The amplitude domain characteristics of wind power generation signals cannot comprehensively characterize long-term fluctuations and local rapid changes in the construction of new power systems. To address this issue, a stochastic time series model of wind power signals was first established; Secondly, a bimodal model of the metastability term and the dynamic term of the current amplitude signal was established, and five characteristic parameters and two characteristic functions were constructed in the run range domain; Finally, 24 typical waveform modes in the run range domain were extracted, and a sample library of wind power typical waveforms was constructed to extract typical features of the voltage and current signals of wind power generation. The results provide a comprehensive characterization of the complex features of wind power generation energy signals.

The output of photovoltaic new energy generation has complex dynamic features, which leads to inaccurate energy measurements. In order to explore the features of photovoltaic new energy power generation, a new energy measurement signal matrix was first established, and an unequally spaced interval STFT algorithm was then proposed based on the finite coverage theorem. A time-frequency domain data representation model was then constructed to represent the real and imaginary parts of the time-frequency domain information of the three-phase voltage and current of the energy measurement signal. This leads to a feature matrix which represents the important features in the time-frequency domain. Finally, important time-frequency domain features of photovoltaic new energy signals were extracted, providing a theoretical basis for determining the time-frequency domain feature parameters of the experimental signal for energy meter error testing.

In order to address the lack of a combined waveform mode model that reflects the characteristics of rapid changes in DC amplitude and the unclear impact of dynamic energy signal excitation on the dynamic error of DC energy meters, this paper adopts a mechanism modeling approach. Firstly, mathematical models are established for the signal preprocessing unit, signal conversion unit, and power and energy measurement unit of the DC energy meter. Secondly, a set of typical DC dynamic waveform modes is defined, and mathematical models for four typical examples are established, followed by the construction of a combination waveform mode dynamic energy test signal model. Then, the impact of the programmable gain amplifier (PGA), power filter, and energy accumulation unit on the dynamic error of the DC energy meter is analyzed through simulations using a dynamic energy test signal as the input. Finally, the dynamic errors in the energy meter caused by the PGA gain switching delay, filter length, and power threshold of the energy accumulation unit under dynamic energy test waveform excitation are analyzed. Our conclusions provide a theoretical basis for improving the dynamic error characteristics of energy meters.