Feasibility of Generic Load Modeling for State Estimation in Distribution System

In the distribution system, the load is always changing. As the penetration of low-carbon technology is getting higher, the load landscape becomes more complex. In this case, the importance of distribution system monitoring is growing in order to increase the stability and reliability of the whole power system. It is expensive for meter installation and maintenance at all distribution nodes as the distribution system generally has large scale. It is expected to find an approach to supplement or even substitute the measurements. A generic load model has been proposed to solve the problem. This model can be considered as input of adaptive state estimation. The feasibility has been discussed in the paper. A preliminary mathematical model has been developed from the real world data. This model proves the level of load uncertainty can be aggregated of 2 different components.

I. Introduction

As the penetration of low-carbon technologies in distribution networks increasing the variability and complexity of loading patterns, it is more important to monitor the distribution system in real-time to enhance security and reliability of distribution system. Transmission systems are generally fully monitored by supervisory control and data acquisition (SCADA), while the number of smart meters in the distribution system is limited. The installation and maintenance would be costly if meters were installed at all nodes in the distribution system. There are three methods proposed to make distribution network completely visible. They are state estimation, load allocation and probabilistic load flow [1] .