Short-time Energy Demand Forecasting And Optimal Operating Strategy Of Regional Integrated Energy System

Since the 21 st century,the worldwide energy pressure has been reached to the top,it is urgent to establish a low-carbon and environmentally friendly energy supply-side structure and improve the overall energy utilization efficiency.The integrated energy system centrally integrates the individual traditional energy supply mode,and can realize the cascade utilization of energy through the organic coordination of multiple energy flows,which has been widely uesd.With the deeping of energy marketization reform,the energy supplier will no longer be the only decision maker,and the user as the main body of energy consumption will also play a key role in the energy transaction.At the same time,the continuous expansion and increasingly complex interconnection modes have also added new challenges to the coordinated optimization of the system scheduling.As the region integrated energy system taken by the main research body,studies about the energy demand prediction technology considering the influence of price incentive and the distributed cooperative optimal scheduling method are carried out.The work is as follows:(1)Considering the impact of price incentive on short-term energy demand forecasting in the process of supply-demand interaction between energy operators and users under the market mechanism,a short-term energy demand forecasting method considering price incentive is proposed.First of all,the structure of recurrent neural network(RNN)is briefly introduced.With the improvement of memory state unit,the long-short term memory network(LSTM)are introduced,and a short-term energy demand prediction model considering historical load data,energy price,meteorology and other factors is established.Secondly,according to the users’ energy preference and the comprehensive benefit of energy operators,the supply and demand interactive model is established with the guiding mechanism based on energy price and energy demand.Considering the price incentive in the process of supply and demand interaction,based on the forecasting value,the stable energy price of the interaction model is obtained,and is used to correct the parameters of the forecast model.Therefore,the circular nesting model with outer demand prediction model and inner supply and demand interaction is formed to obtain stable demand forecast value and energy price.Finally,the effectiveness of the model is proved through the historical load data.(2)In view of the conversion relationship of various heterogeneic energy flow and the information transmission privacy between multi-region integrated energy(MIES)systems interconnection,a distributed collaborative optimization scheduling method based on alternating direction multiplication method(ADMM)is proposed.Considering the characteristics of the geographical distribution,the multi-region system is separated by the link line power.Based on the energy coupling mechanism,the decoupling variables and consistency constraints are set to decompose the multi-region integrated energy system into multiple sub-regions.Then the multiple sub-regions,with the goal to optimize the comprehensive benefits,construct a multi-subject integrated energy co-optimization scheduling model based on the alternating direction multiplication method.Thereby,the optimization scheduling problem of multi-regional integrated energy system with information interconnection is decomposed into two parts: the information interaction layer and autonomous optimization layer;In the information interaction layer,the decoupling variables are exchanged and the Lagrange multipliers are updated to prepare for the next optimization.In the autonomous optimization layer,the scheduling problem is optimized again according to the updated multipliers.The two steps iterate alternately to obtain the stable and convergent optimal scheduling results.The model has been verified by carrying out case study with three coupling regional integrated energy system.