The invention discloses an intelligent tire slip angle prediction method based on GSA-PCA-GPR (Global Sensitivity Analysis), and the method comprises the steps: calculating the global sensitivity index (GSI) of the output voltage of a PVDF (Polyvinylidene Fluoride) sensor at different installation positions based on a global sensitivity analysis (GSA) theory, and determining the installation position of the sensor; the method comprises the following steps of: acquiring a voltage signal of a sensor, extracting an input characteristic value, and performing dimensionality reduction and normalization processing on training data and test data by adopting a principal component analysis (PCA) theory and a linear normalization theory; the method comprises the following steps of: obtaining a side slip angle Gaussian process regression (GPR) model according to the characteristics of training data and test data, and obtaining an optimal solution of hyper-parameters; according to the combined prior definition, combined prior Gaussian distribution is established in combination with the voltage principal component characteristics; and based on posterior distribution, substituting the optimal hyper-parameter solution to obtain a slip angle predicted value and a variance thereof, and finally obtaining a tire slip angle predicted value. The method is low in cost, high in prediction precision, good in stability and good in generalization performance.
本发明公开了一种基于GSA‑PCA‑GPR的智能轮胎侧偏角预测方法,基于全局灵敏度分析(Global sensitivity analysis,GSA)理论计算不同安装位置处的PVDF传感器输出电压的全局灵敏度指标(Global sensitivity index,GSI),确定了传感器的安装位置;采集传感器电压信号,提取输入特征值,采用主成分分析(Principal component analysis,PCA)理论和线性归一化理论对训练数据和测试数据进行了降维和归一化处理;根据训练数据和测试数据的特征得到侧偏角高斯过程回归(Gaussian process regression,GPR)模型,获取超参数的最优解;根据联合先验定义,结合电压主成分特征,建立联合先验高斯分布;基于后验分布,将最优超参数解代入得到侧偏角预测值及其方差,最后得到轮胎侧偏角预测值。本发明成本低、预测精度高、稳定性好、泛化性能好。
Intelligent tire slip angle prediction method based on GSA-PCA-GPR
一种基于GSA-PCA-GPR的智能轮胎侧偏角预测方法
2023-12-19
Patent
Electronic Resource
Chinese
CNN-Transform-based intelligent tire slip angle estimation method
European Patent Office | 2025
|The tire slip angle estimation algorithm based on intelligent tire technology
SAGE Publications | 2024
|