Design And Implementation Of Transient Power Quality Detection System Based On Ridgelet Transform

The safe and stable operation of the power system has an important indicator to the power grid level. Modern power electronic devices produce a series of disturbing questions to the power system, such as voltage swells, sags, harmonics, short-term power outages, pulse transient disturbance, high-frequency oscillatory disturbance, voltage flicker and so on. These problems directly affect many aspects of people’s production and life. Therefore, to correctly detect and identify power system transient phenomena is an important and valuable research content.In this paper, we are focus on eight categories of common typical voltage disturbances. The basic principles of power quality disturbances are studied and the basic characteristics of common power quality disturbances are analyzed. The mathematical model and the corresponding frequency domain detection feature are introduced. In order to resolve the problem of harmonics and features, testing technology in transform domain is used and the common power quality detection method is introduced. The basic principle of wavelet transform and S transform are analyzed. Based on the introduction of them, we introduce a ridge wavelet transform method, and the relationship between the ridge wavelet transform and wavelet transform is discussed. The advantages and disadvantages of ridge wave transform to transient power quality detection are analyzed. On the other hand, to identify the type of power quality disturbances, the method of neural network is study. The basic characteristic of BP neural network and probabilistic neural network is analyzed. On the basis of SOM network, we construct a fuzzy SOM neural network based on fuzzy judgment method and the structural layer was analyzed.On the other hand, most of power quality detection equipment is developed by using application specific integrated circuit implementation. The disadvantages of this method in the secondary development or intelligent visualization are still less efficient, long development cycle, upgrade difficulties and other shortcomings. With the development of virtual instrument, labview and matlab simulation technology become powerful tool in software designed. Based on the method of ridge wavelet transform and network, a power quality detection and identification system is designed, which has many advantages on precision, intelligent, cost, visualization and scalability. Finally, the overall operation of the process and the specific implementation process are described in detail. The function of this system is of multi-disturbance disturbance signal simulation and identification. Simulation and experiment results show that ridge transform can detect power quality multi-disturbance effective and fuzzy SOM neural network can classify it accurately. The problem of voltage superimposed disturbance classification can be resolved successfully from both qualitative and quantitative ways.