Study And Development On Measurement Software For Power Quality

This thesis focuses on describing the importance of the power quality measurement and control. It summarizes the status quo of study today, and also analyses and compares the advantage and disadvantage of several analytic methods of the power quality. The principle diagram of power quality monitoring in the thesis can monitor and classify almost all of the important problems of power quality. Matlab simulating is used to aim at correlative problems of power quality. The simulation results show that the transient and non-stationary signal can be detected and classified using wavelet transform, which can also used in denoising and filtering. The discrete Fourier transform algorithm can analyze time-domain's stationary disturbance signal effectively and confirm the signal's distributing in the frequency domain. So it is effective to analyze and detect stationary and non-stationary disturbance by the integration of the above two methods. Through this idea, we have developed a set of software of power quality measurement which is called PMVIS using C++ Builder6 and SQL-Server2000. The further improving of PMVIS is ongoing.The first part of the thesis introduces the problems' research background, research status and significance. This part illuminates the work of this thesis, too.The second part of the thesis classifies the problems of power quality. This part introduces all power quality problems.The third part introduces and compares analytic methods of the power quality and their application ranges.Simulation has been done through Matlab in this thesis. A new algorithm of power quality measurement is put forward. The future trends of power quality are discussed. This thesis also brings forward the method of power quality measurement, i.e. the integration of WT (Wavelet Transform), FFT (Fast Fourier Transform) and ANN (Artificial Neural Network).This thesis finally concentrates on how to realize the PMVIS software. The algorithm is designed. A set of signal with noise is simulated finally. PMVIS is applied to analyze the example. The error is analyzed in the last.