Coordinated Dispatch Method Of Power System Considering High Proportion Of Wind Power Uncertainty

The utilization of energy in the world is changing towards green and low-carbon,clean and efficient,sustainable and renewable,but the rapid development of clean energy has a negative impact on the security,stability and reliability of power system.As a clean and renewable energy,wind power has the characteristics of strong randomicity,obvious intermittence,large fluctuation range,irregular fluctuation frequency and peak inversion.Therefore,the grid-connected large-scale wind power brings uncertainties to the system,which poses great challenges to the optimal dispatch of the system.The research on coordinated dispatch method of power system considering high proportion of wind power uncertainty is of great significance for making reasonable generation scheduling,ensuring real-time balance of power grid and promoting the integration of high proportion of wind power.Focusing on the modeling method of uncertainty of wind power output,an deeply study has been carried out on the uncertain optimal dispatching technology of power system with high proportion of wind power in this paper,including analysis of probability distribution characteristics of wind power,wind power combination prediction model,operation risk assessment of power system,stochastic reserve dispatch of power system,decentralized and coordinated economic dispatching of inter-regional interconnected power grid.and the concrete contents are as follows:1)It is very important to study the probability distribution characteristics of wind power for constructing the uncertainty model of wind power output.Therefore,based on the measured historical wind power data collected by wind farms,the fluctuation characteristics of wind power at different time and space scales are analyzed,and then the temporal and spatial variation and fluctuation characteristics of wind power are counted.Then heavy tail distribution function is proposed to describe the probability distribution of wind power.Subsequently,relative entropy is introduced as an evaluation index to measure the goodness of fit distribution.By comparing and analyzing the goodness of fit of different probability distribution functions,the rationality of using heavy-tailed distribution function to describe the probability distribution of wind power at different time and space scales is verified.2)In order to make up for the defects of a single wind power prediction model and improve the prediction accuracy of the model,a wind power combination prediction model is proposed by using various time series decomposition techniques and artificial intelligence prediction algorithms.Empirical mode decomposition(EMD).variational mode decomposition(VMD).long short-term memory(LSTM)neural network,deep belief network(DBN)and simulated annealing(SA)are applied to establish the EMD-LSTM and VMD-SA-DBN prediction models of wind power respectively,which are the basic models of the upper prediction model.The extreme gradient boost(XGBoost)algorithm is adopted as the lower prediction model,and then the prediction results of the upper two basic prediction models are combined and optimized to obtain the prediction value of the combination forecasting model.The measured historical data collected by wind farms are used for example analysis,and the effectiveness of the proposed combination forecasting model is verified by comparing the prediction accuracy of different models.3)To quantify the operational risk of high-proportion wind power system,an operational risk assessment model based on Copula function is proposed,which considering the spatial correlation of multiple wind farms.In order to avoid the problem of constructing the joint probability distribution of multi-wind farm output in the generation of multi-wind farm output scenarios with time-space correlation,a Copula function model of wind speed correlation of multi-wind farms is constructed.According to the possibility and influence of node voltage or line transmission active power exceeding the limit,a comprehensive risk evaluation index system is constructed by introducing the probability and severity of exceeding the limit,so as to comprehensively and accurately quantify the system operation risk caused by wind power integration,and the influence of wind power uncertainty on system operation is mapped to the risk index through probability power flow calculation.An example of IEEE 39-bus system is utilized to verify the effectiveness of the proposed risk assessment model.4)Power dispatching needs to reserve enough spare capacity to balance the fluctuation of wind power output,which is used to reduce the operation risk of the system caused by the uncertainty of wind power output.Considering the risk of violating security constraints,a day-ahead stochastic reserve dispatch model is proposed by using the stochastic chance-constrained optimization theory,which taking into account the uncertainty of wind power output and the output characteristics of automatic generation control(AGC)units.First of all.based on Markov chain Monte Carlo(MCMC)method,the wind power scenario was simulated,and the number of generated scenarios was automatically determined by using the violation probability of chance constraints and confidence parameters.Secondly.the reserve output characteristics and operation constraints of AGC units were comprehensively considered in the reserve dispatch model.By incorporating the reserve cost into the objective function,the optimal output plan of conventional units and the optimal allocation coefficient of reserve capacity of AGC units can be obtained simultaneously.Finally,the constructed stochastic reserve dispatch model is equivalent to a semidefinite programming(SDP)problem,and then solved directly.The validity of the stochastic standby dispatch model is verified by an example analysis of IEEE 30-bus system.5)In order to solve the problem of high proportion of wind power grid-connected consumption in multi-area alternating current and direct current(AC/DC)hybrid system,a decentralized and coordinated robust economic dispatch model of cross-regional interconnected power grid with high proportion of wind power is proposed.The affine adjustable robust optimization(AARO)method is used to deal with the uncertainty of wind power output.Firstly,the accommodated wind power interval model is proposed.The accommodated wind power interval and the automatic generation control participation factors are optimized simultaneously to maximize wind power consumption of multi-are AC/DC system.Secondly,a flexible regulation model of high voltage direct current transmission(HVDC)converter in multi-are AC/DC system is proposed.Then,the accelerated alternating direction method of multipliers(ADMM)distributed solution algorithm is introduced to solve the proposed dispatch model to ensure the privacy of information between regional power grids and the independence of dispatch.Finally,two-area 12-bus AC/DC system and Three-area 275-bus AC/DC system are taken as examples to verify the effectiveness of the proposed model.