| As an important component of renewable energy generation,photovoltaic grid connected systems have the characteristics of volatility and intermittency.In addition,the control strategy of power electronic equipment is relatively complex,especially after a fault occurs on the AC side,its control strategy is not clear.Therefore,mastering the characteristics of DC signals under different operating states is a prerequisite for effective monitoring and control of photovoltaic grid connected systems.It is necessary to study DC electrical quantity measurement methods applicable to existing photovoltaic grid connected systems to meet the needs of system monitoring,control,and analysis.In this dissertation,photovoltaic DC electrical quantities as the research object are taken,the time-frequency characteristics of DC signals are analyzed;The measurement method suitable for photovoltaic DC signals is proposed;A DC signal data synchronous transmission protocol is proposed;A DC measurement device is developed,and laboratory tests and field data tests are carried out.In response to the problem of unclear DC signal characteristics caused by numerous factors affecting DC signal measurement in actual photovoltaic grid connected power generation systems,taking a three-phase single-stage photovoltaic grid connected power generation system as an example,the DC signal characteristics are analyzed,and the time-frequency domain behavior rules of DC electrical signals under different operating conditions of photovoltaic grid connected systems are proposed;The characteristics of DC side voltage ripple and its impact on AC side electrical quantity are analyzed;The measurement object of the DC signal and abstract the mathematical model of a typical photovoltaic DC signal is determine;The timefrequency behavior of photovoltaic DC signal is verified by the field data validation,and the effectiveness of the method is verified.As the difficulty that the appearance of noise and harmonics in the field data of PV DC signals has seriously disrupted the accuracy of useful signal measurement,the characteristics and differences of the statistical characteristics of noise and useful signals are analyzed.A denoising method based on Empirical Mode Decomposition(EMD)combined with adaptive wavelet soft and hard thresholds is proposed,A DC signal measurement algorithm based on Blackman-Harris-Hanning(BHH)cosine combination self convolutional window was proposed to suppress the spectral leakage effect in fast Fourier transform;Simulation and actual recording data show that the proposed method can effectively prevent the influence of Gaussian white noise without affecting the measurement of DC and harmonic signals,and achieve accurate measurement of DC and harmonic components.Based on the above research,the framework of the entire DC measurement device is designed,and the DC signal data transmission method is studied.The hardware selection of the measurement device is carried out.A DC measurement device is developed,which can transmit its AC and DC signals based on the DC transformer without the need to measure the ripple components separately.In addition,a laboratory testing platform is established to test the DC signal types under different operating states on simulated data.The test results show that the denoising method proposed in this paper improved the signal-to-noise ratio by nearly 10dB,the sampling value error of the device was less than 0.42%,and the maximum amplitude error was 0.04%.This verified the effectiveness and practicality of the DC measurement device. |
PDF Full Text Request