Automatique
VIBRATION MONITORING FOR FAULT DIAGNOSIS IN ROTATING MACHINERY USING WAVELET TRANSFORM
Vibration analysis is essential in improving condition monitoring and fault diagnostics of rotating machinery. Many signal analysis methods are able to extract useful information from vibration data. Currently, the most of these methods use spectral analysis based on Fourier Transform (FT). However, these methods present some limitations; it is the case of non-stationary signals. In the present work, we are interested to the vibration signal analysis by the Wavelet Transform (WT). The WT is one of the most important methods for signal processing; it is especially suitable for non-stationary vibration measurements obtained from accelerometer sensors. The monitoring results indicate that the WT can diagnose the abnormal change in the measured data. Voir les détails
Mots clés : Vibration analysis, Fault Diagnosis, Rotating machinery, Spectral analysis, Wavelet transform
Application of Wavelet Transform for Fault Diagnosis in Rotating Machinery
Vibration analysis is essential in improving condition monitoring and fault diagnostics of rotating machinery. Many signal analysis methods are able to extract useful information from vibration data. Currently, the most of these methods use spectral analysis based on Fourier Transform (FT). However, these methods present some limitations; it is the case of non-stationary signals. In the present work, we are interested to the vibration signal analysis by the Wavelet Transform (WT). The WT is one of the most important methods for signal processing; it is especially suitable for non-stationary vibration measurements obtained from accelerometer sensors. The monitoring results indicate that the WT can diagnose the abnormal change in the measured data. Voir les détails
Mots clés : Vibration analysis, Fault Diagnosis, Rotating machinery, Spectral analysis, Wavelet transform
Fault detection and diagnosis using principal component analysis. Application to low pressure lost foam casting process
Process fault detection and diagnosis plays a very important role in the production security and the product quality. In this paper, in order to improve the accuracy for fault detection and diagnosis, a new method based on Principal Component Analysis (PCA) is proposed in low pressure lost foam casting process. PCA method reduces the dimensionality of the original data set by the projection of the data set onto a smaller subspace defined by the principal components, the aim of this method is to establish the normal statistical correlation among the coefficients of the data set to detect and diagnose the faults. The process faults are detected and diagnosed using Multivariate Statistical Process Control (MSPC) parameters such as: Hotelling’s T2-statistic, Q-statistic or Squared Prediction Error (SPE) and Q-residual contribution. The monitoring results indicates that the proposed method can be detect and diagnose the abnormal change of the process. Voir les détails
Mots clés : fault detection and diagnosis, principal component analysis, multivariate statistical process control, T2-statistic, Q-statistic, squared prediction error, Q-residual contribution.
Welding Quality Evaluation Using Residual – Based Reference Temperature Distribution Model and Fuzzy Reasoning
A method for welding quality evaluation, which combines model identification and fuzzy sets methods, is proposed. To account for welding quality variations, the proposed approach is based on an optimal reference Gauss distribution temperature along of the welding line in order to take into account the eventual process changes. Fuzzy analysis is then applied to the generated residual data to give an evaluation of the welding condition. This approach is applied to welding process for constructing a complementary condition monitoring system which permits an online quality evaluation. The temperature measurement is carried out using an infrared camera. Simulation results based on the measured surface temperature and generated residual data show that the new approach is easily implementable and gives good evaluation online. Voir les détails
Mots clés : Heat Affected zone of Welding Process, Infrared temperature measurement, Gauss distribution model, Residual generation, intelligent modeling, Fuzzy reasoning, Quality evaluation.
Adaptive support vector machine-based surface qualityevaluation and temperature monitoring. Applicationto billet continuous casting process
A method for surface quality evaluation and tem-perature monitoring of the billet in continuous casting isconsidered in this paper. This method uses the differencebetween the measured and the filtered temperature comput-ed using an adaptive support vector machine method. Thetemperature field, measured by an infrared camera, is affect-ed by an important noise called calamine (a metal oxidegenerated during the cooling process). The quality of thebillet ’ s surface temperature is connected to the secondarycooling behavior, and therefore an evaluation of the cala-mine effect is needed. Methods such as soft sensing andadaptive support vector machine are used for a global eval-uation of calamine intensity on the monitored area of thebillet in continuous casting. This kind of approaches isapplied in continuous casting process for constructing acomplementary condition monitoring system, which allowsan online calamine evaluation. Simulation results, based onthe measured surface temperature and the adaptive supportvector machine analysis, showed that this new combinedapproach is easily implementable and gives good resultswhen applied online. Voir les détails
Mots clés : Continuous casting . Surface billet casting . Infrared temperature measurement . Adaptive support vector machine (ASVM) . Process and quality evaluation
Inferential sensor – Based adaptive principal componentsanalysis for mechanical properties prediction and evaluation
This paper is concerned with a method for on-line quality prediction and evaluation ofmechanical properties in metal testing. This method uses an Adaptive Principal ComponentAnalysis (APCA) as a multi predictor of different sub-models defining the mechanical prop-erties such as constraints limits and elongation. The PCA technique, characterized by itsmultivariate component, is strongly recommended to model a multi-input-multi-outputsystem. The complex system is generally known as a non-linear and unsteady state pro-cess. The PCA method is a linear projection. To adapt it and to improve the prediction accu-racy, a variant of this method is considered based on iteratively using a specific algorithm.This kind of approaches is applied for constructing an inferential model, which allows areliable and accurate predictor. Simulation results, based on the measured and computeddata using the above-cited method, show that the proposed approach is easily implement-able and give an accurate prediction. Voir les détails
Mots clés : Soft sensor, Mechanical testing, Adaptive principal component analysis, Uncertainties evaluation
Modeling and Control of the Wind Energy Conversion Systems Based on DFIG Under Sub- and Super-Synchronous Operation Modes
In this work, the modeling and control of the Wind Energy Conversion Systems (WECS) based on doubly fed induction generator (DFIG) are presented. Firstly, we developed the models of the different elements of the conversion chain. After, we consider the vector control strategy of the active and reactive powers in order to ensure an optimum operation. Finally, the dynamic model of a DFIG and wind turbine grid connected system is determined in the dq-synchronous reference frame. The numerical simulation results obtained with Matlab/Simulink software present the behaviors of the sub-synchronous and super-synchronous operation modes. Voir les détails
Mots clés : wind power generation, doubly fed induction generator, renewable energy, modeling, control.
Task Performance Evaluation for Supervisory Control Systems
Integrated multi-modal Supervisory Control Systems (ISCS) are a new generation ofcomplex and synergistic Human-Machine Interaction Systems (HMIS). This paper deals withmulti-modal interaction and control applied to Human Robot Systems (HRS). A task performanceevaluation technique dedicated for multi-modal interaction and control is proposed. It enablescomparison of task performance carried out by using different selection of control modes or bydifferent operators. Objective and subjective performance measures are defined.Based on the Analytical Hierarchy Process method (AHP) which takes into account qualitative andquantitative attributes and criteria, a task performance evaluation technique has been proposed forsupervisory systems which enables multimodal interaction modes .Experimental results have been carried out and some preliminary results will be presentedconcerning parallel cable-based manipulators. Voir les détails
Mots clés : Task performance evaluation, human factors, Analytical Hierarchy Process, Adaptive supervisory control, Multi-modal interaction, Cable-based robots, Graphical-user interface.
Image-based Control for Cable-based Robots
Some human robot interactive applications involved in tele-robotics, remote supervisory andunmanned systems require specific capabilities. This demand has promoted various interactive modesand high-level control techniques such as tele-manipulation, speech, vision, gesture, etc. Among theseinteractive modes, the image based control which is often named point and click control has proven tobe the most appropriate one that offers multiple advantages. This mode consists of only and simplypointing in an appearing object of an image received from a remote site, to convert this click into a robotcommand towards the corresponding location in the real world space. This mode is suitable for remoteapplications, frees the human operator from being involved into the loop enabling him/her to usecommands in the sense of click and forget. This paper presents, firstly, the design and the realization ofan experimental planar cable-based robot constituted of four cables. Secondly, it presents the designand the implementation of a high-level image-based control. Some typical experiments which havebeen performed prove the simplicity and the effectiveness of the image-based control. Moreover, itopens perspectives for new applications with cable-based robots, particularity for rehabilitation applications. Voir les détails
Mots clés : Cable-based robot, human-robot interaction, image-based control, point and click, robot control.
VIBRATION SIGNAL ANALYSIS USING WAVELET-PCA-NN TECHNIQUE FOR FAULT DIAGNOSIS IN ROTATING MACHINERY
Condition monitoring and fault diagnostics are useful for ensuring the safe running of machines. The demand for monitoring and fault diagnosis of processes and sensors in industrial systems has increased the efforts to develop new analysis techniques. In this paper, a new combined fault diagnosis method that uses Wavelet Transform (WT), Principal Component Analysis (PCA) and Neural Networks (NN) is proposed for rotating machinery vibration monitoring and analysis. In this proposed method, WT is employed to decompose the vibration signal of the sensor measurements into approximations and details coefficients at different levels. These coefficients are then used as inputs to the PCA algorithm in order to perform fault detection and feature extraction using Q-statistic or Squared Prediction Error (SPE) and Q-residual contribution, respectively. After that, NN are applied to further improve the separation between fault classes. The measurements from the vibration process are used to verify the WT-PCA-NN method for detecting and diagnosing faults under typical operating conditions. Simulation results using real sensor measurements from a pilot scale are presented and discussed. Voir les détails
Mots clés : Vibration Signal; Fault Diagnosis; Wavelet Transform; Principal Component Analysis; Neural Networks.