Adaptive Data-Driven Peak Shaving in Smart Grid Electricity Energy by Advanced Metering Infrastructure Data Analytics

In this paper, a novel procedure is proposed to identify the most efficient group of customers for participating in the peak shaving from utility companies’ point of view. This procedure is based on the smart meter data in advanced metering infrastructure (AMI), data mining and pattern recognition algorithms. Studies implies that customers with different consumption behaviors show different effects on the peak load. Consumption pattern recognition in addition to considering networks condition from consumers’ distribution point of view culminates in the most efficient group of them for this aim. The most efficient selection is made when the expected load profile is achieved by affecting the least number of customer as possible. The analysis and results of this paper confirm effectiveness of the proposed data-driven method. This method is able to reduce the number of affected customers in a peak shaving program by identifying the most efficient group of customers in a near real-time data exchanging in the grid. It should be noted that, the proposed method is implemented on a real dataset related to the Irish anonymized households’ consumption data which is provided from Irish Social Science Data Archive (ISSDA).