sound emission
TURKISH JOURNAL OF ELECTRICAL ENGINEERING & COMPUTER SCIENCES
A wavelet packet transform (WPT) is a well-known technique used for data and signal-processing that has proven to be successful in condition monitoring and fault diagnosis. In this study, using feature extraction based on wavelet transformation, sound signals emitted from automobile engines under both faulty and healthy conditions are analyzed. The intention is to categorize sound signals into both healthy and faulty classes. Sound signals are generated from 4 different automobile engines in both healthy and faulty conditions. The investigated fault is within the ignition system of the engines. In addition, there are other possible problems that may also affect the generated sound signals. In the reported study, a set of features is initially extracted from the recorded signals. The more informative features are later selected using a correlation-based feature selection (CFS) algorithm. Results prove the efficiency of wavelet-based feature extraction for the case study of the reported work.
Acoustic-Based Engine Fault Diagnosis Using WPT, PCA and Bayesian Optimization
Applied Sciences, 2020
Engine fault diagnosis aims to assist engineers in undertaking vehicle maintenance in an efficient manner. This paper presents an automatic model and hyperparameter selection scheme for engine combustion fault classification, using acoustic signals captured from cylinder heads of the engine. Wavelet Packet Transform (WPT) is utilized for time–frequency analysis, and statistical features are extracted from both high- and low-level WPT coefficients. Then, the extracted features are used to compare three models: (i) standard classification model; (ii) Bayesian optimization for automatic model and hyperparameters selection; and (iii) Principle Component Analysis (PCA) for feature space dimensionality reduction combined with Bayesian optimization. The latter two models both demonstrated improved accuracy and the other performance metrics compared to the standard model. Moreover, with similar accuracy level, PCA with Bayesian optimized model achieved around 20% less total evaluation time ...
IEEE Transactions on Instrumentation and Measurement, 2011
This work proposes a novel prototype-based engine fault classification scheme employing the audio signature of engines. In this scheme, Fourier transform and correlation methods have been used. Notably, automated audio classification has immense significance in the present times, used in both audio-based content retrieval and audio indexing in multimedia industry. Likewise, it is also becoming increasingly important in automobile industries. It has been observed that real world automobile engine audio data are contaminated with substantial noise and out fliers. Hence, it is challenging to categorize different fault types in different engines. Accordingly, the present paper discusses a methodology where a set of algorithms checks the state of an unknown engine as either healthy or faulty. Fault categorizing algorithm is based on its cross- and autocorrelation coefficient values. Appropriately, in this study, the engine amplitude-frequency values of fast Fourier transform are calculated and subdivided into bands to calculate the correlation coefficient matrix. The correlation coefficient matrix for the unknown engine is then calculated and matched with this “prototype” engine matrix to categorize it into a single or multiple fault(s). It is worth mentioning here that although a rank-based maximum close scheme is adopted for finding the unknown engine's fault, the work can be extended to any other parametric and neural network-based classification scheme. Keeping this background in mind, the present paper discusses the proposed methodology to find a prototype engine, unknown engine classification, implementation on real audio signal for single cylinder engine data, and its results.
Proceeding of the Electrical Engineering Computer Science and Informatics
Improving efficiency and power in an internal combustion engine is always impeded by detonation (knock) problems. This detonation problem has not been explained fully yet. Quick and accurate detection of detonation is also in the development stage. This research used a new method of detonation sound detection which uses microphone sensors, analysis of discrete wavelet transform (DWT), and analysis of the regression function envelope to identify the occurrence of detonation. The engine sound was captured by the microphone; it was recorded on a computer; it was proceeded using a DWT decomposition filtering technique; it was then subjected to normalization and regression function envelope to get the shape of the wave pattern for the vibration. Vibrational wave patterns were then compared to a reference using the Euclidean distance calculation method, in order to identify and provide an assessment decision as to whether or not detonation had occurred. The new method was applied using Matlab and it has yielded results which are quite effective for the detection and identification of detonation and it is also capable of producing an assessment decision about the occurrance of detonation.
Diesel Engine Air-borne Acoustic Signals Analysis Using Continuous Wavelet Transform
This paper studies the characteristics of Diesel engine airborne acoustic signals using time-frequency domain techniques. One analysis technique is investigated: Continuous Wavelet Transform (CWT) which is reviewed from the mathematical point of view, based on its developmental stages, drawbacks and the subsequent improvements. The detection capabilities of this technique are evaluated using airborne acoustic signals collected from diesel engine in acoustically untreated laboratory. Some engine conditions and faults are investigated using CWT techniques. The achieved results prove the technique's sensitivity to engine's speed and load variations. More important, the CWT shows excellent capabilities in detecting engine's injection process and lubrication related faults at early stages. At the end of the paper, summary is given.
Multi-stage Acoustic Fault Diagnosis of Motorcycles using Wavelet Packet Energy Distribution and ANN
2012
Motorcycles generate different sound patterns under dissimilar working conditions. The generated sound pattern gives a clue of the fault. Mainly the parts of the engine that lead to change in sound are cylinder kit, crank, timing chain, and valve. The parts of the exhaust system that change sound under fault are muffler and silencer. In this study, we analyze the sound signals produced by motorcycles to locate the faults in subsystems. The work proceeds in three stages, the first stage detects the fault, the second stage identifies the faulty subsystem and finally the third stage locates the fault. The overall classification accuracy of the first stage is 0.8019. The work finds interesting applications in troubleshooting of machinery, electronic gadgets, musical instruments and the like.
Motor fault detection using sound signature and wavelet transform
International Journal of Power Electronics and Drive Systems (IJPEDS), 2021
The use of induction machines has gained fast popularity in many aspects of today’s energy applications and industrial productions. However, just as with any other machine, failure is expected due to a variety of faults in component and system levels. Therefore, it is necessary to improve machine reliability by performing preventive maintenance and exploring faulty indications in advance to avoid future failures. In normal operation, a distinct machine sound signature can be identify. Therefore, at any faulty operation, diagnosis of potential error can be defined based on output signature sound data analysis. Yet, this process of monitoring induction machine sounds and vibration can be hectic and extensive in terms of collecting data and compiling analysis. That is, a huge number of data samples need to be collected and stored in order to define abnormality operation. Therefore, in this work, wavelet-based algorithms were developed as an analysis process to analyze collected data and identify abnormality, with much fewer data samples and compiling process, as special prosperity of wavelet transform. As a result, MATLAB codes were implemented to analyze data based on sound signature technique and wavelet transform algorithms to show a significant improvement in identifying potential error and abnormality conditions.
Feature extraction using wavelet analysis with application to machine fault diagnosis
2005
Two different approaches have been used to diagnose faults in machinery such as internal combustion engines. In the first approach, a mathematical model of the specific engine or component under investigation is developed and a search for causes of change in engine performance is conducted based on the observations made in the system output. In the second approach, the specific engine or component is considered a black box. Then, by observing some sensory data, such as cylinder pressure, cylinder block vibrations, exhaust gas temperatures, and acoustic emissions, and analyzing them, fault(s) can be traced and detected. In this research the latter approach is employed in which vibration data is used for the detection of malfunctions in reciprocating internal combustion engines. The objective of this thesis is to develop effective data-driven methodologies for fault detection and diagnosis. The main application is the detection and characterization of combustion related faults in reci...
IET Intelligent Transport Systems, 2013
Service station experts examine the sound patterns of the motorcycles to diagnose the faults. Automatic fault diagnosis is a challenging task and more so is recognition of multiple faults. This study presents a methodology for localisation of multiple faults in motorcycles. The sound signatures of multiple faults are constructed by fusing the individual signatures of faults from engine and exhaust subsystems. Energy distributions in the approximation coefficients of wavelet packets are used as features. Among the classifiers used, artificial neural network is found suitable for detecting the presence of multiple faults. The recognition accuracy is over 78% when trained with individual fault signatures and over 88% when trained with combined fault signatures.