This is the authors' abstract. We don't add key points for this paper.
The roller-type residual film collector is the main machine type for collecting residual films.The scarifying tooth harrow bears the largest resistance among all the components of the entire machine, and it directly affects the quality and efficiency of the residual film collector.In this study, the arithmetic optimization algorithm-variational mode decomposition -bidirectional long short-term memory (AOA-VMD-BiLSTM) model was proposed to improve the working condition identification accuracy of the scarifying tooth harrow.First, the locations of three scarifying tooth harrow monitoring points were analyzed, the arithmetic optimization algorithm was introduced to optimize the variational mode decomposition (VMD) parameters, and the optimal parameter combination of the three monitoring points was acquired.Second, the reconstructed signal for the three monitoring points on the scarifying tooth harrow after VMD decomposition was subjected to two-way updating and recursion with bidirectional long short-term memory (BiLSTM) as the core of the model, and iterative training of the scarifying tooth harrow under four working conditions was performed.Last, the new sample set and the original dataset were imported into two deep learning models, namely, BiLSTM and long short-term memory (LSTM), followed by a comparative validation of the training results of VMD-BiLSTM, VMD-LSTM, BiLSTM, and LSTM models.Results demonstrate that among the models, VMD-BiLSTM exerts the best processing effect on complex signals and has a higher convergence rate, better convergence stability, and higher accuracy.In the VMD-BiLSTM diagnostic model, each performance index can exceed 97.3%.This model performs better than VMD-LSTM, BiLSTM, and LSTM models and realizes intelligent classification under four signal conditions of the scarifying tooth harrow: no load, normal operation, mild overload, and serious overload.The proposed method provides great significances for improving the intelligence and technical monitoring levels of residual film collectors and provides a reference for follow-up R&D of intelligent residual film collectors.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Xue et al. (2024) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: