On-line Analysis Method For Integrated Load Characteristics And Development Of Its Automatic Modeling Platform

In recent years,with the rapid development of smart grid and UHV AC-DC hybrid transmission mode,it has put forward new higher requirements for the real-time and accuracy of grid simulation calculation of power grid,and the real-time and accurate simulation calculation results are largely depend on the real-time and accuracy of the load model.Power system load has the characteristics of diversity,dispersion,time-varying,etc.Its structural characteristics and model parameters are difficult to be unified,which will have a greater impact on the operation,analysis and control of the power system.At present,a set of established models and parameters used in simulation analysis and calculation are difficult to meet the requirements of real-time and accurate simulation calculations for load models.To solve the above problems,this paper proposes a complete theoretical system of online analytical methods for comprehensive load characteristics,which aims to provide certain technical support for online modeling of substations.Firstly,aiming at the shortcomings of the daily load curve clustering method in the existing load modeling,a daily load curve dimension reduction clustering method based on feature index dimension reduction and entropy weight method is proposed.The method uses the physical significance of the daily load rate,the daily peak-to-valley difference rate,the daily maximum load utilization time,the peak period,the valley period load rate,and the maximum and minimum load occurrence time as the characteristic index of the curve.The entropy weight method is used to configure the weights of each index,and the feature weighted FCM clustering algorithm(FW-FCM)is used to achieve accurate clustering of daily load curves.Secondly,aiming at the shortcomings of existing integrated load characteristics analysis methods,an online analysis method based on static characteristics of electrical equipment is proposed.The method starts with the static characteristic equation of the electrical equipment,constructs the power balance equations of the active power of the integrated load port and the powerconsumed by its subordinate electrical equipment under a certain period of time,analyzes the composition and proportion of the electrical equipment at the bus port of the integrated load with the least square principle,and realizes the online analysis of the integrated load characteristics.Furthermore,an online analytical method for the proportion of substation industry based on cosine similarity is proposed.The method is based on the power data of the substation and the daily load curve of the user,combines with the user clustering to obtain the clustering feature vectors of each typical industry,and based on the cosine similarity principle to realize the online analysis of the composition ratio of each industry of the output substation.In order to meet the requirements of modeling data for integrated load online modeling,five types of load characteristic data required for modeling: user survey statistics,user power data,substation SCADA/E file data,fault recorder data,WAMS/PMU data.From the three aspects of data source and characteristics,data preprocessing method,database design and warehousing process,the load characteristics database is constructed to lay a good data foundation for the design and development of substation automatic modeling platform.Based on the design concept of structure,modularization and automation,in Visual Studio 2005/MFC,with the help of the C++programming language,SQL Server,ACCESS database server,Serv-U and other tools,ADO data interface technology,WinExec function and other software technologies,developed a fully functional substation integrated load automatic modeling platform.The platform includes five functional modules,including data processing warehousing and substation automatic modeling module,substation manual completion modeling module,data query management module,model parameter export module and database table clearing module.Its main function is to automatically download,parse,preprocess,filter and save multiple load characteristic data to form a load characteristic database.On this basis,the on-line modeling of substation is realized and a complete substation modeling parameter library is finally formed.At the same time,the simulation model parameters can be automatically exported according to the demand for grid simulation analysis and calculation.Compared with similar load modeling platforms,the modeling platform developed in this paper has a clearer interface,simpler operation,more complete functions and more stable operation,which can automatically realize substation modeling.The platform has been put into operation in a provincial control center,and the model parameters of all110 kV and 220 kV substations were provided for the province for one year.