Voltage Quality Disturbances Analysis Algorithm

As a result of the development of national economy and the wide use of new style electronically controlled equipments, power quality has been an important issue to both suppliers and consumers of electricity. Not only the deviation of voltage and frequency from the rating value or other traditionally focused problems, but also the dynamic disturbance phenomena including oscillation and pulse are concerned. In order to improve the power quality, a monitoring and analyzing system must be established to detect, estimate and classify different disturbances correctly. The algorithm of analysis on power quality disturbances has been discussed in this thesis.Firstly, on the basis of a through comparison among several current methods for power quality analysis, a comprehensive analysis method using windowed FFT and wavelet transform is proposed, which uses modulus maxima detection according to the principle of singularity detection. The simulation results demonstrate that the method adopted has good performance in both time and frequency domains. With the method, the harmonic frequencies and phases can be got accurately, the singularity characteristics are presented clearly, and the disturbing-time and degree of disturbance can be informed too.However, this algorithm can't distinguish from the disturbances effectively. After a great deal of research, an improved analysis method is illustrated, which integrates abc-dq transform and the former proposed algorithm. Because of the drawbacks of wavelet transform on the detection of transient disturbances, abc-dq transform is used in the detection of disturbances in most situations. Voltage sags, swells, interruptions, impulses, phase steps, frequency deviations and noises are classified by abc-dq transform, and the algorithm based on modulus maxima detection is applied to classify oscillations and harmonics. In the meanwhile, characteristic parameters are also calculated. Extensive studies by MATLAB simulation evince that, the accuracy rate of the method for nine familiar disturbances types all exceeds 95%. The validity and the accuracy of the method are pretty good. The application prospect of this algorithm is very abroad due to its high accuracy and simplicity.Finally, the research foreground for power quality disturbances analysis and the work that should be done to perfect this thesis are presented.