For Project Enquiry +91 9962 067 067

MACHINE LEARNING METHODS FOR ATTACK

DETECTION IN THE SMART GRID

ABSTRACT

Machine learning algorithms are used to classify measurements as being either secure or attacked. An attack detection framework is provided to exploit any available prior knowledge about the system and surmount constraints arising from the sparse structure of the problem in the proposed approach. Well-known batch and online learning algorithms (supervised and semi supervised) are employed with decision- and feature-level fusion to model the attack detection problem. The relationships between statistical and geometric properties of attack vectors employed in the attack scenarios and learning algorithms are analyzed to detect unobservable attacks using statistical learning methods. The proposed algorithms are examined on various IEEE test systems. Experimental analyses show that machine learning algorithms can detect attacks with performances higher than attack detection algorithms that employ state vector estimation methods in the proposed attack detection framework.

EXISTING SYSTEM	PROPOSED SYSTEM
EXISTING CONCEPT:- Attack detection problems in the smart grid are posed as statistical learning problems for different attack scenarios in which the measurements are observed in batch or online settings. In attack detection methods that employ state vector estimation (SVE), first the state of the system is estimated from the observed measurements.	PROPOSED CONCEPT:- Machine learning algorithms are used to classify measurements as being either secure or attacked. The relationships between statistical and geometric properties of attack vectors employed in the attack scenarios and learning algorithms are analyzed to detect unobservable attacks using statistical learning methods.
EXISTING ALGORITHM:- K-Nearest Neighbor	PROPOSED ALGORITHM:-       Machine Learning Algorithms.
ALGORITHM DEFINITION:- The most frequently observed class label is computed using majority voting among the class labels of the samples in the neighborhood, and assigned as the class label of si.	ALGORITHM DEFINITION:- Machine learning methods have been widely proposed in the smart grid literature for monitoring and control of power systems. We establish the relationship between statistical learning methods and attack detection problems in the smart grid.
DRAWBACKS:- Attack detection problems in the smart grid One of the challenging problems of this approach is that the Jacobian measurement matrices of power systems in the smart grid are sparse under the DC power flow model	ADVANTAGES:- The proposed algorithms are examined on various IEEE test systems. Experimental analyses show that machine learning algorithms can detect attacks with performances higher than attack detection algorithms