Induction motors are critical components for most industries. Induction motor failures may yield an unexpected interruption at the industry plant. Several conventional vibration and current analysis techniques exist by which certain faults in rotating machinery can be identified; however, they generally deal with a single fault only. Instead, in real induction machines, the case of multiple faults is common. When multiple faults exist, vibration and current are excited by several fault-related frequencies combined with each other, linearly or nonlinearly. Different techniques have been proposed for the diagnosis of rotating machinery in literature, where most of them are focused on detecting single faults and few works deal with the diagnosis and identification of multiple combined faults. The contribution of this paper is the development of a condition-monitoring strategy that can make accurate and reliable assessments of the presence of specific fault conditions in induction motors with single or multiple combined faults present. The proposed method combines a finite impulse response filter bank with high-resolution spectral analysis based on multiple signal classification for an accurate identification of the frequency-related fault. Results show the methodology potentiality as a deterministic detection technique that is suited for detecting multiple features where the fault-related frequencies are very close to those analytically reported in literature.

The development of monitoring systems for rotating machines is the ability to accurately detect different faults in an incipient state. The most popular rotating machine in industry is the squirrel-cage induction motor, and the failure on such motors may have severe consequences in costs, product quality, and safety. Most of the condition-monitoring techniques for induction motors focus on a single specific fault. The identification of two or more combined faults has been rarely considered, in spite of being a very usual situation in real rotary machines. On the other hand, information entropy is a signal processing technique that has recently proved its suitability for fault detection on induction motors, and fuzzy logic analysis has extensively been used in combination with several processing techniques in improving the diagnosis of a single isolated fault. The contribution of this paper is a novel methodology that is suitable for hardware implementation, which merges information entropy analysis with fuzzy logic inference to identify faults like bearing defects, unbalance, broken rotor bars, and combinations of faults by analyzing one phase of the induction motor steady-state current signal. The proposed methodology shows satisfactory results that prove its suitability for online detection of single and multiple combined faults in an automatic way through its hardware implementation in a field programmable gate array device.

Outdoor images are used in a vast number of applications, such as surveillance, remote sensing, and autonomous navigation. The greatest issue with these types of images is the effect of environmental pollution: haze, smog, and fog originating from suspended particles in the air, such as dust, carbon and water drops, which cause degradation to the image. The elimination of this type of degradation is essential for the input of computer vision systems. Most of the state-of-the-art research in dehazing algorithms is focused on improving the estimation of transmission maps, which are also known as depth maps. The transmission maps are relevant because they have a direct relation to the quality of the image restoration. In this paper, a novel restoration algorithm is proposed using a single image to reduce the environmental pollution effects, and it is based on the dark channel prior and the use of morphological reconstruction for the fast computing of transmission maps. The obtained experimental results are evaluated and compared qualitatively and quantitatively with other dehazing algorithms using the metrics of the Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity (SSIM) index; based on these metrics, it is found that the proposed algorithm has improved performance compared to recently introduced approaches.

The relevance of the development of monitoring systems for rotating machines is not only the ability to detect failures but is also how early these failures can be detected. Squirrel-cage induction motors are the most popular motors used in industry, consuming around 85% of the power in industrial plants. Broken rotor bars in induction motors are among the major failures that are desirable to detect at early stages because this failure significantly increases power consumption and is responsible for further damage to the machinery. Previously reported works base their analysis on current or vibration monitoring for broken-bar detection up to one broken bar under mechanically loaded motor conditions. The contribution of this paper presents a novel methodology for half-broken-bar detection, which combines current and vibration analysis by correlating the signal spectra to enhance detectability for mechanically loaded and unloaded operating conditions of the motor, which the other isolated techniques are unable to detect. The proposed methodology is implemented in a low-cost field-programmable gate array (FPGA), giving a special-purpose system-on-a-chip (SoC) solution for online operation, with the development of a complex postprocessing decision-making unit. Several cases of study are presented to demonstrate the performance of the implementation.

Power inverters are essential components in modern electrical distribution systems. Switches are key elements of a power inverter, and they are the more affected components when a fault occurs. Breakdown of power-semiconductor devices represents around 38% of a power inverter fault. An open-circuit fault will keep the inverter working, jeopardizing its components and degrading its performance. The diagnosis and classification of power inverter faults have become important research subjects to avoid malfunctioning or damage of other components in an electric system. The Fourier transform and the wavelet-based multiresolution analysis (MRA) are the most applied signal processing techniques for identifying open-circuit faults in these devices. They are usually combined with other methods for increasing its effectiveness, which demands a considerable use of computational resources. This work introduces a novel low-computational-cost technique that directly analyzes the electric phase currents from a power inverter, in the time domain, utilizing three rotatory reference systems and basic arithmetic operations to identify open-circuit faults. The proposed method allows determining the specific power supply phase and the faulty switch, or switches, with high certainty. The obtained results are validated through exhaustive analyses against the Fourier transform and the wavelet-based MRA, demonstrating the proposed method suitability for online-monitoring applications, ensuring high sensitivity and accuracy on detecting and isolating open-circuit faults with low computation cost.