基于 ATML 的特种机任务系统地面测试程序动态调度优化方法

Multi-Objective Optimization Method for Field Test Dynamic Scheduling of a Aircraft Mission System Based on Automatic Test Markup Language

摘要:

针对现有特种机任务系统地面测试程序按固定顺序执行、无法利用机载内置测试诊断先验信息的工程瓶颈,提出了一种基于自动测试标记语言的现场测试程序动态调度多目标优化方法。该方法以故障隔离、器材保障和快速出动三类现场维修场景的实际需求为牵引,建立了关联期望诊断步数、加权测试时间、早停隔离置信度的三目标优化模型,利用基于ATML测试程序平台已有的D矩阵故障模型和先验信息驱动改进了现有NSGA-II求解算法。案例结果表明,相比现有测试方案,将平均诊断步数减少41.5%,早停隔离置信度提升25.6%,动态重优化热启动响应时间不超过2 s,为提升该型装备持续作战能力提供重要决策参考。

Abstract:

The current ground test program for special aircraft mission systems exists two critical inherent drawbacks: test program sets are executed in a fixed order and cannot effectively utilize the prior diagnostic information from airborne built-in tests. To address these drawbacks, this paper proposes a multi-objective optimization method for test program dynamic scheduling based on Automatic Test Markup Language (ATML). Driven by the actual requirements of three typical field maintenance scenarios: fault isolation, equipment support and rapid deployment, this method establishes a three-objective optimization model associated with expected diagnostic steps, weighted test time and early-stop isolation confidence. It uses the existing D-matrix fault model and prior information on the ATML test program to drive and improve the existing Non-dominated Sorting Genetic Algorithm II (NSGA-II) solution algorithm. The case results show that compared with the existing test scheme, the average diagnostic steps are reduced by 41.5%, the early-stop isolation confidence is increased by 25.6%, and the dynamic re-optimization hot start response time is no more than 2 s. This method provides important decision-making reference for improving the continuous combat capability of this type of equipment.