基于太阳能和风能的无线充电站建模和电动汽车充电需求预测

Modeling of Wireless Charging Stations Based on Solar and Wind Power and Prediction of Electric Vehicle Charging Demand

电动汽车(Electric vehicle, EV)和新能源发电相结合,有助于推动可持续能源转型,减缓气候变化,实现环境友好的能源利用和交通方式.为了在未来几年彻底取代传统燃油汽车,必须解决EV的充电和续航问题.为此,提出了基于太阳能和风能的EV无线充电仿真模型,利用太阳能和风能对无线充电站的电池进行充电,并通过2个互耦线圈之间的感应功率对EV进行无线充电.此外,提出了基于改进深度学习算法的EV充电需求预测模型.建立基于编码器-解码器结构的注意力双向门限递归单元(Attention-based Gated Reccurent Unit, Att-BiGRU)框架,对充电站的EV充电需求进行预测.结果表明,新能源无线充电站能够提高EV充电的安全性和舒适性,且所提方法能够准确预测不同时间间隔的EV充电需求.与传统模型相比,所提改进模型在充电需求预测任务中具有更快的收敛速度和更低的误差率,能够有效解决EV充电需求的随机性和波动性.

The integration of electric vehicles(EVs)with new energy generation contributes to driving the transition to sustainable energy,mitigating climate change,and achieving environmentally friendly energy use and transportation.In order to completely replace traditional fuel vehicles in the coming years,the charging and range issues of EVs must be addressed.A simulation model for wireless charging of EVs based on solar and wind energy is proposed.Solar and wind energy is used to charge the batteries of wireless charging stations and EVs are wirelessly charged through the induced power between two coupled coils.Additionally,a predictive model for EV charging demand based on a hybrid deep learning algorithm is proposed.An attention-based gated recurrent unit(Att-BiGRU)framework is established based on the encoder-decoder structure to predict the charging demand of EVs at charging stations.The results indicate that new energy-based wireless charging stations can enhance the safety and comfort of EV charging,and the proposed method accurately predicts EV charging demand at different time intervals.In comparison to traditional models,the proposed improved model demonstrates faster convergence and lower error rates in charging demand prediction tasks,effectively addressing the stochasticity and volatility of EV charging demands.