Power Definition with Asynchronous Sampling

Power definition is established and accepted in electrical literature for quite some time, but non-active power definitions are another story. In this paper we will use two power definitions with improved averaging algorithm for calculating active and reactive power of an arc furnace plant. Both of these definitions are easy to implement in practical applications. Since rated frequency of the plant is 50 Hz, with a 128 µs sampling time, there are 156.25 samples per cycle. Non-integer number of samples would result in cyclic calculation error. This paper proposes an optimized averaging algorithm to account for asynchronous sampling rate. In the algorithm, the remaining 0.25 of the sample is considered as 157th sample with quarter of normal value. Simulated results from sinusoidal and non-sinusoidal voltage and current data show that the optimized algorithm helps to improve accuracy of the calculated power components. Considering an actual Electrical Arc Furnace (EAF) Plant with 90 to 108 MVA total apparent power, the measured voltage and current data are used to calculate active and reactive powers. Using simulation results both methods with normal and optimized algorithm are analyzed