为了有效利用多种检测数据来综合评价轨道的状态,本文应用BP神经网络技术建立了轨道状态评定方法,并采用MATLAB软件编制了具有自学习功能的评价软件。BP神经网络应用于轨道状态的评价中,其学习样本、规模及代表起关键作用,通过大量的样本训练,对“未知”样本神经网络模型的评价具有较高的准确性。理论分析与算例的结果表明,该评价方法是可行的、有效的,为解决轨道状态评定提供了一条新的途径。 In order to effectively use a variety of test data to evaluate the orbit status comprehensively, a method of assessing orbit status is established by using BP neural network technology and a self-learning evaluation software is compiled by using MATLAB software. The application of BP neural network in orbit state evaluation plays a key role in learning sample, scale and representation. Through a large number of sample training, BP neural network has a high accuracy in evaluating the “unknown” sample neural network model. The results of theoretical analysis and numerical examples show that this method is feasible and effective and provides a new way to solve the orbital condition assessment.