【摘要】 为了快速、准确预测不同工况下履带起重机桁架臂结构最大静力响应，提出了一种将BP神经网络和改进的COOT算法(ICOOT)相结合的ICOOT-BP神经网络预测模型。首先，采用Ansys参数化设计语言创建桁架臂在不同工况、杆件尺寸参数和载荷作用下最大静力响应的参数化模型，获取静力响应训练样本；其次，使用Tent混沌映射和自适应变异方法改进原始COOT算法，提高其优化能力，得到了改进的COOT算法(ICOOT)；最后，确定了BP神经网络模型的拓扑结构，利用ICOOT算法优化BP神经网络中的权值和阈值，建立桁架臂静力分析时输入参数与输出响应之间的代理模型ICOOT-BP。研究结果表明：某型履带起重机桁架臂在多种工况下，ICOOT-BP模型能够快速预测桁架臂的最大静力响应，预测结果与有限元分析结果具有高度一致性，位移和应力相对误差绝对值均小于4%，且在预测精度与训练效率方面均显著高于所对比的其他预测模型。所提ICOOT-BP模型极大地提高了履带起重机桁架臂的最大静力响应分析效率，可为桁架臂力学分析与结构优化设计提供准确的结构分析代理模型。更多还原

【Abstract】 In order to predict the maximum static response of the crawler crane’s lattice boom structure in various working cases quickly and accurately, an ICOOT-BP neural network prediction model based on the BP neural network and improved COOT(ICOOT) algorithm was proposed. Firstly, a parametric model of the lattice boom’s maximum static response was created by using the Ansys parametric design language(APDL) to obtain static response training samples under various working conditions, with different bar geometric parameters and loads. Secondly, the original COOT algorithm was enhanced by utilizing the Tent chaotic mapping and adaptive mutation method to improve its optimization ability, resulting in the improved COOT algorithm, namely ICOOT. Finally, the topological structure of the BP neural network model was determined, and the weights and thresholds in the BP neural network were optimized by using the ICOOT algorithm, thereby a surrogate model was established between input parameters and output responses for the static analysis of the lattice boom. The results show that the ICOOTBP model of a special crawler crane’s lattice boom can quickly predict the maximum static response of the lattice boom under various working conditions. The prediction results are highly consistent with the finite element analysis results, and the absolute values of the relative errors of stress and displacement are all less than 4%. Moreover, it is significantly superior to other prediction models compared in this paper in terms of prediction accuracy and training efficiency. The proposed ICOOT-BP model significantly enhances the efficiency of the maximum static response analysis for the crawler crane’s lattice boom, providing an accurate structural analysis surrogate model for mechanical analysis and structural optimization design of the lattice boom.更多还原