论文部分内容阅读
针对现役装备技术状态评估多依赖于手工拆卸的现状,提出一种基于AdaBoost-SVM模式识别算法的在线技术状态评估方法。利用人工后坐在线检测设备对炮闩装置技术状态参数进行检测,在对检测数据进行相关性分析特征提取的基础上,引入支持向量机模式识别方法,建立炮闩装置技术状态评估模型。通过将评估模型与Ada-Boost算法相结合,每次迭代都根据测试精度对分类错误的样本点和各分量分类器的权重重新赋值,在下一次迭代中形成新的分量分类器以优化分类结果,最终将各分量分类器依其权重综合完成评估。实例分析结果验证了评估模型的正确性和有效性。
In view of the status quo of the status quo of the active equipment technology relying on the manual disassembly, a method of on-line status assessment based on the AdaBoost-SVM pattern recognition algorithm is proposed. Based on the correlation analysis of the detection data, this paper introduces the support vector machine pattern recognition method and establishes the technical state assessment model of the breech device. By combining the evaluation model with the Ada-Boost algorithm, each iteration reassigns the wrongly classified sample points and weight of each component classifier according to the test precision, and forms a new component classifier in the next iteration to optimize the classification result. Finally, the component classifiers are evaluated according to their weights. The result of example analysis verifies the correctness and validity of the evaluation model.