With the deployment of advanced metering infrastructure (AMI), an avalanche of new energy-use information became available. Better understanding of the actual power consumption patterns of customers is critical for improving load forecasting and efficient deployment of smart grid technologies to enhance operation, energy management, and planning of electric power systems. Unlike traditional aggregated system-level load forecasting, the AMI data introduces a fresh perspective to the way load forecasting is performed, ranging from very shortterm load forecasting to long-term load forecasting at the system level, regional level, feeder level, or even down to the consumer level. This paper addresses the efforts involved in improving the system level intraday load forecasting by applying clustering to identify groups of customers with similar load consumption patterns from smart meters prior to performing load forecasting.Index Terms-Advanced metering infrastructure (AMI), k-means clustering, load forecasting, load patterns, load profiles, neural network-based load forecasting, smart meters.
Better understanding of actual customers' power consumption patterns is critical for improving load forecasting (LF) accuracy and efficient deployment of smart grid technologies to enhance operation, energy management, and planning of electric power systems. Though technical literature presented extensive methodologies and models to improve LF accuracy, most of them are based upon aggregated power consumption data at the system level with little or even no information regarding power consumption of different customers' classes. With the deployment of Advanced Metering Infrastructure (AMI), new energy-use information becomes available. AMI data introduces a fresh perspective to perform LF, ranging from very-short-to long-term LF at the system level, or down to the consumer level. However, one critical step to realize these benefits is to develop data management and analysis process to transform AMI data into useful information. This paper addresses the efforts involved in preparing residential customers AMI data as inputs for LF, and introduces the idea of how the preprocessed data could be further enhanced by identifying customers' consumption patterns through the application of clustering. Grouping load profiles based on consumption behavior similarities will reduce the variability of load which is going to be predicted, and therefore, reducing the forecasting error.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.