Multi-level Scaled Integration Analysis And Optimization Allocation For Renewable Energy In Power Grid

Traditional power generation with fossil fuels brings more and more serious energy scarity and enviormental pollution problems,while renewable energy can provide electricity without pollution and also is abundant.Hence the capacity of the power generation by renewable energy increases rapidly in power system in recent years,but its randomness and intermittent also deepen the uncertainties of the power system.This dissertation focuses on the integration analysis of renewable energy generation into three levels of power grid including ultra high voltage(UHV)network,extra high voltage(EHV)transmission network and distribution network,and studies the optimization allocation of renewable energy generation in power grid considering uncertainty.The main contribution can be summarized as follows.The steady-state characteristics of large scale renewable energy generation transmitted by UHV half-wavelength transmission system are studied.A quasi-steadystate model of UHV half-wavelength and DC hybrid system is built.Then a voltage and reactive power control method is proposed for combining sender and receiver based on the voltage and current distribution characteristics.Considering various transmission power and load power factors under different scale of renewable energy tansmisssion,the voltage and current distribution characteristics of the half-wavelength transmission and the impacts of power variations of DC lines on the voltage distribution of halfwavelength system are analyzed.As the transmission power fluctuations generated by renewable energy,a combined voltage and reactive power control method for sender and receiver is proposed for UHV half-wavelength and DC hybrid system based on steady-state power flow characteristics by switching DC line filters and adjusting reactive power compensation devices at sender and receiver.The transient characteristics of UHV transmission system and its impacts on power transfer capacity of renewable energy generation are studied.The impact mechanism of power changes in DC fault cases on half-wavelength transmission system power fluctuations are revealed and the method of DC power modulation in system emergency power supporting is proposed.Firstly,the stable operation range of half-wavelength transmission system is analyzed and the rotor angle stability of the hybrid system is discussed under DC commutation failure and DC blocking.In addition,the mechanism of the impact of power changes upon DC commutation failure on half-wavelength transmission system based on impulse response model of second order system is considered.How power changes induced by DC blocking fault impacts halfwavelength transmission system is analyzed based on step response model to enable stable operation of renewable energy system.A system emergency power supporting method via DC power modulation is presented taking advantage of its fast response characteristics.Finally,the mechanism behind DC commutation failure caused by long distance half-wavelength line failure is studied.The simulation results indicate that the proposed method improves the transient stability of the hybrid UHV half-wavelength system and the power tranfer capacity of renewable energy.Considering the transmission network reactive power and storage allocation optimization under the premises of large-scale wind power integration and distribution network equivalent integration,a particle swarm multi-objective optimization algorithm is proposed based on affine PQ flow decomposition and affine number congestion value sequencing.To solve voltage quality problems caused by large-scale wind power integration,and uncertain power integration in distribution networks at the equivalent points and interactions between the transmission networks and distribution networks,an affine-interval model is proposed.Besides,an affine PQ decomposition method is developed to consider the uncertainties.Moreover,an affine number fuzzy dominance considering integration preference is proposed to evaluate and compare the affine number congestion value sequencing method,which is also deployed in the particle group optimization algorithm to replace the adaptability with congestion value.The proposed method is used in power system reactive power and storage optimization allocation.And the test results show the derived Pareto optimal solution sets make up the over-subjective decisions and the lack of diversity and comprehensiveness of existing methods in solving multi-objective uncertainties.The allocation optimization of large-scaled distributed generation(DG)integration is studied.A multi-objective range determination and double-layer optimization method based on D-S evidence theory and genetic algorithm is proposed for DG allocation in distribution network.To model the impacts of the strong fluctuations of DGs and loads,affine model is deployed to represent their uncertainties.A multi-objective uncertainty optimization model is built for DG allocation in distribution network.The objective functions include minimum investment cost,minimum environmental cost,minimum network loss and maximum benefit.Genetic algorithm is utilized for multi-point search to generate multiple allocations in outerlayer optimization,whereas evidence theory method is used in inner optimization to evaluate the outer layer results while revising the evolution direction of outer-layer genetic algorithm.The simulation results verified the effectiveness of the proposed method in adapting to the uncertainties of DGs and loads.