This study aims to investigate the sealing performance of reciprocating seals under the effect of rubber abrasion using ABAQUS simulation software, and to propose a prediction framework based on a hybrid algorithm (GA-PSO-BPNN) to predict the leakage of reciprocating seals of downhole gauging instrumentation under different working condition parameters.
The authors combined the UMESHMOTION user program with the improved Archard wear model to investigate reciprocating seal performance. GA and a PSO were proposed as ways to enhance the BPNN’s predictive model.
The results show that the impact of fluid pressure fluctuations on the wear of the seal lip is more pronounced during the rapid wear phase compared to the steady wear phase. Similarly, variations in compression rate have a greater impact on seal lip wear at different stages of wear. The GA-PSO-BPNN prediction model outperforms the single-prediction model in terms of prediction accuracy.
The authors investigated sealing performance through simulation software and propose a GA-PSO-BPNN-based fault diagnosis method for rotating machinery. To verify the accuracy of the prediction model, a reciprocating sealing test platform for gauge work cylinders is constructed.
The peer review history for this article is available at: https://publons.com/publon/10.1108/ILT-08-2024-0293/
