Power System Anomaly Detection Based on OCSVM Optimized by Improved Particle Swarm Optimization

Abstract

This paper tries to solve anomaly detection, a very important issue in ensuring the safe and stable operation of power system. As the proportion of abnormal data in the operation of power system is very small, a one-class support vector machine (OCSVM) is adopted in this paper, which is suitable for classification of unbalanced data. However, the performance of OCSVM is sensitive to its parameters, and an unsuitable choice will decrease the classification accuracy and generalization ability of it. In this paper, particle swarm optimization (PSO) is used to optimize the parameters of OCSVM. The original PSO algorithm converges slowly and easily falls into local optimum. To overcome this issue, this paper proposes an improved PSO algorithm for parameters optimization, in which adaptive speed weighting and adaptive population splitting are introduced to improve the convergence speed of the algorithm and help the algorithm jump out of the local optimal position. Experiments on standard benchmarks and real power system experimental data sets demonstrate the effectiveness of the proposed algorithm.

Integration of Edge Computing and Blockchain for Provision of Data Fusion and Secure Big Data Analysis for Internet of Things

“…Shang et al [15] used probability density function values and distribution factors to fuse different sensor data but lacked attention to robustness and verifiability. Wang et al [16] used differential privacy combined with deep learning to achieve privacy protection but did not consider robustness. Lin et al [17] proposed a privacy-enhanced data fusion strategy for medical data, but this method is less versatile.…”

Hot-Pressing Furnace Current Monitoring and Predictive Maintenance System in Aerospace Applications

“…This new technique has significant advantages in speed without reducing the generalization performance of the support vector machine (SVM) [ 13 ]. Wang et al improved the effectiveness of a real power system experimentally with parameters optimization particle swarm optimization (PSO) algorithms [ 14 ]. The industry example in the current case [ 15 ] analyzes power consumption data for fault detection, and predictive maintenance, and provides the implementation code on the Web.…”

An electronic transition-based bare bones particle swarm optimization algorithm for high dimensional optimization problems

“…Among these methods, PSO is famous for fast convergence and easy applying. PSO algorithms are widely applied in engineering like route planning [ 8 , 9 ], data clustering [ 10 , 11 ], feature selection [ 12 , 13 ], image segmentation [ 14 , 15 ], power system [ 16 , 17 ], engineering areas [ 18 – 20 ], and so on.…”