Co-optimization Of Generation And Reserve For AC/DC Hybrid Systems With High Wind Power Penetration

With the increasing depletion of fossil energy and the increasing destruction of the ecological environment,actively developing new and renewable energy sources and optimizing energy structure have attracted worldwide attention.As a kind of clean,renewable energy with the most mature technology and the lowest cost,wind power has been rapidly developed worldwide and has become the fastest growing energy source in the world.In 2012,China has become the country with the largest installed capacity of wind power in the world,and its newly installed capacity has consistently ranked first in the world.Together with the rapidly increasing installed wind farm capacity,the increasingly severe issue of "wind curtailment" has emerged.This problem,on the one hand is because of the poor ability to consume wind power nearby,which is due to the adverse distribution of energy and load in our country,on the other hand,is because of the ability to support each other across provinces and regions has not been fully utilized.In view of the above problems,based on the background of bulk AC-DC transmission system in China,this paper focuses on the cooperation of multiple resources,commodating wind power by coordinating multi-area generation and reserve resources,and studies the reserve decision of power system under high ratio of wind power access.A scenario generation and reduction method is proposed,which takes into account the correlations between wind power forecasting errors among different wind farms.Accurately modeling the uncertainty of wind power is an important basis and guarantee for the rational allocation of reserve capacity and correctly developing system scheduling plan.In this paper,kernel density estimation algorithm,Copula theory,Monte Carlo sampling algorithm and improved K-medoids clustering algorithm are applied to generate and reduce wind power scenarios with correlation.Based on the historical statistics of wind power prediction errors of multi wind farms,the kernel density estimation is applied to calculate the probability distribution of wind power forecasting error.The maximum likelihood estimation method is used to estimate the unknown parameters in the joint copula functions.The shortest Euclidean distance method is used to select the most appreciate copula function which best fits the actual correlation of wind power forecasting errors.The Monte Carlo algorithm is usd to generate the scenarios.Aiming at the shortcoming of the slow convergence rate of conventional clustering algorithms,this paper proposes an improved K-medoids clustering algorithm which combines variance method and density method to optimize the initial center for scenario reduction.The example uses wind power forecasting error data of five wind farms in China for generation and reduction of scenarios with correlation,which verifies the effectiveness of the proposed method.Considering of the poor ability to consume wind power nearby and the shortage of spinning reserve of thermal unit,a single area coordinated dispatching method of generation and operating reserve considering multiple resource coordination is proposed.The theory of Conditional Value at Risk(CVaR)is introduced to assess the risk of loss of load and the risk of wind power curtailment.Under the conditional risk framework,a reserve plan is made,which better balances the relationship between risk and cost and better accounts for low-probability and high-risk events.The charging process of electric vehicles and proactive wind power control to provide reserve capacity are modeled.A two-stage stochastic unit commitment model considers correlation between wind power forecasting errors and coordination of electric vehicles and proactive wind power control is constructed.The example analyzes the influence of wind power forecasting error correlation on reserve plan,and verifies the positive effect of the proposed model in this paper on mitigating the reserve shortage of thermal power generation.In order to alleviate the problem of wind curtailment which is caused by adverse distribution of energy and load and inadequate cross-regional wind power consumption.In the context of the AC/DC hybrid system,this paper proposes a decentralized co-optimization model of generation and reserve in multi-area power systems considering the flexible operation characteristics of HVDC transmission lines.Firstly,the net load forecast error scenario is regarded as a multi-state unit,and the expression of sub-area reserve demand and lost load and wind power curtailment is established to decouple the calculation of reserve demand from optimization process,which reduces the calculation scale.Then,taking into account cross-regional consumption of wind power and reserve support between adjacent regions,a centralized co-optimizasion model of generation and reserve in multi-area power systems considering the flexible operation characteristics of AC/DC tie-lines is proposed.Finally,an improved analytical target cascading algorithm is proposed.The centralized model is decomposed into an upper-level master problem and lower-level sub-problems.The master problem decides the dayahead transmission plan for the HVDC/HVAC tie-line and the cross-regional reserve plan,and the lower-level dispatch centers independently solve their SCUC problems in parallel.The example analysis verifies the effectiveness of the distributed algorithm proposed in this paper.By comparing with the regional independent scheduling model,it verifies the positive role of the cross-regional dispatch model in alleviating wind power curtailment,promoting clean energy consumption,and optimizing allocation of resources within the interconnected region.