Modeling And Optimized Management Of Distributed Microgrid Using Multi-agent Systems

In traditional power systems,centralized control technique is usually employed to supervise and control the entire power systems.These arrangements are largely recognized to be inactive/passive since power flows outward from the utility grid to the consumers or loads.Electrical systems composed of numerous and usually multifaceted components which are difficult to operate and control by efficient ways at the centralized level having problems such as adaptability,mobility,and fault tolerance.Upgrading and redisposition of the power industries allied with increasing saturation of renewable energy resources(RERs)and other traditional power generators at the microgrid(MG)level,the way of power flows in the grid changes.Such arrangements are recognized as active systems since power can flow in bi-directions either from the distributed grid to the central power grid(CPG)or vice-versa.Likewise,for centralized control technique,it may not appropriate to proficiently operate distributed generators(DGs)and accomplish the comprehensive tasks at the MG level as it is ineffectual and might be difficult to control the DGs.Thus,a distributed/decentralized control technique is adopted as an alternative to centralized control method with a foremost objective to establish the MG more competent and could effectively be supervised in a distributed way while the monetary assistances of the MG are exploited simultaneously.At present,energy generation is evolving into a smart distribution system that assimilates several green energy resources at a distributed level assuring to generate clean energy without producing any harmful gases,to have consistent operational procedures,and to improve energy management and supervision arrangements.The distributed energy resources(DERs),such as,solar photovoltaic(PV/SPV),solar combine heat and power(CHP),wind turbines(WTs),alternative grid supporting generators(AGSG),fuel cells(FC)technologies,other micro-turbines(MTs),and energy storage system(ESS)are growing in power systems.An appropriate operational procedure along with proper controlling and supervision of such electrical arrangements have positive impact on the power systems.For optimal practice and management of electrical power in MGs,multiagent systems(MAS)technology has been approved that best supervise the schemes at the distributed level and has numerous applications in the power systems.In this thesis,firstly,a novel MAS based model and optimal management of a MG integrated with RERs at distributed level is proposed.Power generation at distributed level comprises of numerous DERs that deliver power to the consumers critical and non-critical loads.A controlled architecture of a MG based on MAS technique is employed for the finest operations of the grid,energy management and power delivery,and also offers intelligence to the MG at distributed level.For validation of the proposed model,the power generation within the MG was evaluated by simulation under the capabilities of RERs power production,critical and noncritical load demands,and several grid instabilities.The simulation results prove that the proposed model for the MG management based on MAS technique at distributed level offers robustness and high-performance supervision and control than centralized arrangements.Secondly,due to high environmental concerns and ever-increasing fossil fuel costs,generation of clean energy and its optimal management considering the demand side responses are very essential to upsurge the productivity and reliability of the distribution grids.Hence,a multi-energy generation grid(MEGG)has been proposed that investigate the optimal resource management in a grid under various operational conditions to minimize emissions and operational costs.The problem is supervised by agents-based technique considering the generation capabilities and demand side responses.The three-layered technique is presented for the distributed optimal resource management based on MAS framework,where different agents operate and perform tasks autonomously.An optimal energy management scheduling strategy is accomplished considering actions that guarantee the maximum use of generated renewable energy,eradicate the excess use of waste burning power plant(WBPP)and FC,minimize AGSG operational time,and full-load requests satisfaction during the entire day.Similarly,coordination between the MEGG and CPG has also been accomplished for the purpose to exchange power.In order to validate,three scenarios are presented,which show that the comprehensive utilization of DERs can generate a large amount of electricity with less emissions,recover the cost of power generation during peak hours,and enhance the reliability of the entire power grid.Finally,multi-agents based optimal energy scheduling technique for the electric vehicle(EV)aggregator and domestic loads in the residential zone has been presented.This new technology aims to minimize the total cost associated with domestic energy consumption and EVs charging in terms of specific market prices and battery degradation costs subject to the EVs state of charge(SOC)limits.Firstly,agents based optimal technique is presented for the distributed resource management where autonomous agents operate and accomplish their tasks independently that guarantee the maximum extraction of electrical power from the RERs to accomplish the energy demands and make the system more intelligent and reliable.Secondly,to model the actual grid voltage and price uncertainties,the proposed technique is applied in a low distribution network considering the upper and lower bounds of the grid prices instead of the average/estimated prices.The proposed grid has the ability to generate and provide enough power to the domestic consumers and shape the EV aggregator,allowing maximum power charging to fully charge the EV batteries at potentially lower cost.The problem is solved by linear programing considering the generation capabilities of the RERs and EV SOC during the day-ahead period of 24 hours.Besides,to deal with the domestic load and EVs SOC uncertainties,the simulation is carried out based on their consumption's periods during the day while the EVs initial SOC's are estimated on their daily mileage.For validation,the proposed technique is employed at a low voltage residential area and compared,which shows that the proposed technique total profit raised by 16.92% and 5.60% in comparison with the uncoordinated and stochastic techniques respectively and guarantee the optimal energy scheduling that satisfy the consumers load demands efficiently.