Research On Optimal Configuration Planning Of Multi-energy Microgird Based On Source-load-temperature Scenario Deep Joint Generation

To deal with environmental disruption and energy exhaustion,the multi-energy complementary model of energy conservation,environmental protection,centralization,multiple integration,distributed collaboration,and efficient configuration has gradually become the development direction of micro-energy grids.In order to improve the multi-energy complementary capability in a multi-energy microgrid system and the reliability of energy supply in extreme scenarios,a multi-energy microgrid system optimization configuration method based on source-load-temperature depth scenario joint generation is proposed.First,the working principle and operating characteristics of the equipment in the multi-energy microgrid system are analyzed considering the diversity of equipment in a multi-energy microgrid system.Mathematical models of photovoltaic,combined heat and power,electricity storage system,heat storage system,electric cooler,electric heat pump,absorption cooler,and gas turbine are established.Secondly,in order to exploit the coupling characteristics of multiple energy sources under complex meteorological conditions,improve the system economy and energy supply reliability,cluster analysis is carried out on historical meteorological data.Typical meteorological type clusters are obtained.At the same time,the extreme temperature types are divided by the temperature that has the highest correlation with the energy load.Then,based on the historical data of radiation,electric load,and temperature,for each meteorological module in typical scenarios and extreme scenarios,the joint generation of radiation-electric load-temperature depth scenarios based on the modular denoising variational autoencoder was carried out.This can accurately characterize the coupling characteristics of various energy sources in a multi-energy system and can enhance the complementary potential of multiple energy sources.Temperature is used to determine the cooling and heating load.Source-load-temperature scenarios under different weather scenarios are obtained.Extreme scenarios with fewer data samples are expanded.Among them,modular modeling can make the generated model more targeted.Furthermore,the denoising criterion can enhance the robustness and generalization of the model.The new method can generate massive potential typical and extreme scenarios that are similar to and different from the probability distribution and correlation of historical data.Thus,it reflects the uncertainty of energy consumption in different meteorological scenarios and covers potentially high energy consumption scenarios under limited extreme temperature data.Finally,cluster analysis is used for scenario reduction.Typical days for different typical scenarios and extreme temperature scenarios are determined.The scenario weight is determined by the historical occurrence probability of various scenarios.A multi-energy microgrid system optimal configuration model with the lowest total annual cost of the multi-energy microgrid system as the objective function is established.Experimental analysis shows that the generated source-load-temperature joint scenarios can reflect the differences under different meteorological conditions,and can deeply tap the complementary potential between energy sources to improve the economics of optimal configuration and reliability of energy supply.