车身高度调节中的滑模控制神经网络消抖技术
打开文本图片集
中图分类号：TB9;U461.4 文献标志码：A 文章编号：1674-5124（2025）06-0150-10
The SMC neural network anti-vibration technology during vehicle height adjustment
ZHU Maoyuan¹， ZHU Honglin¹，LIU Xiaoya²，DING Weiping' （1. School ofMechanical Engineering， Southwest Jiaotong University， Chengdu 61oo31， China; 2. Sichuan Ningjiang Shanchuan Machinery Co.， Ltd.， Chengdu 61o1oo， China）
Abstract：The oscillatory phenomenon inherent in sliding mode control （SMC） has the potential to significantly degrade the precision of damping dynamics matching during vehicle height adjustment， consequently impacting the smoothnessof theride experience. Initially，itis posited that effective mitigationof these oscilations hinges upon the characteristics of the switching term within SMC，demanding atributes of continuity and adaptability.While the utilization of a continuous saturation function as the switching term， coupled with carefully designed boundary layer configurations， shows promise in atenuating oscilations， the inherent linearity of the control mechanism within the boundary layer poses limitations on adaptability and robustness.As a result， a novel methodology is proposed，entailing the integration of Radial Basis Function （RBF） neural networks into the SMC framework， denoted as SMC-RBF， where RBF serves as the pivotal switching term. Leveraging the nonlinear mapping capabilities of RBF ensures the continuity of the switching term， while harmessng its capacity for online weight updates via closed-loop feedback enables adaptation to time-varying systems. Consequently， this approach yields a seamlessly adjustable damping force curve， thus augmenting the precision of damping dynamics matching.Finally，from the vantage points of algorithmic oscillation suppression efficacy，as well as the ramifications of delays and dynamic factors on practical applications， comprehensive simulations encompassing 1/4 vehicle models， corresponding testing platforms， and full vehicle simulations are conducted.The results unequivocally demonstrate that， in comparison to conventional SMC and SMC-Fuzzy strategies， SMC-RBF affords superior oscillation suppression capabilities and，under conditions ensuring the preservation of handling stability，significantly enhances ride smoothness. Keywords： height active adjustment; damping dynamic matching; SMC; vibration; neural network
0 引言
车身高度与阻尼集成可调空气悬架系统是一种有效协调汽车平顺性与操稳性的技术解决方案，备受学术界与工业界的关注。（剩余10631字）