Research On Compressed Sensing Of Power Quality Disturbance Signals

With the scale of power increasing rapidly and the development of the science and technology,new power electronic devices and a large of non-linear load are applied in the power grid, making the harm of power quality in power system serious, the power quality problems become more and more frequent. At the same time, power quality problems have been more extensive attention. In order to solve the problem of power quality, only first to analyze and identify the data of power quality signal can be carried out for the next step. But the traditional electric energy quality analysis and processing process is based on the Nyquist sampling theorem that acquisition, compression, decompression and signal processing. While this tends to produce large amounts of data and occupy a large storage space. Compressed sensing theory is emerging in recent years, the theory, which breaks through the framework of Nyquist sampling theorem, by a few measurement signals to reconstruction of the original signal, greatly easing the pressure of storage space, reducing the computational complexity.The main research of this paper is to apply compressive sensing theory to the analysis of power quality problems. The main research work is as follows:(1) according to the power quality standard of IEEE, the mathematical model of 6 kinds of common power quality disturbance signal is constructed, and the sparsity of these 6 kinds of disturbance signals are verified by the FFT and the DCT.(2) In view of the traditional measurement matrix that the complexity of the calculation is high, the hardware implementation is complex and so on.There are two measurement matrixs are builded that the computational complexity is low, the hardware realization is simple and with the structure feature, the experimental results show that the reconstruction results slightly better than the traditional random measurement matrix.(3) using a compressed sensing theory of harmonic detection method based on by harmonic signals of the unknown observation, and then solving sparse, finally get the harmonic signal in the harmonic information, through simulation verified the validity and accuracy, but also has a good anti noise property.(4) Proposed a new power quality disturbance signal compression perception method. By extracting the feature of the obtained sparse vector and then into neural network classifier can achieve classification and recognition, and it is proved by the simulation, the method of 6 kinds of disturbance signal has a high recognition rate, in the noise free case, the recognition rate reach 98.71%.