Soft-sensing of Cable Collision Speed of Obstructed Object of Blocking Device Based on CNN-GRU
As an important guarantee for the safe stopping of the blocked object,the blocking device cannot obtain the speed of the rope collision of the blocked object in real time.To solve this problem,a new CNN-GRU soft sensor model is proposed.First-ly,according to the data characteristics of the blocking device,the sequence is expanded into a three-dimensional matrix.Then,the channel attention mechanism is combined with the residual module to mine the sequence features and reduce the data dimen-sion.Finally,the extracted features are re-converted into sequences and the collision speed of the obstructed object is deduced through the gating cycle unit.The experiment shows that this method has a good effect in the prediction of the collision speed of the obstructed object of the arresting device.
万方数据
维普
