Claude Delpha

Incipient fault detection and diagnosis based on Kullback-Leibler divergence using principal component analysis

Most of fault indicators are devoted to detect deviations related to specific features but they fail to detect and estimate unpredictable slight distortions often caused by incipient faults. The Kullback-Leibler divergence is characterised with a high sensitivity to incipient faults that cause unpredictable small changes in the process measurements. This work has two main objectives: first estimate the amplitude of incipient faults in multivariate processes based on the divergence and second evaluate, through detection error probabilities, the performance of the divergence in the detection of incipient faults in noisy environments.Throughout all the paper, the Fault-to-Noise Ratio (FNR) has been referred to as a comparative criterion between the fault level and noise; particularly the region around 0 dB of FNR is of interest in the evaluation. A theoretical study is developed to derive an analytical model of the divergence that considers the presence of Gaussian noise and allows obtaining a theoretical estimate of the fault amplitude. After application on a simulated AR process, the fault amplitude estimate turns out to be an overestimation of the actual amplitude, therefore guaranteeing a safety margin for monitoring. Accurate fault severity estimation for an eddy currents application shows the effectiveness of this approach. HighlightsWe propose to enhance the fault detection approach based on the KLD modelling with the introduction of the noise.Based on the aforementioned model an estimator of the fault amplitude is developed and validated.The performances of the detection are studied in a noisy environment with the introduction of the Fault to Noise Ratio (FNR).The robustness of the proposed method is evaluated with the computation of the miss-detection and false alarms probabilities.A performed validation of this approach with a simulated AR model is given.

Collusion resilient spread spectrum watermarking in M-band wavelets using GA-fuzzy hybridization

This paper proposes a collusion resilient optimized spread spectrum (SS) image watermarking scheme using genetic algorithms (GA) and multiband (M-band) wavelets. M-band decomposition of the host image offers advantages of better scale-space tiling and good energy compactness. This bandpass-like decomposition makes watermarking robust against frequency selective fading-like gain (intelligent collusion) attack. On the other hand, GA would determine threshold value of the host coefficients (process gain i.e. the length of spreading code) selection for watermark casting along with the respective embedding strengths compatible to the gain of frequency response. First, a single bit watermark embedding algorithm is developed using independent and identically distributed (i.i.d) Gaussian watermark. This is further modified to design a high payload system for binary watermark image using a set of binary spreading code patterns. Watermark decoding performance is improved by multiple stage detection through cancelation of multiple bit interference (MBI) effect. Fuzzy logic is used to classify decision magnitudes in multiple group combined interference cancelation (MGCIC) used in the intermediate stage(s). Simulation results show convergence of GA and validate relative performance gain achieved in this algorithm compared to the existing works.

Improved Fault Diagnosis of Ball Bearings Based on the Global Spectrum of Vibration Signals

This research deals with the discrimination between conditions of faults in rolling element bearings based on a global spectral analysis. This global spectral analysis allows to obtain spectral features with significant discriminatory power. These features are extracted from the envelope spectra of vibration signals without prior knowledge of the bearings specific parameters and the characteristic frequencies. These extracted spectral features will then be the global spectral signature produced by the bearing faults. Since the signature produced by the faults in bearing balls is very weak, and hard to be detected and identified, this paper proposes the linear discriminant analysis as part of the global spectral analysis method in order to improve the diagnosis of ball faults. The application on experimental vibration data acquired from bearings containing different types of faults with different small sizes shows the proficiency of the overall method. The Bhattacharyya distance is used to confirm the efficiency of the obtained results.

PMSM Drive Position Estimation: Contribution to the High-Frequency Injection Voltage Selection Issue

High-frequency injection (HFI) is an alternative method to estimate permanent magnet synchronous motor (PMSM) rotor position using magnetic saliency. Once the maximum fundamental electrical frequency and the power converter switching frequency are set, the HFI voltage amplitude tuning is generally based on trial and error. This paper proposes a methodology to select the appropriate high-frequency signal injection voltage amplitude for rotor position estimation. The technique is based on an analytical model taking into account the noise in the voltage supply to derive the resulting currents containing the information on the rotor position. This model allows setting the injection voltage amplitude that leads to the maximum acceptable position error for a given signal-to-noise ratio and a speed range. The approach is validated with the analytical and the global drive models through extensive simulations. Experimental results on a 1-kW PMSM drive confirm the interest of the proposed solution.

An optimal fault detection threshold for early detection using Kullback-Leibler Divergence for unknown distribution data

The incipient fault detection in industrial processes with unknown distribution of measurements signals and unknown changed parameters is an important problem which has received much attention these last decades. However most of the detection methods (online and offline) need a priori knowledge on the signal distribution, changed parameters, and the change amplitude (Likelihood ratio test, Cusum, etc.). In this paper, an incipient fault detection method that does not need any a priori knowledge on the signals distribution or the changed parameters is proposed. This method is based on the analysis of the Kullback-Leibler Divergence (KLD) of probability distribution functions. However, the performance of the technique is highly dependent on the setting of a detection threshold and the environment noise level described through Signal to Noise Ratio (SNR) and Fault to Noise Ratio (FNR). In this paper, we develop an analytical model of the fault detection performances (False Alarm Probability and Missed Detection Probability). Thanks to this model, an optimisation procedure is applied to optimally set the fault detection threshold depending on the SNR and the fault severity. Compared to the usual settings, through simulation results and experimental data, the optimised threshold leads to higher efficiency for incipient fault detection in noisy environment. HighlightsWe propose an incipient fault detection method that does not need any a priori information on the signals distribution or the changed parameters.We show that the performance of the technique is highly dependent on the setting of a detection threshold and the environment noise level.We develop an analytical model of the fault detection performances (False Alarm Probability and Missed Detection Probability).Based on the aforementioned model, an optimisation procedure is applied to optimally set the fault detection threshold depending on the noise and the fault severity.Compared to the usual settings, a performed validation of this approach with through simulation results and experimental data is given.

SVM based diagnosis of inverter fed induction machine drive: A new challenge

In fault diagnosis studies two main approaches are mostly used. The first one consists in designing the full physical or empirical model of the system in healthy and faulty conditions. The major drawback of this approach is the difficulty to obtain an accurate model reflecting all the operating conditions and phenomena. The second approach, used in this work, consists in using signal processing techniques for the characterization of the healthy and faulty behaviors. This paper deals with the study of a fault detection and isolation procedure on a three phase inverter feeding an induction machine drive using pattern recognition techniques. The diagnosis procedure relies on the use of classifiers after the collection of the output currents of the inverter flowing in the machine windings. The proposed classifiers are based on Support Vector Machines (SVM). We show in this paper how it is possible to tune the SVM and also the influence of the data normalisation to perform an effective diagnosis with experimental data.

Discrimination of a refrigerant gas in a humidity controlled atmosphere by using modelling parameters

Abstract For real life conditions of use, the application based on tin oxide gas sensors must be humidity controlled, because humidity is a very influent parameter which may cause false gas detection. In our application, we want to distinguish a refrigerant gas (Forane R134a) in an air conditioned atmosphere using a TGS gas sensor array. At first, this paper presents a summary of the sensor array time-dependent response in a Forane R134a gas concentration (0–1000 ppm) for a humid air atmosphere (0–85%). We show that these responses can be well fitted by a double exponential model for which we extract five modelling parameters. These variables are coupled or not with two experimental parameters, the steady-state conductance and the conductance dynamic slope, and then arranged in data bases. Afterwards, these data bases are treated by using two complementary pattern recognition methods: the principal component analysis (PCA) followed by the discriminant factorial analysis (DFA). We show the ability to discriminate the target gas, whatever the humidity rate, when the modelling parameters are coupled or not with the experimental parameters. Finally, the identification of unknown cases is described.

Informed stego-systems in active warden context: Statistical undetectability and capacity

Several authors have studied stego-systems based on Costa scheme, but just a few ones gave both theoretical and experimental justifications of these schemes performance in an active warden context. We provide in this paper a steganographic and comparative study of three informed stego-systems in active warden context: scalar Costa scheme, trellis-coded quantization and spread transform scalar sosta Scheme. By leading on analytical formulations and on experimental evaluations, we show the advantages and limits of each scheme in term of statistical undetectability and capacity in the case of active warden. Such as the undetectability is given by the distance between the stego-signal and the cover distance. It is measured by the Kullback-Leibler distance.

An Electronic Nose for the Identification of Forane R134a in an Air Conditioned Atmosphere

An electronic nose based on a TGS type sensor array for the main detection of Forane R134a has been characterised under closely controlled gas temperature and humidity conditions. This paper presents the dependence of the TGS sensor array to the gas temperature and the relative humidity rate values. A model is proposed for the sensor array behaviour for each of these parameters. Afterwards, the importance of these two atmospheric parameters is underlined and the need to control or to include them into a database is proven. We present the ability to identify the target gas with the discriminant factorial analysis method even if the relative humidity or the gas temperature differs from the nose database learning process.

An electronic nose for the discrimination of forane 134a and carbon dioxide in a humidity controlled atmosphere

The great interest for the development of performed gas sensors has now an increasing application field -- the electronic nose. We propose here to develop such a system for an environmental application in the field of the air quality control. Our application is based on the study of the learning process of our sensor system for a performed discrimination of a refrigerant gas forane 134a and carbon dioxide in a humidity and temperature controlled atmosphere. In this paper, we present first the characterisation results obtained for the metal oxide sensors used in this application. Afterwards, we propose to compare the discrimination results of the gases with several representative variables by using two pattern recognition methods: principal component analysis (PCA) and discriminant factorial analysis (DFA). We show the importance to use two types of complementary variables: the steady-state conductance and the conductance dynamic slope. We also propose a new method to reduce the drift effect and then we show the discrimination of the gases and also the identification of unknown cases by the way of the created decisive law.