Study On Electric Signal Digital Measuring Technology And Its Application

Modern power industry demands that power systems and electric devices must run under digital supervise, control, and protection, which are based on the real-time, accurate, and reliable analyzing and measuring of electric signals. Applying digital signal process theory, micro-computer technology, and power system automation technology, this dissertation researches on theories and methods for more accurate, faster, less cost, and more practicable electric signal measuring, and pay more attention on improving signal sampling methods and developing reasonable electric parameter estimating algorithm. The main work and the contributions are given as follows.1) The traditional synchronizing methods of AC sampling, including hardware synchronous sampling, software synchronous sampling, and asynchronous sampling, are examined; the factors that cause synchronization error and uniformity error are then found out. Some novel methods, designed to minish these errors and improve measurement accuracy, are whereafter proposed. Especially to the software synchronous sampling, which is most common used now, three novel methods are designed. The first adjusts the sampling space-time dynamically so that the sampling moments approach the ideal ones; the second modifies the actual samples toward an ideal sampling sequence by means of linear interpolation algorithm; the third evaluates the signal fundamental period in real-time and modifies the sampling period. All the methods are easy to implement, can be widely applied and need less additional work. It is helpful for the soft- and hardware design of the electric measuring instruments.2) In order to remove the non-period components from the electric signals at fault conditions, a new approach that modifies the sampling data is present. Compared with other approaches, this one is suitable for any algorithms because it is not designed to a special algorithm. It can be easily applied to microprocessor-based protection to eliminate the effect of non-period components.3) According to the characteristics of power electric signals, a novel DFT correction algorithm, which is based on the improved DFT spectrum correction technology and suitable for electric measuring, is present. The implement technique of the algorithm is detailed through its "short time-window" and "long time-window"algorithms. By choosing different length of data window, sampling frequency, and corresponding windows, the algorithm can meet the different demands of computing speed and precision at various applied situation, such as measurement, control, and protection in power system. The algorithm can. be easily and widely employed to get a rapid and precise estimating of electric signals.4) Based on the "short time-window" algorithm of DFT correction, a new algorithm designed for automatic synchronization device is present. It performs DFT on two sampling sequences of a voltage signal, and then makes use of the difference of two DFT analysis results to estimate signal frequency, amplitude and phase. Even if the sampling period is fixed, the algorithm can get high estimating precise when the signal frequency changes. It solves the problem that the traditional algorithm can not estimate the parameters of both generator and power system voltages when the two signal's frequencies are different and, as a result, the synchronous sampling is difficulty to perform.5) A new window family, called as Rectangular Self-Convolution Windows (RSCW), is present. The m-order RSCW is developed by convolving m rectangular sequences, and the values of 1 to m-1 order derivatives of its amplitude-frequency characteristic is zero. As a result, the windows of this family are especially suitable for electric signal analyzing.6) An improved algorithm for estimate power harmonics parameters is present. The algorithm first weights the sampling sequences with RSCW to reduce the interferences between the harmonics due to spectrum leakage, in which boosts the precision of harmonics estimation. Whereafter, a DFT correction based interpolation algorithm suitable for RSCW is given, and some novel measures are taken to reduce the computational load.7) A new algorithm is present to estimate the RMS value of voltage and current and the active power. This algorithm converts the problem of estimating RMS value and active power to the one that extract the zero frequency component from a specially built signal, and then designs a milt-passband digital filter, whose response is a RSCW, to remove the harmonics components from the signal. In this way, accurate measurement can be obtained even if the synchronous sampling errors occur.8) The results of the above researches have been applied in the practices of electric signal measuring; some devices, such as "the equipment for reactive power compensating and power harmonics filtering", "the microcomputer-based monitor for power harmonics filter", "the microcomputer-based automatic synchronization...