Digital Signal Processing Techniques In The Detection Of Power Quality Applications

Nowadays, electric power has been widely used in our society and it is the milestone for the modernization of a country. Coming along with the development of technology and economy, consumer's demand for electric power has been increasing day-by-day. Meanwhile the power quality has become a hotspot that both the consumer and the power corporation are concerned with.After the concept of power quality (PQ) was defined, people have made great effort to research the new PQ detection technologies, which have much better performances than ever before. In the beginning analog technologies played a very important role in PQ detection, but today most detection technologies are based on digital technologies, such as digital signal processing etc.At first, this thesis briefly introduces the power quality's concepts, definiens and standards. Then the research scope of this thesis is set, according to the actuality and the trend of the development of PQ detection technologies. As a technical background introduction, this thesis gives some analysis of those digital signal processing algorithms, which are common in PQ detection, such as Fourier Transform, Short Time Fourier Transform, Wavelet Transform etc.The third part of this thesis focuses on the PQ detection under condition of stable state. By using non-sine signal as the research object, some successful detection algorithms, such as virtual value of U/I, power and degree of unbalancedness etc, are obtained. Moreover frequency detection, the core of the third part, is analyzed in detail and based on the complex magnitude of harmonic signal, a novel algorithm for computing the frequency deviation in real-time is proposed.The PQ detection under transient state is discussed in the fourth part of this thesis. Flicker is a very typical transient PQ disturbance and can be regarded as same as the interharmonic disturbance in mathematics. Considering its short time duration, a new Flicker detection algorithm based on CWT is presented. In order to detect the transient harmonic, a novel algorithm 'Wavelet packet transient harmonic detection algorithm based on anti-aliasing db wavelet filters' is provided.At last, simulations' results are given showing that those new detection algorithms, derived from this thesis are effective, simple and fast ways which can meet our requirements. Their performances are better than some commonly used detection algorithms.