Intelligent techniques for energy management and deduction of various intrusions in smart grid

Abstract

This thesis suggests a hybrid methodology for energy management newlinemethod in smart grid. The suggested hybrid methodology is the combination of newlinetwo intelligent techniques Artificial Cell Swarm Optimization (ACSO) and Vapor newlineLiquid Equilibrium (VLE) and hence, it is named as ACSO-VLE. The main newlineobjective of the recommended method is to minimize the generation cost. The newlinepower demand is frequently monitored and tracked by ACSO. The VLE improves newlinethe perfect association of micro grid by providing better schedule between the newlinerenewable energy sources. In the micro grid operation, the first approach is the newlineplanning of scheduling between the renewable energy sources to reduce the charge newlineof electricity. The aim of the second approach is to satisfy the power balance newlineequation and reduce the impacts of forecasting errors. The suggested model is newlineworked out in MATLAB/Simulink platform. The effectiveness of the ACSO-VLE newlinemethod has been examined through the comparison of the proposed method with newlinethe existing methods. newlineAn Intrusion Detection System (IDS) has been used to find or avoid newlineprobable cyber-attacks in a specific time without upsetting the usual function of newlinethe system. Some existing methodologies are there to notice the disturbances in newlinesmart grid framework, though they have consumed an old dataset. Due to this, the newlinerate of detection is low and the accuracy is poor in the results. Consequently, some newlinemachine learning techniques can be used along with new Knowledge Data newlineDeveloper (KDD-99) dataset for finding different anomalies in the smart grid. The newlineinformation from smart grid cloud storage and KDD99 are pre-processed and newlineoptimized by Improved Aquila Swarm Optimization (IASO) technique. For the newlinecalculation and classification of intrusions, the probabilistic Recurrent Neural newlineNetwork(RNN) classifier has been used. newline