PRZEGLĄD ELEKTROTECHNICZNY, ISSN 0033-2097, R. 92 NR 2/2016 121 Tomasz ADRIKOWSKI, Dawid BUŁA, Marian PASKO Politechnika Śląska, Instytut Elektrotechniki i Informatyki doi:10.15199/48.2016.02.34 The use of p-q control in single-phase active power filter for dynamic reactive power compensation Abstract. The subject of the discussion is the use of single-phase active filter for compensating dynamic changes of reactive power which may occur in office buildings. This circuit consumes reactive power with high dynamic changes at the level of minutes or seconds and fractions of seconds, making it impossible to efficiently compensate reactive power with standard capacitors-based compensators. Modified p-q algorithm was used to control single-phase active power filter. The considerations were supported by simulations carried out in MATLAB-Simulink environment. Streszczenie. Przedmiotem rozważań jest zastosowanie 1-fazowego filtru aktywnego do kompensacji dynamicznych zmian mocy biernej jakie mogą wystąpić w przypadku odbiorów w budynkach biurowych. Do sterowania 1-fazowym filtrem aktywnym zostanie wykorzystany algorytm p-q. Rozważania zostały parte symulacjami przeprowadzonymi w środowisku MATLAB-Simulink. (Zastosowanie sterowania p-q w 1-fazowym energetycznym filtrze aktywnym do dynamicznej kompensacji mocy biernej). Keywords: single-phase active power filter, extension p-q theorem, dynamic reactive power compensation, power factor correction. Słowa kluczowe: 1-fazowy energetyczny filtr aktywny, teoria mocy p-q, dynamiczna kompensacja mocy biernej, korekcja współczynnika mocy Introduction Costs resulting from the poor power quality on a global basis can be substantial [4]. Therefore measurements and identification of problems associated with power quality has become increasingly important. These problems depend on the type of the used loads. Different events will occur in heavy industry and different in housing estates or office buildings. From the point of view of the consumer, those parameters are important which may cause generation of additional costs in form of penalties. This is, for example, reactive power for which the relevant provisions in contracts with electricity provider are applied [6]. There are additional charges for inductive and capacitive reactive power consumption. The paper presents an analysis of reactive power variation on the basis of the measurements in a system composed mainly of office computers [2], carried out in one of the buildings of Faculty of Electrical Engineering of the Silesian University of Technology. The obtained measurement results were used to develop a simulation model of single-phase active power filter (APF) intended to compensate dynamic changes of reactive power. The model of APF was simulated in Matlab- Simulink. For control algorithm the modified p-q method has been used. As part of the simulation study reactive power compensation for a sample of variability reactive power load was carried out. Characteristic of reactive power variation of loads in office buildings Electrical loads in buildings such as offices are mainly computers, printers and small electronic equipment but also UPS devices, servers, air conditioning and ventilation equipment. All of these devices can cause problems with the power quality. Significant here is the large variability of consumed active and reactive power and harmonic generating in currents. Admittedly most of them are low- power devices but they may be hundreds or even thousands of them in one place. Specific for office buildings is also load distribution during the week. Figure 1 shows the total power P tot (10 minutes average) for the analyzed building in a period of one week. As shown the load reaches a maximum value during the day but decreases and has a constant value during the night. In the case of reactive power one can highlight two specific aspects of the analyzed network. The first one is the character of reactive power consumed per day. Loads in office buildings during the day consume inductive reactive power. At night, while the majority of office equipment is in standby state they consume capacitive reactive power. Because the capacitive reactive power imposed fees are charged independently of the active power these values can be significant [9]. Figure 2 shows the average reactive power Q tot in a period of one week. Fig. 1. Average total active power for the measured building in 1 week period Fig. 2. Total average reactive power Q tot for the measured building in 1 week period 11:35 11:30 11:25 11:20 11:15 11:10 11:05 11:00 10:55 10:50 10:45 10:40 Q min , Q max , kvar 11.0 8.8 6.6 4.4 2.2 0 Q min Q max 11:40 t, hh:mm Fig. 3. An example of the maximum and the minimum value (Q max , Q min ) of reactive power during one hour