A high-precision digital integrator based on the Romberg algorithm.

An integrator is widely used for measurement in the field of power systems, and it is a key technology in signal processing. According to research on the digital integrator based on the traditional Newton-Cotes algorithm, the high-frequency response of the low-order Cotes formula is usually poor and the design of the transfer function introduced by the high-order Cotes formula is too complex. In this paper, we analyze the error between the composite Newton-Cotes algorithm and the ideal transfer function. One signal was sampled using the normal sampling frequency and the other signal was sampled using half the normal sampling frequency. The two signals were weighted based on the Romberg algorithm. Thus, the precision of the digital integrator was improved, and the design difficulty was reduced for algorithms of the same order. The simulation and test results show that the proposed digital integrator has better transient and steady performance, and also has a lower error, which is less than 0.01%.