Error analysis for electrical power measurements with non-integer-period sampling

When measuring the electrical characteristics of periodic signals using intelligent instruments based on the correlation principle, errors caused by non-integer-period sampling cannot be eliminated completely due to hardware restrictions. Such errors can be reduced with a quantitative correction method designed for precise measurements of the electrical power. For the signals with various harmonic components, the definition of the average power in the time domain and the first-order Taylor formula on sinusoidal function are used to develop an analytical expression for the measurement error and a correction formula. The results indicate that the error can be reduced substantially by properly selecting the sampling frequency and number of periods to reduce the period bias and controlling the initial sampling time so that the product of the initial voltage and current are close to the measured electrical power.