Inverter fed sensorless controlled variable speed drives with induction machine are widely used in the industry applications, also in wind power generation and electric vehicles. On-line self diagnostic systems implementation is needed for early stage fault detection and avoiding a critical fault. Diagnostic algorithms in modern DSP-based controllers can operate simultaneously with control system functions. In the closed-loop controlled drive measured and estimated signals can be used in diagnosis. Broken rotor of induction machine is not statistically the most common fault, but its relatively long propagation time enables early stage detection of the fault and setting limitations in the control system to avoid a critical fault. Early stage rotors diagnosis is needed also when testing casting inaccuracies in a squirrel-cage rotors related to gas bubbles presence. When designing the drive for electric vehicle usually temporarily overloading of the machine is assumed. Thermal stress and electromagnetic forces are the main causes of rotor damage. Broken rotor detection methods has been investigated for many years [1]. However most of them are designed for open-loop operation of the drive. Closed-loop and sensorless operation gives a new symptoms of the fault which can be used for diagnostic reasoning. Diagnostic of the machine operating in the sensorless control system requires taking into account influence of state observers and control system, regarding the uncertainties associated with nonstationary diagnostic signals, nonlinear model of machine, machine parameters estimation accuracy and tuning parameters of the observers and controllers. A few articles concerning FOC or DTC closed-loop control system with broken rotor has been published so far. The paper deals with sensorless control system of the induction machine with broken rotor. A new symptoms of the fault in the control system are showed and experimentally verified. Compensating effect of the control system is also investigated. Possibilities of broken rotor symptoms utilization in the diagnostic system are discussed.
Authors
Additional information
- Category
- Aktywność konferencyjna
- Type
- publikacja w wydawnictwie zbiorowym recenzowanym (także w materiałach konferencyjnych)
- Language
- angielski
- Publication year
- 2010
