Stationary and Dynamic Behaviour of a Speed Controlled Synchronous Motor with cosℐ - or Commutation Limit Line Control

Abstract

In actual realizations of thyristor-controlled synchronous machines for high speed ranges, converters with D.C. link are usually applied. Normally, the converter for the synchronous motor is realized in a three phase thyristor bridge, and the gating of the thyristors is controlled by a position transducer on the rotor shaft. The reactive power for the commutation of the rectifier is supplied by the synchronous machine driving the synchronous motor in an overexcited state. The control angle of the synchronous motor, which is defined as the angle between the armature field and excitation field, is applied as constant value in actual implemented equipments, so that the synchronous motor is working by nominal power on the commutation limit line of the converter, according to the best possible cosℐ of the synchronous motor. In all other states it is working with a rather worse cosℒ than by nominal power, which represents a disadvantage of these methods.In this paper a control method is presented allowing to control the synchronous motor with constant resulting flux in the machine and cosℐ = 1 in the whole speed and torque range, by changing the control angle in function of the state of the machine. The condition for such an operation is a self-commutated converter on the machine side of the D.C. link.For applications, in which for reason of cost the converter with forced commutation cannot be used, a further control method is proposed, wherein the cosℐ of the synchronous motor is optimized by maintaining the resulting flux of the machine on a constant value, so that the converter is operating on the commutation limit line within the whole speed and torque range.The control characteristics for the variation of the control angle and the excitation in function of the state of the machine are derived from the computation of the stationary characteristics of the synchronous motor. The dynamic behaviour of the synchronous motor controlled by means of these methods is depending, as shown in this paper, on a large scale on the working point of the machine.Based on the computation of the linearized transfer functions of the thyristor-controlled synchronous machine it is shown how the control parameters can be optimized for different working points by means of the state-variable representation of the process.