Data Preprocessing Using Hybrid General Regression Neural Networks and Particle Swarm Optimization for Remote Terminal Units

作者：Chen, Wen-Hui^*; Chen, Jun-Horng; Shao, Shih-Chun

来源：International Journal of Control Automation and Systems, 2012, 10(2): 407-414.

DOI：10.1007/s12555-012-0220-0

• General regression neural networks
• particle swarm optimization
• supervisory control and data acquisition (SCADA) systems
• remote terminal units

Data corruption in SCADA systems refers to errors that occur during acquisition, processing, or transmission, introducing unintended changes to the original data. In SCADA-based power systems, the data gathered by remote terminal units (RTUs) is subject to data corruption due to noise interference or lack of calibration. In this study, an effective approach based on the fusion of the general regression neural network (GRNN) and the particle swarm optimization (PSO) technique is employed to deal with errors in RTU data. The proposed hybrid model, denoted as GRNN-PSO, is able to handle noisy data in a fast speed, which makes it feasible for practical applications. Experimental results show the GRNN-PSO model has better performance in removing the unintended changes to the original data compared with existing methods.

• 出版日期2012-4