The Research And Application Of The On-line Correction Method Of The Power System Comprehensive Load Model

In recent years,smart gird is going to be the orientation of future power grid in worldwide,also in China,the construction and development of smart gird is deepening.Since the smart gird take benefit to economy development and resource utilization efficiency,a brand new challenge in load modeling of power system field was posed by the real-time interaction of smart grid.How people to model the load model which combined the real-time load data and correctly reflect the real-time load characteristics,under the setting of smart gird,be a reliable basis of power system operation,planning and scheduling,is a researching problem remained.An integral synthesis load modeling and real-time correcting parameters of load model theory,which aim to achieve real-time modeling technology for smart gird,was introduced in this paper,and the introducing aspects of the theory was posed in followings:to begin with,introduction of classical load modeling theory,component-based approach and measurement-based approach,was given;what's more,the feasibility,basic principle and process flow of the load model are expounded.It is proved that the method could effectively revise the load model parameters,and thus the real-time characteristics of the load could be reflected.Secondly,the FCM(fuzzy C-means)clustering algorithm,used in approach of substation composition analysis of the typical electricity industries,was discussed in depth;The FCM clustering algorithm is weak on sensitivity to initial values,easily fall into local optimum,the artificial definition of numbers of clusters breaking objectivity,etc.,due to the weakness of FCM clustering algorithm,density-function-based clustering algorithm was purposed in this paper,it taking the advantages of quickly and efficiently cluster plus no need to define the numbers of clusters,etc.These two clustering algorithm mentioned are such complementary so that an enhanced FCM clustering algorithm,integrated these clustering algorithm,was proposed;by introducing the fuzzy clustering validity coefficient,the effectiveness and superiority of enhanced-FCM clustering algorithm are proved through an example of coefficient contrast.Moreover,the article introduces the key method of the typical industry composition ratio analysis method of the substation comprehensive load model parameter online revision theory,and expounds the basic principle and the step flow 'of the method.The results show that the method is applied to the prediction of load model parameters.It also proves that it has certain application value.To end with,the article shows the software platform which can realize the online correction of load model parameters.The platform in the Windows operating system Visual Studio environment,functional modules including:substation and power industry classification and integrated functional modules,load model parameter identification function module,model parameters online correction function module.The platform has been operating in the provincial power grid scheduling control center is successful and poses practical significance.