A method for early warning of inter-turn short-circuit faults in excitation windings of synchronous motors

Abstract

The invention realizes the fault pre-warning of the inter-turn short circuit of the excitation winding of a synchronous motor based on a long-short-term memory network (LSTM). After the data is collected, it is normalized; then the LSTM network is established and the network model is trained to determine the fault warning value, thereby judging whether the excitation winding is normal. The present invention uses 9 physical quantities as independent variables and 3 physical quantities as output quantities to train the LSTM network; then performs data fusion on the 3 output quantities to realize fault warning. The invention fully considers the physical quantity related to the short circuit between turns of the rotor, avoids the inaccurate measurement of a certain physical quantity or the inaccurate prediction value caused by the interference of the signal, and overcomes the existence of single criteria such as the rotor vibration change method. It can provide early warning for the early fault of the inter-turn short circuit of the excitation winding, and detect the fault even when the load of the synchronous motor fluctuates greatly, and does not need to add an additional measuring device in the synchronous motor, which has the advantages of simplicity and ease of operation.

Description

technical field [0001] The invention relates to the technical field of synchronous motors, in particular to the field of diagnosis of inter-turn short-circuit faults of excitation windings of synchronous motors. Background technique [0002] The normal operation of a synchronous motor is an important factor for maintaining the stability of the power grid, and the inter-turn short circuit of the field winding is the most common fault in a synchronous motor. The main reasons for the inter-turn short circuit of the field winding are: when the winding is subjected to the centrifugal force of the rotor rotation, the burrs contained in the winding will pierce the winding insulation; poor ventilation will cause the field winding to overheat and cause thermal deformation; design defects in the overall fastening of the rotor end winding ;Oil mist, particles and other foreign matter enter the interior of the rotor, etc. When the short-circuit fault is mild, it has little impact on th...

AI Technical Summary

Problems solved by technology

The ideal application state of the detection coil method is the situation of no-load or three-phase short-circuit of the synchronous motor. At this time, the iron core of the synchronous motor is in an unsaturated state, but in actual operation, the magnetic circuit is in a saturated state when the synchronous motor is running with load, and the rotor leakage field Compared with the armature reaction magnetic field is small, so the detection sensitivity is reduced, this method has certain limitations

Images