Modelling And Optimization Of Wind-Hydro-Thermal Day-Ahead Generation Scheduling

With development of economic and improvement of living standard,energy demand in China is steadily increasing.In the meantime,environmental problems caused by the energy consumption become serious.In order to improve energy structure,wind power industry in China has developed rapidly.However,owing to the uncertainty of wind power,wind power integration brings a great challenge to day-ahead generation scheduling of power system.In this paper,an optimization method for short-term wind-hydro-thermal load dispatch is developed considering the forecast errors of load and wind power.Moreover,considering the demand of environmental prediction and emission reduction,a mathematical model for stochastic multi-objective day-ahead generation scheduling with pumped-storage stations and wind power integrated is built.At first,the characteristics of several forms of power generation,as well as the development of wind power industry in China,are summarized.Meanwhile,the researches about economic dispatch,unit commitment,short-term hydro-thermal generation scheduling and generation scheduling with pumped-storage stations and wind power integrated are summarized.Secondly,the mathematical model of short-term hydro-thermal scheduling and the basic principle of evolutionary predator and prey strategy(EPPS)are introduced.EPPS is applied to solve short-term hydro-thermal scheduling problem.The simulation results show that compared with other published methods,EPPS can find a better solution.The next,based on the mathematical model for short-term hydro-thermal generation scheduling,a mathematical model for short-term wind-hydro-thermal generation scheduling based on chance-constrained programming is built.The model can help to set reasonable spinning reserve capacity based on probability distribution characteristic of net demand and given confidence level.In order to prove the efficiency of EPPS,EPPS is applied to solve short-term wind-hydro-thermal generation scheduling problem.Besides,the model for short-term wind-hydro-thermal generation scheduling based on chance-constrained programming provides basis to set spinning reserve capacity.System operators can comprehensively consider the security and economic of power system,and set reasonable confidence level in order to determine the spinning reserve capacity.At last,this paper proposed a mathematical model for stochastic day-ahead generation scheduling(SDAGS)with pumped-storage stations and wind power integrated.Based on the probability distribution characteristic of the forecast error of net demand,Latin hypercube sampling(LHS)is utilized to generate a large number of scenarios,and scenario reduction technique is applied to reduce the number of scenarios.Thus,tipical scenarios are retained.Four objectives,including minimizing the mean value of thermal units cost in different scenarios,minimizing the variance of thermal units cost in different scenarios,minimizing the average emission of carbon dioxide,and minimizing the average emission of polluted gas,are considered in the model.Besides,Multi-objective group search optimizer with adaptive covariance and Lévy flights(MGSO-ACL),priority list method and primal-dual interior point method are applied to optimize the SDAGS over 24-hour period,aiming at reaching a compromise between the minimization of expected total cost of thermal units and the variance of them.Furthermore,a decision making method based on evidential reasoning(ER)approach is utilized to determine a final optimal solution considering average carbon dioxide emission and expected polluted gas emission.The simulation results show that the proposed method can effectively solve the SDAGS,and that evidential reasoning approach can obtain a final optimal solution successfully.