A Study Of The Joint Optimization Problem Of Unit Commitment And Reserve Decision

At present,as the energy shortage and environmental pollution problems are aggravating,the development of renewable energy has become the first target of all countries in the world.In recent years,the proportion of installed capacity of renewable energy in China’s power grid has been rising year by year,so the high proportion of renewable energy grid-connected power generation has become an inevitable trend of its energy development strategy.On the one hand,large-scale renewable energy grid-connected power generation can effectively alleviate the energy shortage and improve the climate and environment.On the other hand,the uncertainty of its power output brings huge challenges to the optimization and dispatch of the power system.In this context,the traditional deterministic dynamic economic dispatch is no longer applicable,and new optimal dispatch methods are urgently needed.Therefore,this paper is devoted to the optimal dispatch of grid unit combination and reserve decision considering uncertainties in order to realize the complete consumption of a high percentage of renewable energy as well as the optimal control of each power source in the system.The main work is as follows:(1)A day-ahead unit commitment and reserve joint optimal scheduling model based on scenario analysis and conditional value-at-risk is proposed.Use multi-scenario technology to simulate the uncertainty of system load and renewable energy output,analyze system power generation and reserve demand through wind power random output and load forecasting scenarios and error scenarios,and target the volatility of renewable output and load in the system,Introduce conditional value-at-risk(CVa R)theory to quantify the risk of uncertainty to system scheduling operation.Considering the balance between system unit commitment,reserve decision and risk loss,an optimal dispatching plan that takes into account economy and reliability is obtained.The analysis of calculation examples verifies the rationality and effectiveness of the proposed model and method.(2)The conditional value-at-risk theory is introduced to establish a joint optimization model of multi-type power unit commitment and reserve with integrated consideration of nuclear power and stored energy.Also,the CVa R theory is applied to quantify the risks arising from system uncertainty on scheduling operation,and the impact of different scheduling schemes on the economics and reliability of system scheduling operation is compared and analyzed.Meanwhile,the impact of different risk attitudes on system dispatch is studied.Ultimately,the example results show that the unit combination and reserve decision of different types of power sources can significantly improve the system operation economy,reliability as well as reduce the risk loss under extreme conditions of the system.Besides,it can meet the scheduling decision requirements for large-scale renewable energy grid connection.(3)To address the problem of large errors in the prediction accuracy of renewable energy day-ahead power,the day-ahead and intra-day unit commitment scheduling model based on model predictive control theory is proposed.With economic optimality as the objective function,the day-ahead optimal scheduling obtains the day-ahead unit start/stop plan and generation plan by solving the optimal flow,as well as the day-ahead optimal scheduling results as reference values are distributed.Afterward,to cope with the system uncertainty,the intra-day optimal scheduling takes the scheduling plan value issued in the corresponding period of the day-ahead optimal scheduling stage as the reference value,and the objective function is to minimize the deviation of the day-ahead optimal scheduling result,so as to establish the intra-day optimal control model and call the GAMS quadratic programming QCP solver to solve it.More importantly,the day-ahead scheduling plan value is continuously revised to realize the full consumption of renewable energy in the grid.