In today's image centric world processing has become the centre of attention for various real-life applications. Embarking from car thefts, breaking traffic rules to contravene restricted expanse, Image Processing given us a conviction put grinding halt these malpractices. Inspired by feature learning capabilities Convolution Neural Networks (CNN), preeminent work is detection and recognition plate number which accomplished dint Network (CNN). The images can be procured still camera. Self...

Purpose The purpose of this paper, is to predict the various phases and crystal structure from multi-component alloys. Nowadays, concept strategies development multi-principal element alloys (MPEAs) significantly increase count potential candidate alloy systems, which demand proper screening large number systems based on nature their phase structure. Experimentally obtained data linking elemental properties resulting for MPEAs profused; hence, there a strong scope...

Reliability of the traditional analytical model building techniques for Robotic Manipulators is debatable with higher Degrees Freedom (DoF) and under dynamic, uncertain environments. Keeping these uncertainties inaccuracies in backdrop, researchers have been encouraged to use supervised machine learning as a better alternative data-driven learning. The main advantage models lies their adaptability cope variations real-time. Considering proven superiority Recurrent Neural Networks (RNN)...

An attempt has been made to understand the kinetics concerning manufacturing schedule of high strength 304 stainless steel comprising formation strain-induced α′ martensite in course cold rolling (25%) at 0°C and its reversion during annealing temperature range 300–800°C for 1 h. Increase with increasing without any perceptible grain growth up 800°C demonstrated by microstructural investigation. X-ray diffraction analysis revealed that 25% resulted 20% deformation induced which after reduced...

This paper proposes a FPGA friendly fused DWT-FFT based technique for broken rotor bar fault detection of induction motor. Discrete wavelet transform is implemented in platform realizing its time-frequency perseverance quality. The smallest length `Haar' capable detecting with satisfactory level accuracy. Since Haar used the computational burden and resource utilization are reduced as compared to others. In addition that significant reduction FFT computation analyze spectrums specific band...

Artificial Neural Networks (ANN) provides a simple and efficient method to implement highly non-linear complex systems due its "Universal Function Approximation" capabilities. However lack of hardware design that is capable adopting any changes in operating environment the system limits applicability ANN automotive industrial environment. The most challenging task for implementation embedded plat-form realization sigmoidal activation function. This paper aims address various issues terms...

Application of machine learning in the area fault diagnosis induction motor is somewhat popular. However major challenges this domain are restricted by handmade statistical features, which limit performances classifiers immensely. On other hand deep learning, a new era has opened horizon, where self-synthesized features from raw signals have already proven their superiority various real life applications. This paper exploited merit and reports preliminary findings detection using novel Quasi...

The use of data driven intelligent system is gaining importance in the area condition monitoring electrical equipment. However, irrelevant and redundant input features make bulky, computation intensive provides poor classification accuracy. Data mining feature selection techniques play an important role to reduce these problems. Not only but also clustering quality selected actually guides engineer pick up best for developing real time applications. This paper have proposed effort...

Well-dispersed exfoliated white graphene (h-BN) nanosheet in transformer oil was prepared at various weight percentages. The nanofluid showed excellent stability over long time duration and significant improvement of thermal conductivity (>45% for 0.05 wt.%)due to large surface area high h-BN nanosheets.

Abstract Learning of better feature representation instinctively by Convolutional Neural Networks (CNN) has inspired to address the unsolved issues in stator current based multi-class fault diagnosis induction motor drives. Current envelope acquired using Hilbert transform is proven be effective pre-processing method handle complex data pattern and reveal masked defect information. The self-synthesized quality features through deep convolution layers outperforms reaches an unmatched accuracy...

FPGA based embedded system for continuous online monitoring has gained importance in recent years. The existing methodologies rely on transient analysis, which unnecessarily gives undue stress to the motor. Also, FFT is used consumes large resource hardware unit. In this paper a DWT algorithm designed and implemented detect Broken rotor bar fault using vibration signal at low loading condition steady state. main contribution of work considerable reduction by use 2-length filter RMS energy...