روش داده کاوی برای تشخیص عدم تعادل در عرضه موتور القایی
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|21433||2009||6 صفحه PDF||سفارش دهید||4360 کلمه|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Expert Systems with Applications, Volume 36, Issue 9, November 2009, Pages 11808–11813
This paper describes an approach for detection of the supply unbalance condition in induction motors by using data mining process. Simulation results have shown that a good indicator of the fault is the amplitude of the second harmonic of the supply frequency component (2f) in the signal obtained by the differences in supply current zero crossing instants. In the study, linear regression (LR), pace regression (PR), sequential minimal optimization (SMO), M5 model tree, M5’Rules, KStar, additive regression and back propagation neural network (BPNN) models are applied within the data mining process for determining the condition of the motor supply voltage. All data mining algorithms were applied using WEKA software. The best result for the determination of the fault related dominant parameter was obtained by using the M5P algorithm model.
The squirrel cage induction motor has been widely used in all kinds of industry because of its reliability, robustness and simplicity of its construction. However condition monitoring of the motor becomes a necessity to prevent any unplanned stops and breakdowns. The effects of unbalanced supply voltage on the induction motors are stated as reduction on efficiency, temperature rises, and increments on the rotor losses. Voltage unbalance generates negative sequence component in the voltage due to a reverse rotating air gap field in the opposite direction of rotor rotation. The reduction in efficiency of a three phase induction motor supplied by unbalanced voltages was studied in Williams (1954). This leads to a temperature rise and shorter life time of the machine (Gafford, Duestenhoef, Mosher, 1959). Supplying the motor with unbalanced voltages also decreases the rating of the motor (Berndt & Schmitz, 1964). A basic method was applied in order to study the impact of unbalanced voltages on the losses and its negative effect on the insulation of the motor (Woll, 1975). Protection of the motor against these risks and adjusting the relays were proposed in reference (Cummings, Dunki-Jacobs, & Kerr, 1985). In Cummings et al., 1985, Kersting and Philips, 1997 and Kersting, 2000, the impact of unbalanced voltage on the losses of the motor has been investigated. Kersting and Philips have presented a work about discussion of the effect of 0–5% unbalanced factors (Kersting & Philips, 1997). Unbalances on the terminal voltages cause a considerable effect on the stator and rotor copper losses. An increase in the unbalanced voltage at the terminals can increase rotor losses more than stator losses. The reason is that the rotor currents have a larger deviation than the stator currents. In addition, the resistance of the rotor bars is higher for unbalanced voltage (Gillbert, 1980). Even 5% unbalanced voltage increases the losses considerably, and lead to temperature rises that could damage the motor. Therefore, use of a motor connected to a large unbalanced voltage is not normally allowed. In reference (Lee, 1999), the temperature rise due to unbalanced voltages has been studied experimentally. The results of this study showed that monitoring amplitude of the only single component (2f) is enough for the detection of unbalanced voltage in induction machines. It is implemented by measuring the amplitude of fault indicator component appeared in the spectrum of stator current zero crossing instants (ZCT). The frequency component is found to be function of the level of voltage unbalance. In this system, for the classification purpose and determination of the most dominant parameter in the detection of the supply unbalance data mining process is used. Data mining applied in a wide range of applications such as in the field of prediction, production control, and in the area of medical. However, in the available literature, there are very limited numbers of studies on the fault detection in the electrical motors with using data mining approach. The unbalance voltage is similar caused by extensive motor failure or even catastrophic phenomena, differing only in degree of unbalance. In order to detect unbalanced phase voltages or currents are readily identified, by the abnormality in induction motors several approaches are looking for the presence of negative phase sequence component. A small voltage unbalance produces a large negative sequence current flow in induction motor that will produce excess heating. The 5% voltage unbalance produces a stator negative sequence currents of 30% of full load of current. With this extra current, the motor experiences a 40% to 50% increase in temperature rise. In this paper, the authors propose the data mining approach to investigate the level of unbalanced voltages. This approach was successfully used to predict the dominant parameter. In order to analyze the effects of the supply voltage unbalance, the three phase induction machine was simulated with 5% of unbalanced supply with and without any other fault.
نتیجه گیری انگلیسی
It has been reviewed that the voltage unbalance has a potential source of effects on motor efficiency, rotor losses, and temperature rises. For unbalanced motor, fault detection is implemented by monitoring the amplitude changes, in stator current at (1 + 2s)f component and at 2f component in ZCT spectrum with the usage of data mining approach. For full load condition whereas amplitude changes (at 2f) in motor current spectrum are very small, the magnitude changes of 2f component in ZCT spectrum are more evident even for even small supply unbalance condition (in a % 1). With the development of the fault, this value increases. This variation is still in detectable level for various load conditions. In this paper, a M5P algorithm part of data mining approach is successfully applied to determine the supply unbalance condition. The R2R2 value for the predicted unbalance is 0.9639, which can be considered very satisfactory.