Predicting droplet vibration in oil subjected to chaotic pulse group electric field excitation using LSTM

¹ School of Mechanical Engineering, Chongqing Technology and Business University, Chongqing, 400072, China
² Engineering Research Centre for Waste Oil Recovery Technology and Equipment of Ministry of Education, Chongqing Technology and Business University, Chongqing, 400072, China
³ School of Environment and Resources, Chongqing Technology and Business University, Chongqing, 400072, China

^* Corresponding author: This email address is being protected from spambots. You need JavaScript enabled to view it.

Abstract

Oil-water separation is the key step in the process of converting waste oil into a resource.The chaotic pulse group (CPG) electric field is now regarded as an emerging method for oil-water separation.However, droplet vibration under CPG electric field excitation involves complex parameters that are challenging to analyze. Traditional numerical methods are hampered by high computational demands, lengthy processing times, and limited output, restricting further exploration of vibration characteristics. To address these limitations, this paper proposes a prediction model for droplet vibration under CPG electric field excitation using a long short-term memory (LSTM) neural network. The research results show that the LSTM model can effectively extract and learn the long-term dependencies in the droplet vibration sequence, capture the relationship between droplet vibration and chaotic pulse group electric field signals as well as time, thereby achieving accurate predictions and saving computing resources. Moreover, the LSTM model outperforms BP, GRU, and CNN models on the test set in terms of prediction accuracy.