Modelling of Load Devices and Studying Load/System Characteristics

This thesis deals with deriving load models from laboratory measurements and studying the impact of tap-changers on both load and system characteristics. The laboratory measurements were carried out on various types of lighting, thermostat regulated radiators and refrigerators. The results show that lamps have no pronounced dynamic voltage dependence and that a large group of the tested lamps have an almost quadratic static voltage dependence in active power. Both fluorescent and mercury lamps have also a strong voltage dependence in reactive power. The bimetallic regulated radiators act as a constant impedance load during a time period of tens of seconds, but have a pronounced dynamic behaviour in a time range of minutes. Unexpectedly, the mean power was changed in an opposite direction to the voltage, i.e. when the voltage step was negative the mean power increased and vice versa, and the original mean power as it was prior to the disturbance was not regained. This effect is explained by two resistors in the bimetallic switch. The electronically regulated radiators also act as a constant impedance load during a time period of tens of seconds. The dynamic behaviour after a voltage step, in a time range of minutes, is a damped oscillation in mean power , where the prior mean power is regained. Refrigerators show no dynamic voltage dependence and active power has a static voltage dependence between constant current load and constant impedance load. Refrigerators have a poor power factor, are uncompensated, and the reactive power has a rather high voltage dependency. The studies on the impact of tap-changers on load and system characteristics are performed using load characteristics and the well-known up-curves. The studies are intended to elucidate the processes described (an increase in load and a decrease in feeding voltage) in an illustrative way.