In this article, an effective machine learning-based fault diagnosis method is developed for induction motors driven by variable frequency drives (VFDs). Two identical 0.25 HP induction motors under healthy, single-, and multifault conditions were tested in the lab with different VFD output frequencies and motor loadings. The stator current and the vibration signals of the motors were recorded simultaneously under steady-state for each test, and both signals are evaluated for their suitability for fault diagnosis. The signal processing technique, discrete wavelet transform, is chosen in this article to extract features for machine learning. Four families of machine learning algorithms in the MATLAB Classification Learner Toolbox, decision trees, support vector machines, k -nearest neighbors, and ensemble, with 20 classifiers are evaluated for their classification accuracy when used for fault diagnosis of induction motors fed by VFDs. To allow fault diagnosis for untested motor operating conditions, the feature calculation formulas are developed through surface fitting using experimental data of a range of tested frequencies and loadings of the motor to provide training data for untested conditions.

Single- and Multi-Fault Diagnosis Using Machine Learning for Variable Frequency Drive-Fed Induction Motors

According to current energy efficiency labels, the majority of light-emitting diode (LED) lamps on the market are considered as highly efficient devices. This makes it difficult to distinguish between lamps with different operational characteristics and performance. This paper introduces a comprehensive experimental-based labeling methodology for comparing LED lamp performance with reference to two additional important characteristics: light flicker and power factor. The new labeling methodology reveals that there is high diversity between different LED lamps with different circuit topologies but also for a given topology with different design choices. General consumers and design engineers can benefit from the simple and clear information presented by the set of comparative labels when comparing LED lamp performance.

/pdf/light-flicker-and-power-factor-labels-for-comparing-led-lamp-1z4g2mmctg.pdf

Light Flicker and Power Factor Labels for Comparing LED Lamp Performance

Voltage unbalance and mechanical overload generate negative effects on induction motors, producing thermal damages in the stator insulation and a reduction of the motor lifetime. In this regard, the development of protection devices is crucial as they can help to maintain the integrity of the motor and avoid irreversible damages such as insulation system breakdown, short-circuits, and so on. In this work, a methodology to obtain a time-protection model for induction motors from voltage unbalance is presented. The methodology is based on the thermal impact on the motor produced by mechanical overload and voltage unbalance conditions. In general, it consists of the following steps: i) induce in the motor different overloads and voltage unbalance levels and monitor their thermal profiles at the stator winding, ii) obtain the time–overload curve of the motor, iii) determine both the thermal level according to both the time-overload curve and the overload thermal profile, and iv) estimate the parameters of the time-unbalance model using both the thermal level obtained and the unbalance thermal profiles. The model is validated through its implementation in an online protection scheme, the results show a similar behavior between overload and unbalance protections in the safe thermal level defined by the protection models. The protection scheme is carried out over a 750 W three-phase induction motor under different overload and voltage unbalance operating conditions.

Thermal-Impact-Based Protection of Induction Motors Under Voltage Unbalance Conditions

Power quality (PQ) is becoming more and more critical issue in shipboard microgrid systems (SMGs) Especially, the impact of voltage unbalance combined with harmonic distortions on the SMG behavior has not been fully investigated In this paper, simple PQ assessment models and a series of controlled experiments are proposed and carried out in a real ship under sea-going conditions The ship experimental results are presented and discussed considering nonlinear bow thruster load and high power ballast pump loads under unbalanced and harmonic voltage conditions In addition, the analysis of bow thruster current harmonic surges during the ballast pump start-up is presented Furthermore, the voltage/current distortions of a working generator, bow thruster and pump loads are analyzed The paper provides a valuable analysis for coping with PQ issues in the SMG

/pdf/power-quality-assessment-in-shipboard-microgrids-under-17dsym2p2j.pdf

Power Quality Assessment in Shipboard Microgrids Under Unbalanced and Harmonic AC Bus Voltage

The ability of an electric powertrain to perform according to mechanical specifications is equally important as assessing its thermal protection limits, which are affected by its electrical and thermal properties. Although rated parameters (such as power, torque, etc.) are easily accessible in catalogs of equipment producers, more specific properties like mass/length of copper winding, heat dissipation factor, etc., are not available to customers. Therefore, an effective selection of drivetrain components is limited due to the lack of sufficient data and the need to consult critical design decisions with suppliers. To overcome this limitation, we propose a method to estimate the temperature rise of motor drives based on popular loadability curves, which are provided in catalogs. A simple first-order thermal model is applied to represent heating/cooling phenomenon of motor drives. The parameters' identification process is formulated as a nonlinear optimization problem and solved using commercial software products. Within the proposed approach, it becomes possible to include the effect of reduced torque availability at low speeds in self-ventilated motors during design of electric actuation systems. Contrary to using a discrete set of permissible overload conditions from the catalogs, the current methodology allows for evaluating a temperature rise of a motor drive for any overload magnitude, duty cycle, and ambient temperature. This greatly improves flexibility of the design process and facilitates communication in a supplier-customer dialog. The discussed method is verified against reference overload recommendations, yielding the same thermal protection levels, and validated using the experimental results, producing identical motor temperature rise profiles as the ones measured on the laboratory test bench.

Identification and Experimental Validation of an Induction Motor Thermal Model for Improved Drivetrain Design

This paper describes efficiency measurement in induction motor and variable-frequency drives systems under imposed voltage unbalance at the drive input. A number of drives from different manufacturers are tested with matching sizes of motors and the results are compared. The results highlight differences in the effect of voltage unbalance on drive performance. The results also indicate that intrinsic motor performance has important influence on motor-drive system efficiency.

Operation of variable frequency drive motor systems with source voltage unbalance

This paper presents an experimental investigation of the performance of induction motors subjected to single phasing on the primary side of a distribution transformer that feeds the motors. While single phasing caused by, for example, loss of fuse or bad contact is permanent in nature, that caused by recloser operation is momentary and is the focus of this paper. Several motors and transformer connections were used to investigate the effects of the momentary interruptions of a phase on the performance of the induction motors. The conditions under which the motors are able to ride through the single phasing are presented and discussed.

Induction Motor Single-Phasing Performance Under Distribution Feeder Recloser Operations

This paper shows the results of a real life case-study which demonstrates also that there is a need for new Flickermeters able to cope with new lamps' technologies different from standard incandescent, in particular with LED lamps. The case-study refers to a LED lamp that was visibly flickering even if measurements taken onsite with a standard PQ instrument reported neither THD nor Pst issues. Numerical tests using a new Flickermeter proposed by some of the authors and laboratory experiments by means of an objective light Flickermeter have demonstrated the reasons of this singular case.

/pdf/a-real-life-light-flicker-case-study-with-led-lamps-1dvyu97qu0.pdf

A real life light flicker case-study with LED lamps

Laboratory tests were carried out on seven voltage monitoring relays to determine their operational characteristics. The objective was to gain an understanding of the impact of identical power quality events on different voltage monitors commonly installed on utilization equipment. The tests focused on typical monitors found on a local utility's customers' equipment including device models that have previously encountered power quality issues. A similar experiment was carried out on nine thermal overload relays that are widely used for motor protection. The results indicate that both the thermal overload and voltage monitoring relays are prone to erratic response and may be responsible for some nuisance trips observed on utilization equipment.

Joseph Grappe

Papers

Operation of variable frequency drive motor systems with source voltage unbalance

Induction Motor Single-Phasing Performance Under Distribution Feeder Recloser Operations

A real life light flicker case-study with LED lamps

Response of motor thermal overload relays and phase monitors to power quality events