Research On Multi-stage Optimal Allocation Of Hybrid Energy Storage Capacity In Wind Storage System

The proportion of wind power generation in China is increasing year by year,but due to the intermittent and fluctuating nature of wind power output power,large-scale wind power grid connection will cause a large impact on the power grid,thus affecting the safety and reliability of the power system.By achieving high accuracy wind power prediction,it is convenient for power-related departments to make scheduling plans,optimize storage capacity allocation,and improve the utilization rate of wind power.Thus,this paper proposes a short-term wind power prediction method for wind storage systems and realizes multi-level optimal capacity allocation for hybrid energy storage systems,as follows:The wind power prediction method is studied for the problems of volatility,randomness and low prediction accuracy of wind power.First,Bidirectional Long-Short Term Memory(Bi LSTM)neural network is selected as the base model for short-term wind power prediction.In order to address the problems of information loss and gradient disappearance caused by the long training time of Bi LSTM model,Attention Mechanism(AM)is introduced to highlight the important influencing factors of prediction and increase the weight of prediction feature information;secondly,to address the problems of hyperparameter selection and optimization search difficulties of prediction model,Whale Secondly,the Whale Optimization Algorithm(WOA)is used to optimize the hyperparameters of the prediction model,and then the WOA-AM-Bi LSTM prediction model is established.After example analysis,the WOA-AM-Bi LSTM prediction model proposed in this paper has an explanation capability of 96.4% and an average absolute percentage error of 4.93%,thus verifying the effectiveness of the prediction model for shortterm prediction of wind power.Aiming at the problem that it is difficult to accurately allocate the power of hybrid energy storage system,this paper proposes a method for multi-level allocation of hybrid energy storage power.Firstly,the reference power of the hybrid energy storage system is accurately allocated using the ensemble empirical modal decomposition and the variational modal decomposition;secondly,this paper takes the minimum whole-life cost of the hybrid energy storage system as the optimization objective,considers the charging and discharging efficiency of the energy storage equipment,the rated power and other constraints,and uses the carnivorous plant algorithm to find the optimal objective function,and finally obtains the optimal hybrid energy storage capacity optimization allocation scheme.