Enhancing Power Grid Data Analysis with Fusion Algorithms for Efficient Association Rule Mining in Large-Scale Datasets

Abstract

Against the backdrop of the rapid development of information technology, the total amount of data has exploded, and efficient association rule mining methods for large-scale datasets have been studied. Conventional rule mining algorithms are subject to electrical constraints when working, and their convergence speed and data noise are currently the main problems they face. In order to accelerate the working process of the algorithm, this study introduces a data warehouse into the K -Means algorithm. The time series and voltage interaction functions are connected with the long -and -short-term memory network for efficient information analysis of power grid data, generating fusion algorithms. When mining association rules in electrical data sets, the pruning strategy is used to effectively reduce the search space, thus improving the efficiency of the algorithm. The pruning strategy applied in this study is confidence pruning, and the rules that do not meet the conditions are deleted by calculating confidence. For large-scale data sets, the distributed computing framework can dispatch tasks to multiple computing nodes, thus accelerating the computing speed. In the voltage interaction experiment, the storage and processing cost of electrical data is very high, so the research on using compression technology to reduce the dimension of storage space will reduce the computational complexity of the algorithm. The study conducts experiments on the Netloss dataset and three models, including long -and -short-term memory networks, to verify the superiority of the fusion algorithm. Under the same experimental voltage, the circuit power flows of the four models were 0.37, 0.64, 0.79, and 0.82A, respectively, indicating that the algorithm effectively controlled the electrical dataset. Its measurement accuracy was the highest among the four models, at 91.7%. The experimental results showed that the fusion algorithm proposed in the study had precise control ability in power grid datasets, and effectively mined association rules on large-scale datasets. This paper proposes a novel approach for efficient association rule mining in large-scale power grid datasets. A fusion algorithm is developed combining clustering, neural networks, and association rule learning. The technique incorporates data warehousing, time series modeling, and voltage interaction analysis to enhance information extraction from electrical data. Experiments on the Netloss dataset demonstrate the algorithm's effectiveness for power flow control and measurement accuracy compared to standard methods. Results show significant improvements in active power loss and network loss metrics as well. This research provides an important foundation for scalable analytics in smart power systems.