With the growing trend of emerging technologies and the global concern for increasing greenhouse gas emissions from fossil fuels, the concept and implementation of smart grids over the past few years have been presented. According to the American National Technology and Energy National Laboratory, the intelligent network must have some features such as self-healing, energy efficiency, smart management, penetration of renewable energy sources, and so on. These features make the future power grid a highly complex network of cyber-physical aspects and propose some demanding requirements in the information infrastructure of the new generation power grids. Some of these requirements are 1- the rapid response to faults, errors and cyber-attacks 2- extensive information management 3- data security and 4- more efficient management of energy sources. The purpose of this paper is to examine and analysis the aforementioned features in the smart grid besides each other. In this regard, we will show the effects of cyber-attacks on smart meter’ data as well as the losses that can influence the demand dispatch programs, and finally detect them based on the density-based OPTICS algorithm using LOF index. The proposed system is an off-grid network that consists of wind turbines, solar system, diesel generator and battery, and its customers which have some dispatchable and non-dispatchable loads that are connected by intelligent meters to the control center. In order to simulate the proposed method, the actual wind energy and solar energy data obtained from the NREL laboratory climate database are used. The results will show for two general modes that are Demand Dispatch program with and without cyber-attack to smart meters' data.

Keywords