组合导航系统性能评估方法的研究

发布时间：2018-12-31 18:09

【摘要】：随着微电子技术、导航技术、控制理论的发展,使得体积小、成本低、功耗低、功能多、精度高、可靠性高的组合导航系统成为新的发展趋势,并且广泛应用在军事、航空、航天、航海、交通等领域。组合导航系统主要任务是为运载体提供准确可靠的位置、速度、姿态等信息。随着自动化程度越来越高系统的结构越来越复杂,系统内设备之间的联系越来越紧密,其性能直接关系到运载体及载体全体人员的安全。因此,通过评估系统的性能,为组合导航系统的优化设计奠定基础,对确保系统高效可靠的工作具有重要意义。首先,本文以组合导航系统为例,在阅读大量相关文献的基础上,介绍了课题研究的背景和意义,总结了导航系统及其评估理论的发展现状,阐述了系统评价过程中需要注意的问题及几种常用的系统评价方法,探讨了组合导航系统的工作原理,分析了系统性能指标,建立了组合导航系统性能评估指标体系,给出了指标权重的确定方法,通过采用层次分析法确定了指标权重。其次,针对层次分析法得出的指标权重客观性较差,模糊综合评判法确定的评判等级可区分度不强等问题,运用了基于熵的模糊综合评判法,建立了系统性能评判模型,该方法运用熵权法修正了指标权重,选取加权平均算子进行模糊运算,根据指标权重使得每个指标对系统的综合性能都有所体现,采用区间量化方式增强了处于同一评判等级系统的可区分性,通过仿真实验证明了该方法可以准确、客观的评估出系统的性能,并与模糊层次分析法进行了比较分析。然后,针对组合导航系统性能评价的非线性和复杂性问题,BP神经网络对非线性系统有较强适应能力,运用了基于BP神经网络的综合评判法,建立了系统性能评价模型,确定了组合导航系统的网络模型结构,无量纲化处理了样本数据,通过仿真实验表明该方法简单,易于操作,克服了评估过程中容易出现的模糊性和不确定性,并与基于熵的模糊综合评判法的评估结果进行了比较,验证了该模型是合理、可信的。最后,针对BP神经网络评估方法需要大样本、计算量大等局限性,运用了基于灰色理论的综合评价方法,分别建立了灰色关联度法、灰白化权和模糊数学法、灰色聚类法的评估模型,通过仿真实验表明了所提出的三种方法是正确合理的,都可以准确的评判出组合导航系统的优劣,而灰色聚类综合评估法解决了灰色关联度综合评估法不能确定系统评判等级的问题,弥补了灰白化权和模糊数学综合评估法客观性差及确定的评判等级可区分度不强等缺点,提高了组合导航系统性能评估的客观性和有效性。
[Abstract]:With the development of microelectronic technology, navigation technology and control theory, the integrated navigation system, which is small in size, low in cost, low in power consumption, high in function, high in precision and high in reliability, has become a new development trend, and has been widely used in military and aviation. Space, navigation, transportation, etc. The main task of integrated navigation system is to provide accurate and reliable position, speed and attitude information for the vehicle. With the increasing degree of automation, the structure of the system is becoming more and more complex, and the relationship between the equipment in the system is becoming more and more close. The performance of the system is directly related to the safety of the carrier and all personnel. Therefore, it is of great significance to evaluate the performance of the system and lay a foundation for the optimal design of the integrated navigation system. Firstly, taking integrated navigation system as an example, this paper introduces the background and significance of the research, summarizes the development status of navigation system and its evaluation theory, based on reading a large number of related documents. This paper expounds the problems needing attention in the process of system evaluation and several common system evaluation methods, discusses the working principle of the integrated navigation system, analyzes the system performance index, and establishes the performance evaluation index system of the integrated navigation system. The method of determining the index weight is given, and the index weight is determined by the analytic hierarchy process (AHP). Secondly, aiming at the problems such as poor objectivity of index weight obtained by analytic hierarchy process (AHP), and weak degree of judgment grade determined by fuzzy comprehensive evaluation method, the system performance evaluation model is established by using fuzzy comprehensive evaluation method based on entropy. In this method, the index weight is modified by entropy weight method, and the weighted average operator is selected for fuzzy operation. According to the index weight, the comprehensive performance of the system is reflected by each index. Interval quantization is used to enhance the discriminability of the system in the same judgment level. The simulation results show that the proposed method can evaluate the performance of the system accurately and objectively, and it is compared with the fuzzy analytic hierarchy process (FAHP). Then, aiming at the nonlinearity and complexity of the performance evaluation of integrated navigation system, BP neural network has strong adaptability to nonlinear system. The comprehensive evaluation method based on BP neural network is used to establish the system performance evaluation model. The network model structure of the integrated navigation system is determined and the dimensionless sample data is processed. The simulation results show that the method is simple and easy to operate and overcomes the fuzziness and uncertainty in the evaluation process. Compared with the fuzzy comprehensive evaluation method based on entropy, it is proved that the model is reasonable and reliable. Finally, aiming at the limitation of BP neural network evaluation method, such as large sample and large amount of calculation, the comprehensive evaluation method based on grey theory is used, and the grey correlation degree method, gray whitening weight method and fuzzy mathematics method are established, respectively. The evaluation model of grey clustering method is proved to be correct and reasonable by simulation experiments, and the advantages and disadvantages of integrated navigation system can be judged accurately. The grey cluster comprehensive evaluation method solves the problem that the grey relational degree comprehensive evaluation method can not determine the system evaluation grade, and makes up for the shortcomings of the grey and white weight and fuzzy mathematics comprehensive evaluation method, such as the objectivity of the evaluation method and the degree of discrimination can not be strong, and so on. The objectivity and effectiveness of integrated navigation system performance evaluation are improved.
【学位授予单位】：哈尔滨工程大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN967.2

【参考文献】