| Global electricity consumption is expected to increase to 225 percent by 2050 as fossil fuel consumption continues to rise.The traditional power generation method is centralized power generation based on fossil fuels.However,as the world energy crisis becomes more and more serious,the subsequent problems such as carbon emissions,social and political pressure and energy security limit the possibility of further expansion of traditional energy.Therefore,it is necessary to vigorously develop renewable energy to alleviate the energy crisis.Microgrid is a kind of distributed power grid,which is composed of various distributed renewable energy generators.Due to the power generation characteristics of renewable energy,the management mode of traditional power grid cannot adapt to the distributed structure of microgrid.In microgrid,factors such as different scenes,distribution locations,and power generation period are complex,leading to serious imbalance in power generation and consumption of different users,and low enthusiasm for user participation,which is not conducive to the implementation and use of microgrid.Therefore,the participation experience,economic benefits and environmental benefits of microgrid users are optimized for the problems mentioned above in microgrid scenarios.In addition,in the above two application scenarios,it is necessary to predict the photovoltaic output,so as to achieve the purpose of absorbing the photovoltaic energy of the microgrid.The specific work is as follows:(1)A Principal Component Analysis(PCA)method was proposed to reduce the dimension of pv power generation data,which not only reduced the model processing time,but also eliminated the data characteristic noise with low correlation.At the same time,the processed data is put into the Long and Short Time Memory Network(LSTM).Through the simulation experiment,the average absolute error of the simulation prediction is 0.787,which is 0.457 higher than that of the traditional LSTM method.The experiment shows that the pv output prediction model has good accuracy and stability.(2)Aiming at the problem of household energy management in photovoltaic power generation scenario,a multi-objective optimization model was proposed,which considered economic cost,environmental cost and user comfort.By constructing load model and setting economic and comfort objective function and constraint function,the optimal energy consumption strategy of smart home equipment was obtained.This strategy can reduce power consumption cost and carbon emission to the greatest extent while considering user comfort.Through two simulation cases,the effectiveness coefficient method is used as a comprehensive evaluation index to verify the correctness and validity of the developed results.(3)In view of the low degree of active participation of users in P2 P market,a psychological motivation model is proposed to formulate electricity price pricing strategy and user priority strategy for different scenarios to improve the participation of consumers.Given the volatility of energy and the constraints of local communities,there is an energy imbalance in the trading market when energy is traded among fragmented consumers.This paper proposes an optimal alliance matching mechanism for microgrid in smart distribution system,and designs a hierarchical alliance forming mechanism to form a microgrid user alliance.By matching the optimal alliance,the transaction matching degree between users is improved and the energy transfer within the alliance is maximized to improve the energy use efficiency.Through simulation and analysis,when optimize the user selects the optimal model,can effectively improve the trade matching degree between users,by users and their constraints to model,the trading scheme can improve the user economic and environmental benefits,at the same time,the income can be a positive feedback,the psychological motivation model to form a virtuous cycle,To achieve the purpose of constantly encouraging users to participate. |
PDF Full Text Request