Research On Electrical Signal Parameter Measuring Algorithm And Its Application

Timely and accurate measurement of power system electrical signal parameters, this is an important prerequisite to ensure the security and reliability of power system and a variety of power equipments, and this can also provide accurate scientific basis. The modernization, automation and intellectualization realization of power system and electrical equipments are based on the real-time and accurate measuring of electrical signal. The analysis and research of many problems can not be executed without the electrical signal measurement in the field of power system. Therefore, a detailed analysis and in-depth research of the electrical signal measuring theories and its implementation methods are carried out in this paper, and several improved algorithms are proposed to enhance the real-time and accuracy of electrical signal measurement on this basis.The sampling of input signal is a very important step which should be firstly carried out in the signal measuring. The common software synchronous sampling methods have some defects such as sampling interval error, error accumulation increasing and synchronization error, and the influence of these errors on the measurement accuracy can not be ignored. In order to inhibit this synchronization error, an improved software synchronous sampling method is proposed in this paper. The real-time change of electrical signal cycle can be exactly tracked by the zero-crossing-over triggering circuit, and the synchronization error can be reduced by some simple software calculations accurately. So, the approximate synchronous sampling data can be obtained.In the fault situation of power system, the non-periodic DC component which is decayed according to the exponential function is often existed in voltage and current signal. When the traditional full-cycle Fourier algorithm processes this fault signal directly, some errors will be produced, thus an improved Fourier algorithm which can eliminate the errors accurately is proposed in this paper. This improved algorithm is based on the principle that integral of input signal including the decaying DC component and summation of the sampling data are not zero in a full-cycle, this algorithm does not need addition of the sampling data and the approximation calculation in its derivation process, and this can calculate the algorithm error caused by the decaying DC component and its parameters, this algorithm error is subtracted from the algorithm result of input fault signal.In the spectral analysis of electrical signal and the digital filter design, application of a window function is an important issue. On the important role of a window function, the general principle of performance optimization for the cosine combination window function and its selection method are presented by the mathematical formula derivation in this paper. Spectral leakage caused by the truncation processing of input signal is related to the problem that the window function and its derivatives have continuity at the boundary respectively. If the order of derivative with the continuity is high, the data truncated by a window function is smoother at the boundary, thus the spectral leakage can be effectively suppressed.The discrete Fourier transform is widely used in the harmonic spectral analysis of power system. But this algorithm is afflicted by the spectral leakage and the picket fence effect in the asynchronous sampling condition. In order to suppress this phenomenon, the frequency domain estimation method of harmonic signal, which is based on the windowing interpolated Fourier algorithm, is considered in this paper. The spectral leakage can be further reduced by the window function whose frequency-domain characteristic is good, but the algorithm process using this window function is very complex. For this reason, the resolution method which can solve the high order equation of the frequency estimation error is proposed in this paper, the proposed method can be easily carried out by microcontroller. In addition, the rectification formula of the frequency estimation error with Blackman window is derived to reduce the interference influence of the fundamental component spectral leakage on the other harmonic components.Hilbert-Huang transform theory is a new time-frequency analysis method. This method is widely used to analyze the non-stationary signal whose frequency is variable according to the time. In this paper, the electrical signal measurement method based on Hilbert-Huang transform is thoroughly considered. The envelope fitting problem exists in the empirical mode decomposition process of this transform, thus an improved algorithm which can eliminate the undershoot phenomenon exactly is proposed in this paper. Firstly, this improved algorithm determine the undershoot intervals which are generated by the cubic spline interpolation. Secondly, the envelope intervals of undershoot regions are compensated by the monotone piecewise cubic polynomial interpolation. The modified envelope can wrap the original signal tightly, that does not affect the characteristics of signal itself, and it can keep the smoothness of previous envelope as well as possible.The simulation comparison analyses of the proposed algorithms are carried out to verify the superiority and practicability in this paper.