Font Size: a A A

Fault Diagnosis Of Squirrel Cage Induction Motor Fed With Variable Frequency Power Based On Information Fusion

Posted on:2009-04-21Degree:MasterType:Thesis
Country:ChinaCandidate:M LuoFull Text:PDF
GTID:2132360245970585Subject:Control theory and control engineering
Abstract/Summary:PDF Full Text Request
It is well known that squirrel cage induction motors (SCIM) are widely used in modern industry. The electric equipments'fault diagnosis technology and SCIM's fault theories have been developed decades. To improve the dynamic performance and adapt motors to the frequent startup and stop, moreover the speed adjustment advisably in the real-time status, the variable frequency power serving motors is so prevalent that it may be the absolutely necessary subsystem of drive system. But it makes more problems in motors fault diagnosis.In this paper, based on the recent development of theories and practices on broken rotor bars fault diagnosis of SCIM, the mechanism of fault is analyzed particularly. A mathematics simulation modeling of SCIM supplied with variable frequency power is introduced, which respectively in normal condition and in broken rotor bars conditions, grounded on the principles of multi-loop model and parameters, sinusoidal pulse width modulation. The inverter power harmonics in motor's stator currents and instantaneous power spectrum affect the fault diagnosis specially, which makes the difference between the variable frequency power and single industrial frequency power. Classical current spectrum signature analysis approach deduces the data information fusion diagnosis: the extended Park's vector approach and the mean instantaneous power approach. The relevant simulation experiment results are given. Though the integral synthesis and compare, the mean instantaneous power approach is better than the extended Park's vector approach on broken rotor bars diagnosis of SCIM fed with variable frequency power.Wavelet analysis is introduced in the further analysis of the features extraction of fault signal spectrum. Because the conventional Fourier transform and short-time Fourier transform are good at stationary signal processing, but wavelet transform has the better analysis ability in time-frequency domain and subdivision ability in local spectrum, which makes it better than Fourier transform and more suitable for analyzing the motor's spectrum full of frequency components. At the same time, Morlet's complex wavelet is used to extract features in startup current spectrum for decision of the fault happening exactly, which is an auxiliary method.
Keywords/Search Tags:variable frequency power supply, squirrel cage induction motor, fault diagnosis, information fusion, wavelet analysis
PDF Full Text Request
Related items