| As an important driving force for national development,electric energy has an important impact on the progress and development of all walks of life in modern society.With the advancement and development of science and technology,the grid access load has become increasingly diversified,and the emergence of new power grids has also had a certain impact on the power quality of power distribution systems.Scientific assessment of power quality is the basis for solving power quality problems.It is not only important for improving the power quality of power distribution systems,but also an important means to improve the power market system.Aiming at the problem that the existing power parameter acquisition system can not meet the multi-dimensional evaluation of distributed power grid,combined with the relevant national standards of power quality testing equipment,a distributed electrical information full parameter acquisition system suitable for microgrid is developed.The system can collect the main power parameters and send the data to the cloud for storage through Industrial Ethernet,which can realize simple calculation and visualization.The system has obtained relevant certificates after being tested by Zhejiang Institute of Metrology,and can achieve the accuracy standard of 0.5S class meters.According to the characteristics of power quality data,four different load curves such as stable,gradual,catastrophic and multi-variable in power quality evaluation are given,and the correlation characteristics of different load curves are analyzed.A method for measuring the similarity of power quality indicators based on data Boolean operation is proposed.The method obtains the similarity measure between indicators by the geometric features of the time series to be evaluated on the coordinate axes.According to the similarity value,the evaluation data can be subjected to dimensionality reduction processing,thereby reducing the calculation amount of the model and improving the calculation efficiency.Aiming at the problem that the power quality evaluation result can not reflect the dynamic characteristics of the data,a dynamic directed evaluation model of power quality based on LIHM is proposed.By analyzing the characteristics of power quality data,a homogenization processing method for power quality indicators considering index margin is proposed.Aiming at the inability of dynamic estimation of traditional power quality assessment methods,a power quality directed assessment model based on dynamic fuzzy theory is proposed.The current power quality performance is obtained and the possible changes of power quality in the future.In order to provide reference for the power sector to conduct related power quality behavior.Aiming at the increasing power quality data,a power quality evaluation method based on SAX-Hist is proposed.The symbolic polymerization theory is introduced into the power quality evaluation,and the power quality data is symbolized and smoothed by the symbolic aggregation approximation operation,thereby realizing the power quality data compression.The SAX-processed data is further normalized by the Hist method,so that the power quality data distribution can be better reflected.The power quality assessment method based on SAX-Hist can perform denoising and dimensionality reduction on the basis of maintaining the original sequence information and basic morphology,thereby reducing the computational cost and improving the efficiency of data processing,mining and analysis.The distributed power quality acquisition system developed in this thesis realizes the full parameter acquisition of the power distribution system,and has the characteristics of universal versatility.Based on the analysis of the characteristics of power quality data,two different power quality evaluation methods are proposed,which can evaluate the power quality from different angles and have certain practical value. |
PDF Full Text Request