RESEARCH ON MODELING OF SIMULATION ERRORS FOR AIRCRAFT BAY DOOR HYDRAULIC SYSTEM BASED ON DATA-DRIVEN

As hydraulic drive technology advances, there is an increasing demand for accurate modelling of hydraulic drive systems for design analysis and full life cycle management. In this paper, an AMESim simulation model is established for an aircraft bay door hydraulic drive system test bed. Since there is a large error between the output of the simulation model and the actual, an error compensation model is established for the AMESim output and the physical output. This paper proposes a combined GRU-Attention-ELM model for the establishment of the error compensation model, and compares it with other commonly used machine learning models, the fitting effect is better than other models, and the accuracy of the compensated simulation output is greatly improved.