Research on Traction Performance Prediction of Wheel-Drive Tractor Based on the BWO Algorithm

Abstract: To address the limitations of existing traction performance prediction models for wheeled tractors, particularly their low accuracy and insufficient integration of soil mechanics, we propose an advanced, comprehensive set of models. These models span the entire process of system modeling, prediction optimization, and empirical validation. By incorporating multiple physical domains, including soil mechanics, tire dynamics, and the tractor's transmission system, we derive three core models, the tractor dynamics model, the wheel-soil interaction model, and the tractive force model—forming a robust prediction framework applicable to wheel-driven tractors. To enhance prediction precision, we introduce a novel optimization approach based on the Beluga Whale Optimization (BWO) algorithm, incorporating a multi-dimensional iterative strategy to minimize the machine';s slip rate. The accuracy of the model is further validated through real-time online optimization. Additionally, offline validation is performed using a YTO tractor as the test prototype. Experimental results demonstrate that, in comparison to existing models, the BWO-based traction performance prediction method achieves substantial improvements, with errors of 1.33% and 2.58% for maximum tractive force and tractive force at the characteristic slip rate, respectively, and 1.24% and 1.15% for rolling resistance at the unloaded tractive force and characteristic slip rate. In this paper, a tractive performance prediction model suitable for wheeled-driven tractors is established, which effectively improves the accuracy of the current tractive performance prediction model.