The accurate prediction of mine water inflow plays an important role in ensuring safe production and protecting the groundwater environment in coal mines.In order to improve the accuracy of time series prediction of mine water inflow,a combined prediction model of mine water inflow based on variational modal decomposition(VMD)and the optimization of bidirectional long short-term memory network(BiLSTM)by Bayesian algorithm was constructed.Firstly,VMD was used to decompose the time series data of mine water inflow into multiple subseries,and then the subseries obtained by decomposition were input into the BiLSTM model.At the same time,Bayesian algorithm was introduced to optimize the hyperparameters of each model.Finally,the prediction results of each subseries were superimposed and summed to obtain the final predicted value,and compared with the prediction results of other models.The results show that the model has obvious advantages in single-step prediction,and its performance in multi-step prediction is also quite good.The prediction accuracy can meet the production demand,which verifies the validity and applicability of the model in the time series prediction of mine water inflow.
关键词
变分模态分解/双向长短期记忆网络/贝叶斯算法/矿井涌水量预测/时间序列
Key words
Variational modal decomposition/Bidirectional long short-term memory network/Bayesian algorithm/Prediction of mine water inflow/Time series
维普
万方数据