Research Of Power Quality Intelligent Detection And Disturbance Classification

Recently, with the development of technology and national economy, the power quality has been hihgly concerned. At the same time, due to the use of large number of power grid load, many power quality disturbances are produced, disturbances have made power quality decline. Power quality can be affected by many aspects. Power quality problem has been more serious by now, has become the focus and attracted attention of many other fields.In this paper, it firstly introduces some concepts of power quality and national standards, and expatiates on the definition, features and causes of power quality problems. Because power quality disturbances can be divided into steady state and transient power quality disturbances, detection of power quality means the detection of steady state and transient power quality problems.The steady-state power quality disturbances include voltage deviation, frequency deviation, three-phase imbalance, and harmonics, etc. This paper gives national standards and measurement methods of different steady-state disturbances, and focuses on the harmonic detection method. It uses the windows and interpolated Fast Fourier Transform to reduce the calculation error caused by the non-synchronous sampling and algorithm defects, and compares calculation errors with different window functions.According to the international standards developed for the transient power quality disturbances, the transient power quality disturbances include voltage sag, swell, interruption, transient pulse and oscillation. Based on the detection methods, this paper uses S transform to detect and locate the transient power quality disturbances, extracts the line of the highest frequency from the S mode matrix as the basis for disturbance detection and location.While transient power quality disturbances are made classifications, single feature can not be fully representative of transient disturbance, it may cause errors.In order to overcome the weaknesses, this paper extracts features of transient power quality disturbances based on S transform and wavelet packet decomposition, and gets the combination of features. Finally, it designs a binary tree support vector machine classifier with clustering ideas. And it also compares the simulation results with the other classifiers’.Actually, power quality disturbances are complex in the power system, power quality disturbances always exist with other disturbances and multiple disturbances happens. This paper creates some typical multiple disturbance models, uses different methods to analysis multiple disturbances, improves the feature extraction method and builds the classifier with dynamic tree structure. Finally, the classification of multiple disturbances is achieving. This paper also designs the software to verify detection algorithms of the steady-state and transient power quality and the implementation process of the software algorithms when steady-state or transient power quality problems happen. This paper also designs the system architecture of the hardware platform of power quality detection, and gives the hardware design of each module.