Data Models based on Time-domain Multi-parameter Information Fusion and Machine Learning Algorithms

In the case of uncertain system parameters, it is not possible to directly perform state estimation. The state estimation of the system model requires known model parameters and noise variance. Therefore, before conducting state estimation, it is necessary to identify the unknown parameters in the model and input the identification results into the system model for state estimation. This article applied various new recognition technologies to fuse time-domain multi-parameter information, thereby reducing data dimensions and improving recognition accuracy. However, the current data obtained is often not accurate and comprehensive enough to fully reflect system characteristics. Therefore, it is particularly important to use information fusion methods for state estimation. This study introduced a mathematical model based on principal component analysis (PCA) and clustering algorithm, which combined PCA and genetic algorithm (GA) to improve the clustering algorithm and enhance the effectiveness of data preprocessing and analysis. The effectiveness of the introduced GA-PCA model in improving classification accuracy and computational efficiency was verified through simulation experiments on multiple datasets.