首页>
外国专利>
Battery Output Voltage Response and State-of-Charge Forecasting Optimization Method using Hybrid VARMA and LSTM
Battery Output Voltage Response and State-of-Charge Forecasting Optimization Method using Hybrid VARMA and LSTM
展开▼
机译:电池输出电压响应和使用混合Varma和LSTM的充电状态预测优化方法
展开▼
页面导航
摘要
著录项
相似文献
摘要
The method for optimizing the prediction of the output voltage response and the state of charge of a battery using hybrid VARMA and LSTM according to an embodiment of the present invention uses actual data including the battery voltage response and the battery output current as input variables and uses a model generation program. A first step of obtaining an initial vector autoregressive moving average (VARMA) model; A second step of predicting a battery voltage response and a battery output current using the initial vector automatic regression moving average model but only a pure vector automatic regression (VAR(p)) model; A third obtaining residuals by subtracting the battery voltage response included in the actual data by the predicted battery voltage response predicted in the second step, and subtracting the battery output current by the predicted battery output current predicting in the second step. step; A fourth step of re-estimating a battery voltage response and a battery output current using the residual of the third step and a VARMA(p, q) model; And a fifth step of performing optimization by applying the predicted battery voltage response value, the predicted battery output current value, the real data, and the least squares estimation method to obtain the parameters of the optimal VAR and VMA models. .
展开▼