Transient motor current signature analysis has become a mature technique for fault detection in induction motors. By using start-up transients, the whole range of slip in the machine is exploited to generate well-defined fault frequency patterns. However, in the inverter-fed motor case, the fault-patterns are always close to the supply frequency and often of low amplitude. Therefore, it is difficult to distinguish and localize the fault-patterns. In this paper, a novel method is proposed to create a new fault-pattern; the proposed technique can concentrate the fault-harmonic in a specific frequency bandwidth and avoid the spectral leakage by reducing the supply frequency amplitude. The methodology has been validated through experimental tests carried out to detect broken rotor bar in an induction motor started through a voltage source inverter.

One of the most critical devices in an electrical system is the transformer. It is continuously under different electrical and mechanical stresses that can produce failures in its components and other electrical network devices. The short-circuited turns (SCTs) are a common winding failure. This type of fault has been widely studied in literature employing the vibration signals produced in the transformer. Although promising results have been obtained, it is not a trivial task if different severity levels and a common high-level noise are considered. This paper presents a methodology based on statistical time features (STFs) and support vector machines (SVM) to diagnose a transformer under several SCTs conditions. As STFs, 19 indicators from the transformer vibration signals are computed; then, the most discriminant features are selected using the Fisher score analysis, and the linear discriminant analysis is used for dimension reduction. Finally, a support vector machine classifier is employed to carry out the diagnosis in an automatic way. Once the methodology has been developed, it is implemented on a field-programmable gate array (FPGA) to provide a system-on-a-chip solution. A modified transformer capable of emulating different SCTs severities is employed to validate and test the methodology and its FPGA implementation. Results demonstrate the effectiveness of the proposal for diagnosing the transformer condition as an accuracy of 96.82% is obtained.

Induction motor fault identification is essential to improve efficiency in industrial processes improving costs, production line and maintenance time. This paper presents a novel motor fault detection methodology based on Quaternion Signal Analysis (QSA). The proposed method establishes the quaternion coefficients as the value of motor current measurement and the variables x, y and z are the measurements from a triaxial-accelerometer mounted on the induction motor chassis. The method obtains the rotation of quaternions and applies quaternion rotation statistics such as mean, cluster shades and cluster prominence in order to get their features, and these are used to classify the motor state using the tree classification algorithm. This methodology is validated experimentally and compared to other methods to determine the efficiency of this method for feature detection and motor fault identification and classification.