Intelligent prediction for digging load of hydraulic excavators based on RBF neural network. (January 2023)

• Record Type: Journal Article
• Authors: Huo, Dongyang; Chen, Jinshi; Zhang, Han; Shi, Yiran; Wang, Tongyang
• Is Part Of: Measurement. Volume 206(2023)
• Publication Date: 2023-01
• Subjects: Load prediction -- Intelligent excavator -- Machine learning -- Hardware-in-loop experiment -- Digging load characteristics; Weights and measures -- Periodicals; Measurement -- Periodicals; Measurement; Weights and measures; Periodicals; 530.8
• DOI: 10.1016/j.measurement.2022.112210
• Languages: English
• ISSNs: 0263-2241

Abstract

Highlights: A digging load prediction method for excavators based on RBF neural network is proposed. The method does not require prior knowledge of soil parameters and can be implemented in real-time environment. BPNN, coupled DEM-MBD simulation, and analytical model are used for comparative study. The proposed approach is validated by hardware-in-loop experiments. Abstract: Traditional modeling methods for digging load of excavators are often computationally expensive and require prior knowledge of soil parameters, which severely limits their engineering applications. According to the digging load characteristics in typical digging tasks, this paper presents an intelligent prediction method for digging load based on radial basis function (RBF) neural networks. The recursive least-squares (RLS) algorithm is used for weights updating. Back propagation neural network (BPNN), coupled discrete element method (DEM) and multi-body dynamics (MBD) simulation, and analytical model are applied for comparative studies. The simulation results illustrate that the RBF neural network model outperforms other comparative models in terms of prediction accuracy and computational cost. The hardware-in-loop (HIL) experiments are conducted to validate the proposed approach. Experimental results demonstrate that the error in the dynamic behavior of the excavator under the predicted digging load is less than 7%. This paper lays the foundation for digging load prediction in intelligent excavators.