Dynamic Model Uncertain Parameter Identification for A Curved Chain Rotary Shell Magazine
To accurately simulate the nonlinear dynamics of a curved chain rotary shell magazine during the delivery of ammunition, a dynamic model containing uncertain parameters is established according to the system topology and control principle. An uncertain parameter identification model for the dynamic model is developed based on the system test data using the optimization design idea. A functional time series similarity is proposed as the identification criterion. The high-dimensional model representation based on the radial basis function and the radial basis function are used to construct the surrogate models from the uncertain parameters of the mechanical and control systems to the identification criterion, respectively. The sparrow search algorithm is embedded into an island model for multi-population structuring to form an island sparrow search algorithm for optimization. The uncertain parameters of the mechanical and control systems are identified successively by taking the test data of working condition 1 as the benchmark. The results show that the outputs of the identified dynamic model for the two operating conditions are similar to the test data, which verifies the accuracy of modelling and the validity of identification, and provides a reliable sample data source for action reliability analysis and fault diagnosis research.
curved chain rotary shell magazineparameter identificationfunctional time series similarityisland sparrow search algorithmhigh dimensional surrogate model
万方数据
维普
