با استفاده از داده AMR برای برآورد بار برای تجزیه و تحلیل سیستم توزیع

With the development of distribution automation (DA) and other advanced applications in distribution systems, the real-time monitoring and control of distribution systems becomes possible. Now there are only a limited number of real-time measurements on the distribution systems. The load monitoring and estimation of customers can be an important source of information used by the distribution analysis applications. In recent years, an increasing number of automated meter reading (AMR) systems have been installed. AMR can provide customer consumption information and other data such as confirmations for outages and restoration. In this paper, a load estimation algorithm is discussed. The proposed algorithm makes use of the above information that AMR provides as its input. It also incorporates time series forecasting method and the use of the customer load curves to improve the accuracy of individual customer real-time load estimates. This method with the use of AMR data has excellent load estimation results. This method demonstrates how AMR data can be used for other functions besides billing.

To effectively control the power distribution system that is becoming more and more complicated, operators should have comprehensive, accurate and real-time knowledge about the system. Many distribution automation (DA) systems have been and will be installed in the distribution systems. Some of these systems can transmit real-time data back to the control center. If enough measurements can be obtained accurately, continuously and reliably, the operator or an advanced application can understand the exact status of the system and decide how to most effectively manage the system. Transmission level loading allows for real-time values of conditions. While the distribution system does not have that feature today, one of the opportunities is to use the available AMR data to provide load models and estimations for various distribution system analysis techniques. Besides distribution state estimation as mentioned below, load estimations for individual customers can be useful for load balancing, restoration-cold load pick up and planner forecasting for upgrades for lines, transformers and other upstream power system equipment. These applications do not need exact values but estimates that provide a first-order snapshot of the distribution system as it operates day-to-day. Additional metering can be done on key locations to move from estimations and pseudo-measurements to real time data. However, as this is not practical for all the distribution system, using the AMR data is a useful initial tool. Various constraints make it impossible to have a perfect picture of the system. First, for economical reasons, measurement instruments cannot be installed at every place where the measurements are needed, so the data are incomplete. Second, the measured data are subject to error or lost communication, so the data may be inaccurate, unreliable and delayed. State estimation is one approach to reduce these constraints. State estimation techniques have been developed and used on the transmission level for over 30 years. Transmission system state estimation is considered to be the heart of a modern energy management system (EMS). Now it becomes possible and important to apply state estimation techniques to distribution systems as well. The distribution system state estimation can provide the real-time system states to a distribution management system (DMS) enabling operators to monitor and control the operation of the distribution systems [1], [2], [3] and [4]. However, because of the limited number of real-time measurements in the distribution systems, the state estimator cannot acquire enough real-time measurements, so pseudo-measurements are necessary for a distribution system state estimator. One good source of the pseudo-measurements is the load estimation at the transformer node, providing the combination of several customer loads below it. Many techniques and approaches have been investigated in load modeling and estimation in the last three decades [5], [6], [7], [8] and [9]. Some previous algorithms make use of the transformer capacity or the customer billing information combined with coincidence factors to estimate the real-time load [5]. Others use weather information [6] and [7]. Miu and Wan [9] developed load estimation techniques using the limited measurements available including some meter measurements. In recent years, an increasing number of automated meter reading (AMR) systems have been installed. AMR can provide customer consumption information and other data such as confirmations for outages and restoration. And these data can be used to simplify the load modeling algorithm and make it more accurate. AMR is the remote collection of consumption data by meters over telecommunication lines, radio, power line or other links. Meters are installed on the customers’ houses. The meters transmit the data to the intermediate communication controllers at very short time intervals, and the data are routed by these controllers to a central computer [10]. In this paper, a real time load estimation algorithm is discussed. The algorithm estimates the load of a metering point. The proposed algorithm makes use of the above information that AMR provides as the input. And it also incorporates a time series method and the use of the basic customer class load curves to improve the accuracy of load estimates. With more and more utilities installing AMR equipment, one of the challenges will be to develop algorithms that use this valuable information for load modeling, monitoring and system analysis. The next section outlines the load estimation modeling algorithm using AMR data.

The development of AMR systems provides a great resource for load estimation in the distribution system. The proposed research activities have extended theory in several areas. First the algorithm uses historical data from a given meter (not just a class of load) to help predict the load for the future on that device. The algorithm then uses the time series algorithm coupled with historical data to model and estimate the future load patterns. The proposed load estimation algorithm utilizes the newly installed AMR system to calculate more accurate load estimations. The algorithm is a first step toward a more comprehensive individual customer load model for use in distribution state estimation. This data analysis provides an opportunity to evaluate how we use AMR data in distribution system operations in the future. While many of the AMR installations are being justified for billing purposes, this research shows that the AMR data can be used to develop load estimations that can be extended to look at distribution load forecasting related to maintenance and upgrading of systems, pseudo-measurements for a real-time distribution state estimation and extended analysis of the distribution systems to help maintain power quality and reliability. Future work includes testing the algorithm on more AMR data as well as developing trending data models predicting constant overall increases or decreases in daily power consumption based on prolonged weather conditions. The key to these further applications will be getting open meter reading data sets that provide information about the customer as well as locations so weather data can be integrated directly into the algorithm.