Israel Cruz-Vega

Vibration Analysis of Partially Damaged Rotor Bar in Induction Motor under Different Load Condition Using DWT

The relevance of the development of monitoring systems for rotating machines is not only the ability to detect failures but also how early these failures can be detected. The purpose of this paper is to present an experimental study of partially damaged rotor bar in induction motor under different load conditions based on discrete wavelet transform analysis. The approach is based on the extraction of features from vibration signals at different level of damage and three mechanical load conditions. The proposed analysis is reliable for tracking the damage in rotor bar. The paper presents an analysis and extraction of vibration features for partially damaged rotor bar in induction motors. The experimental analysis shows the change in behavior of vibration due to load condition and progressive damage.

FPGA implementation of Orthogonal Matching Pursuit algorithm

In this paper, the Orthogonal Matching Pursuit (OMP) algorithm is implemented on a Field Programmable Gate Array (FPGA) to obtain the number and position of the atoms in a dictionary. The dictionary, obtained by K-Singular Value Decomposition (K-SDV) algorithm and developed in Matlab, reconstructs a signal from its Sparse representation. With the atoms selected by the OMP algorithm, a signal classification is obtained by means of the FPGA. Some experimental tests are presented to verify the proper operation.

Intelligent identification of induction motor conditions at several mechanical loads

This paper presents an assessment of three classification models, all based on computational intelligence techniques, for the automatic identification of three possible conditions found in induction motors: healthy, with a half broken rotor bar or with one broken bar. Motors with full-load, half load and no load were considered. Based on evidence previously reported, the power spectral densities of the absolute value of a motor currents and vibrations are suitable as signatures of possible damages. However, a simple statistical analysis over such raw signals is not enough to accurately identify such conditions. In order to obtain good identification performance, we looked for accurate characterizations of vibration signals, as well as for suitable classifiers, in order to improve performance rates previously reported. We found that a feature extraction method, based on simple statistics of Discrete Wavelet Transforms (DWT), was able to characterize well the conditions with different loads. Classifiers analyzed were based on three strategies: one-nearest neighbors (1-NN), Support vector machines (SVM) and multi-layer perceptrons (MLP). Our 3-fold validated experiments reported up to 100% accuracy for motors with full load and no load; a 99.87% of accuracy was obtained in motors with half load, using 1-NN.

Automatic stellar spectral classification with multiple intelligent classifiers

Stellar Classification is based on their spectral characteristics. In order to improve performance rates previously reported, like those based on statistical analysis or data transformations, classifiers based on computational intelligence provide a high level of accuracy no matter the presented high level of non-linearity or high dimensionality characteristics of data. In this paper, the stars classification is based on the use of three main strategies: multi-layer perceptrons (MLP), one nearest neighbor (1-NN), and support vector machines (SVM). A strategy of one-vs-one (OVO) for binary classification and directed acyclic graphs (DAG), improves the accuracy of classification and reduce the computational cost.

FPGA-based broken bar detection on IM using OMP algorithm

In this paper, the classification of three Induction Motor conditions using the Orthogonal Matching Pursuit algorithm is presented. The OMP algorithm was implemented into a Field Programmable Gate Array to obtain the Sparse representation of the signal given a Dictionary. Then, the signal information obtained from the Sparse representation is evaluated and classified. The dictionaries were obtained by K-Singular Value Decomposition (K-SVD) algorithm developed in Matlab software. The FPGA implementation is low-complexity, cheap on hardware, compact, and works at 100MHz.

Spectral Characterization of Content Level Based on Acoustic Resonance: Neural Network and Feedforward Fuzzy Net Approaches

Free vibration occurs when a mechanical system is disturbed from equilibrium by an external force and then it is allowed to vibrate freely. In free vibrations, the system oscillates under the influence of inherent forces on the system itself. Free vibrations are associated with natural frequencies that are properties of the oscillating system, quantified in parameters such as mass, shape, and stiffness distribution. A number of these mechanical characteristics can be inferred from vibration patterns or from the generated sound using the adequate sensors. It is well known that liquid level inside a container modifies its natural frequencies. Unfortunately, other container characteristics such as shape, composition, temperature, and pressure modifies the natural frequencies of vibration making the task of level measurement nontrivial. Preliminary experiments aiming to do measurement of liquid content level and container characterization are presented in this work. Spectral analysis in Fourier domain is used to perform feature extraction, with the feature vectors containing information about the frequencies having the greatest amplitude in the respective spectral analysis. Classification has been carried out using two computational intelligence techniques for comparison purposes: neural network classification and a fuzzy logic inference system built using singleton fuzzifier, product inference rule, Gaussian membership functions and center average defuzzifier. Preliminary results showed a better performance when using the neural network-based approach in comparison to the fuzzy logic-based approach, obtaining in average a MSE of 0.02 and 0.09, respectively.