Power fundamental frequency detection based on orthometric average chirp-z transform

Chirp-z transform (CZT) is usually used to refine frequency spectrum to analyze special frequency. The spectral resolution by CZT can be small as one wants but sometimes is wrongly considered as the same as accuracy. This paper discussed some characters of CZT about accuracy for monotone frequency signals. The CZT error decreases when the sampling time increases, and fluctuates periodically and diminishingly until reaches the resolution. The error distribution is related to the initial phase of the sampled signal. The spectral results are approximately symmetrical for two orthometric signals. A new frequency analyzing method, orthometric average chirp-z transform (OACZT), is presented to analyze the power fundamental frequency more exactly. Simulation results show that the accuracy of OACZT is about 0.003Hz with sampling time of 0.2s, better than CZT with an accuracy of about 0.1Hz.