F. Filippetti

Advances in Diagnostic Techniques for Induction Machines

This paper investigates diagnostic techniques for electrical machines with special reference to induction machines and to papers published in the last ten years. A comprehensive list of references is reported and examined, and research activities classified into four main topics: 1) electrical faults; 2) mechanical faults; 3) signal processing for analysis and monitoring; and 4) artificial intelligence and decision-making techniques.

Recent developments of induction motor drives fault diagnosis using AI techniques

This paper presents a review of the developments in the field of diagnosis of electrical machines and drives based on artificial intelligence (AI). It covers the application of expert systems, artificial neural networks (ANNs), and fuzzy logic systems that can be integrated into each other and also with more traditional techniques. The application of genetic algorithms is considered as well. In general, a diagnostic procedure starts from a fault tree developed on the basis of the physical behavior of the electrical system under consideration. In this phase, the knowledge of well-tested models able to simulate the electrical machine in different fault conditions is fundamental to obtain the patterns characterizing the faults. The fault tree navigation performed by an expert system inference engine leads to the choice of suitable diagnostic indexes, referred to a particular fault, and relevant to build an input data set for specific AI (NNs, fuzzy logic, or neuro-fuzzy) systems. The discussed methodologies, that play a general role in the diagnostic field, are applied to an induction machine, utilizing as input signals the instantaneous voltages and currents. In addition, the supply converter is also considered to incorporate in the diagnostic procedure the most typical failures of power electronic components. A brief description of the various AI techniques is also given; this highlights the advantages and the limitations of using AI techniques. Some applications examples are also discussed and areas for future research are also indicated.

Quantitative evaluation of induction motor broken bars by means of electrical signature analysis

The paper reports the comparison and performance evaluation of different diagnostic procedures that use input electric signals to detect and quantify rotor breakage in induction machines supplied by the mains. Besides the traditional current signature analysis based on one-phase current spectrum lines at frequencies (1/spl plusmn/2s)f, the procedures based on analysis of the line at frequency 2sf in the spectrum respectively of electromagnetic torque, space vector current modulus and instantaneous power are considered. These last procedures have similar features and the comparison is developed on the basis of instantaneous torque. It is shown that the speed ripple introduces two further terms in the instantaneous torque, decreasing the accuracy of the diagnosis. It is shown that there is a link between the angular displacement of the current sideband components at frequencies (1/spl plusmn/2s)f. This allows a more correct quantitative evaluation of the fault and to show the superiority of the sideband current components diagnostic procedure over the other proposed methods.

Trends in Fault Diagnosis for Electrical Machines: A Review of Diagnostic Techniques

The fault diagnosis of rotating electrical machines has received an intense amount of research interest during the last 30 years. Reducing maintenance costs and preventing unscheduled downtimes, which result in losses of production and financial incomes, are the priorities of electrical drives manufacturers and operators. In fact, both correct diagnosis and early detection of incipient faults lead to fast unscheduled maintenance and short downtime for the process under consideration. They also prevent the harmful and sometimes devastating consequences of faults and failures. This topic has become far more attractive and critical as the population of electric machines has greatly increased in recent years. The total number of operating electrical machines in the world was around 16.1 billion in 2011, with a growth rate of about 50% in the last five years [1].

On-field experience with online diagnosis of large induction motors cage failures using MCSA

The experience gained by ENEL Produzione (previously the Italian Electric Board) on monitoring the cage condition of large induction motors is reported in this paper. The diagnostic procedure is based on the motor current signature analysis and, in particular, on the two sideband current components near the frequency fundamental line that appear in the current power spectrum when a rotor bar/ring breakage occurs. According to the developed procedure, a diagnostic index obtained from these components is stored and its trend as a function of time allows for the detection of the occurrence of a failure in most cases. This event is clearly shown by the overcoming of a prefixed and triggered threshold. Moreover, machines with particular rotor magnetic structure are considered. In this case, unexpectedly high sideband components appear, even in the presence of healthy cages, and the test procedure was adapted to account for these conditions.

Neural networks aided on-line diagnostics of induction motor rotor faults

An improvement of induction-machine rotor fault diagnosis based on a neural network approach is presented. A neural network can replace in a more effective way the faulted machine models used to formalize the knowledge base of the diagnostic system with suitably chosen inputs and outputs. Training the neural network by data obtained from experimental tests on healthy machines and from simulation in the case of faulted machines, the diagnostic system can discern between healthy and faulty machines. This procedure replaces the formulation of a trigger threshold, required in the diagnostic procedure based on the machine models.<<ETX>>

High Frequency Resolution Techniques for Rotor Fault Detection of Induction Machines

Motor current signature analysis (MCSA) is the reference method for the diagnosis of medium-large machines in industrial applications. However, MCSA is still an open research topic, as some signatures may be created by different phenomena, wherein it may become sensitive to load and inertia variations, and with respect to an oscillating load torque, although suitable data normalization can be applied. Recently, the topic of diagnostic techniques for drives and low to medium size machines is becoming attractive, as the procedure can be embedded in the drive at no additional thanks to a dedicated firmware, provided that a suitable computational cost is available. In this paper, statistical time-domain techniques are used to track grid frequency and machine slip. In this way, either a lower computational cost or a higher accuracy than traditional discrete Fourier transform techniques can be obtained. Then, the knowledge of both grid frequency and machine slip is used to tune the parameters of the zoom fast Fourier transform algorithm that either increases the frequency resolution, keeping constant the computational cost, or reduces the computational cost, keeping constant the frequency resolution. The proposed technique is validated for rotor faults.

Advanced Diagnosis of Electrical Faults in Wound-Rotor Induction Machines

The aim of this paper is to present a diagnosis methodology for the detection of electrical faults in three-phase wound-rotor induction machines (WRIMs). In the considered application, the rotor windings are supplied by a static converter for the control of active and reactive power flows exchanged between the machine and the electrical grid. The proposed diagnosis approach is based on the use of wavelet analysis improved by a preprocessing of the rotor-voltage commands under time-varying conditions. Thus, the time evolution of fault components can be effectively analyzed. This paper proves also the importance of the fault components computed from rotor voltages in comparison to those coming from rotor currents under closed-loop operation. A periodical quantification of the fault, issued from the wavelet analysis, has been introduced for accurate stator- or rotor-fault detection. Simulation and experimental results show the validity of the proposed method, leading to an effective diagnosis procedure for both stator and rotor electrical faults in WRIMs.

Development of expert system knowledge base to on-line diagnosis of rotor electrical faults of induction motors

The authors consider the development of a knowledge base branch related to rotor electrical faults in squirrel cage machines, to be implemented in an expert system (ES), utilizing instantaneous values as input data. The knowledge base is organized in two levels: in the first level diagnostic indexes for the orientation of the ES inference engine toward the appropriate branch of the fault tree are utilized. The second level includes the deep knowledge with a data set obtained on the basis of a complete faulty machine model. The diagnostic indexes of the first level concern how to distinguish faulty events from the healthy signals due to the unavoidable manufacturing asymmetries. They are pointed out through a simplified model of a faulted rotor that needs few machine parameters. Some diagnosis examples are reported to describe the sequence of operations of the diagnostic system.<<ETX>>

Monitoring of induction motor load by neural network techniques

This paper deals with the electric tracing of the load variation of an induction machine supplied by the mains. A load problem, like a torque dip, affects the machine supply current and consequently it should be possible to use the current pattern to detect features of the torque pattern, using the machine itself as a torque sensor. But current signature depends on many phenomena and misunderstandings are possible. At first the effect of different load anomalies on current spectrum, in comparison with other machine problems like rotor asymmetries, are investigated. Reference is made to low frequency torque disturbances, which cause a quasistationary machine behavior. Simplified relationships, validated by simulation results and by experimental results, are developed to address the current spectrum features. In order to detect on-line anomalies, a current signature extraction is performed by the time-frequency spectrum approach. This method allows the detection of random faults as well. Finally it is shown that a neural network approach can help the torque pattern recognition, improving the interpretation of machine anomalies effects.