Search

Real-time monitoring is a key tool for several automated features in distribution system. However, due to the limited number of real-time measurements, distribution system monitoring generally relies on load modeling procedures. This paper presents field results for a real-time monitoring tool that was implemented on a Brazilian distribution utility to provide load values for service restoration software. The implemented tool comprises two steps. The first step consists of preprocessing routines that provide the required information for a real-time load estimator that is executed in the second step. The field results provide evidence of the practical viability of the implemented real-time monitoring tool, demonstrating the tool overcomes some usual limitations of real-time load estimation procedures. More specifically, the implemented tool is able to handle multiple and different types of real-time measurements, treat real-time measurements on the feeder and not only in the head of the feeder, handle obviously bad data, and to satisfy the performance requirements on practically sized distribution systems. Besides proposing and confirming a real-time monitoring tool on a real distribution system, this paper shares valuable utility experience on providing new insights and the challenges faced for real-time monitoring projects. [ABSTRACT FROM PUBLISHER]

The art of estimating future values of a random process, based upon previously observed or estimated values, is usually known as a priori estimation, prediction, or forecasting. Power system state estimation process can be enhanced if state/measurement forecasts are incorporated into it. Important research efforts have been made in this direction bringing a fresh perspective to the state estimation problem. This paper (Part I) presents a comprehensive survey of forecasting-aided state estimators. It gathers up-covering a period of three decades-research results on the main benefits achieved by state estimators with forecasting capability regarding: data redundancy, innovation analysis, observability, filtering, bad data, and network configuration and parameter error processing. Aspects of modeling, forecasting techniques, and computational effort are also addressed. The second of this two-paper series presents the results of the implementation of a forecasting-aided state estimator in the energy management system of the LIGHT Services of Electricity, a company which provides electric energy to Rio de Janeiro, Brazil. [ABSTRACT FROM AUTHOR]

This paper is the second part of a two-part paper addressing power system forecasting-aided state estimation. Part I presented a comprehensive survey on the subject, including future prospects. Part II is devoted to presentation and discussion of numerical results obtained with the implementation of a forecasting-aided state estimator in the energy management system of LIGHT Services of Electricity, which is company that supplies Rio de Janeiro, Brazil. This estimator is submitted to different situations, such as: normal operating conditions, presence of spurious measurements (including those which are critical), and occurrence of unpredictable sudden changes in the state of the LIGHT system. [ABSTRACT FROM AUTHOR]