Research On Probabilistic Production Simulation Considering Wind Power Correlation

With the large-scale utilization of wind energy,wind power penetration in each country and area is gradually increasing.Consequently impacts on the power system brought by the highly variable wind power and the problem of abandoning wind resulting from kinds of restrictions become increasingly obvious.Power system probabilistic production simulation(PPS)could simulate and assess the power system on its economy and reliability in the long term,and is helpful for the analysis on impacts of wind power integration and on wind power accommodation.But the current PPS methods for the power system with wind power seldom consider the wind power correlation and system's chronological information at the same time,resulting that the simulation results are not detailed and accurate enough to reflect the actual situation.In response to this problem,a new probabilistic production simulation method considering wind power correlation is studied in this research.Firstly,a probabilistic time series model of wind farms' output is established.For wind farms with correlated output,a reference wind farm is selected and according to its wind speed history data,a probabilistic time series model of its wind speed is established.Then based on the joint distribution between wind speed of the reference wind farm and output of each wind farm,the conditions probability of wind farm output is achieved.Combining the wind speed model and the wind farm conditional probability,the probabilistic time series model of each wind farm's output,which could reflect their correlation,is eventually developed.Secondly,probabilistic time series models of power sources are established and probabilistic production simulation is conducted.The universal generating function(UGF)is introduced to develop the model.For conventional power sources,random failures are considered to establish the UGF model.For correlated wind farms,UGF models are established based on the aforesaid probabilistic time series model of wind farms' output.The UGF models are consolidated into one model representing an equivalent power source.Then the power source is matched with the load chronologically,and the probabilistic production simulation is conducted to calculate the energy production and reliability index of the power system.Finally,UGF-based method of wind power accommodation ability calculation and wind power abandon analysis is put forward.System's daily minimum technical output is calculated and modeled by UGF based on the PPS method mentioned before.This model is calculated with the UGF models of load and wind power output,achieving the probability distributions of wind power accommodation ability and the expectation of abandoned wind energy.Suggestion on improving the accommodation ability is put forward according to results of the simulation case.The proposed UGF-based probabilistic production simulation method could take both wind power correlation and the power system temporal information into account,and give a more detailed and accurate reflection of the actual operation of the system.Therefore the research has certain theoretical and practical values for studying on wind power integration.