Research On Power Load Forecasting And Wind Power Grid Connection Based On Verhulst Grey Theory

In recent years,with the continuous improvement of China's economic level and people's quality of life,the speed of energy consumption is also accelerating.At the present stage,the energy demand of all countries has exploded,and energy trade has become an important factor affecting the diplomatic relations of all countries.On the basis of changing China's energy structure and improving energy efficiency,wind power generation brings many risks to power grid operation due to its own technical problems,that is,the active power output of wind power generation has the characteristics of randomness,volatility and discontinuity.After wind power is connected to the grid,the power quality of regional power grid becomes poor,it is difficult to achieve the effective peak load regulation of the grid,which causes many negative factors to the stable operation of the grid.How to choose a better scheme to reduce the negative impact of wind power generation on the stable operation of power system is an urgent problem.First of all,this thesis selects the newly planned Tieling long-term wind farm as the research object,according to the local wind power acceptance capacity,and takes the wind farm access system scheme as an example to analyze the power grid in this area.When selecting the scheme of wind farm access system,the requirements of regulations such as planning and design of wind power access system of Liaoning power grid in 2011(discussion draft)shall be met.The selection and implementation of the accurate grid connection scheme can not be separated from the accurate load forecasting and on this basis,the supply and demand balance analysis of power supply and load.When analyzing the regional economic situation and the trend of load and electricity growth,this thesis studies and constructs the Verhulst grey model prediction method based on particle swarm optimization algorithm to predict and analyze the future power load growth of the local power grid.Relying on the accurate prediction results,the reasonable operation mode of the power grid and the selection of the appropriate wind farm into the regional power grid access point.Through the study of the above model,the proposed algorithm can meet the accuracy and reliability of load forecasting under the condition of expanding wind power output,so as to improve the trust degree for the next selection and analysis of wind power grid connection scheme,and then carry out a detailed research and Analysis on Tieling long wind farm connected to the grid system.Secondly,calculate the regional power balance under the condition of the proportion of wind expansion power output.Study the wind power capacity of the local power grid and make a detailed analysis.In the calculation,the power balance in Tieling area should be carried out according to the results of regional power load prediction and the power supply installation plan.The power balance of thermal power unit should be carried out according to the full capacity and 20% of the actual unit capacity in the current year as the reserve capacity and the plant load respectively;considering the randomness of the wind farm treatment,the regional wind farm should be carried out according to the 10% of the installed capacity participates in the power balance calculation,and then through the comparative analysis of the results of power balance,the current wind power capacity of Tieling grid is concluded.Finally,according to the local situation at that time,this thesis proposes the corresponding access scheme for the wind power access to the local grid system,and compares it through the power flow mode,access point voltage level and grid connection economics,Finally,the acceptable access scheme is determined,and the validity and feasibility of the scheme are verified by the BPA power flow program of China(version 4.0)compiled by China Electric Power Research Institute to calculate the power flow of regional power grid,"N-1" operation mode verification,voltage value of each node and other data,and the advantages and disadvantages of the method used and described in the summary and Prospect of this thesis,and further suggestions are put forward improvement methods.