Research On Key Technologies For Military Automatic Test System And Its Uncertainty Evaluation

Military Automatic Test System (ATS) has developed to necessary guarantee for complex systems and weapon equipments keeping going. It has already become one of leading technologies studied by many countries. While corresponding Automatic Calibration System (ACS) has developed to provide with metrology support, in order to guarantee the accuracy and consistency of the ATS measurement result. Therefore, the research on ATS and its uncertainty evaluation has significant theoretic and practical value, which includes setting up military ATS or ACS, evaluating measure quality and serving for test & calibration task.This paper, based on the engineering practice of various missiles' ATSs and integrated with signal-oriented testing, mainly does the research on the key technologies of ATS test and calibration.Firstly, starting from stimulation, response and test of UUT information model, and the testing realization, simulating test technology based on signal model is put forward. Taking signal as basic unit, it can provide ATS with UUT working conditions and equivalent signals, forming self-stimulating, self-responding and self-testing. The methods such as double-buffer mechanism, analytical simulation and hardware-integrated simulation are adopted to realize it and improve the testing for UUT. It can reduce the requirement for real hardware, increase the development progress and validate the capability of testing and fault location.Secondly, ATS testing model based on signal routing is established to optimally configure the test information, which includes the resources capability, the signal routings and UUT test requirements. Especially, the design of switching subsystem based on signal routing is presented. It describes the switching subsystem to route signals by start-point sets, end-point sets, routing attributes, matrix relation and paths sets. All test resources and UUT signals can be assigned uniformly. It can achieve signals matching, paths reachability and resources optimization, and improve the flexibility, expansibility and integrity for system foundation.Thirdly, system integration of various ATS for tactical missiles is realized. In the testing process for TV goniasmometer, a target equivalent method is put forward, which can supply the appropriate target and background. The light source system of Cassegrain is improved by substituting inverse-folding for reflecting. It is the first time to carry out the testing for TV goniasmometer in the field condition via virtual monitor, arriving at miniaturization, rapidness and portability.Fourthly, based on Monte Carlo (MC) simulation, the uncertainty propagation and evaluation in the virtual instrument (VI) measurement chain are achieved. The analytical method and MC simulation are adopted to evaluate the uncertainty of measurement chain, such as signals sources, signals conditioning module, data acquisition module, data processing module and so on. The result shows the good consistency and the validation of MC simulation. It is possible for ATS to automatically supply the measurement result with an associated uncertainty.Moreover, according to reverse analysis idea and MC simulation for distribution propagation, sampling reverse analysis for uncertainty distribution is presented. The method realizes distribution simulation from output to input and converse process of distribution propagation, integrated with the estimation of probability density function (PDF) with different samples scale and the uncertainty evaluation. It can indirectly accomplish the uncertainty evaluation from testing signals to source signals, and primarily resolve the on-station calibration puzzle on signals sources in the metrology process of ATS.Finally, based on the measurement reliability index, the determination on calibration intervals of measuring instruments or test system is studied. Taking the impact of decision risk in the calibration process and different parameters into account, the simple response method (SRM) and Weibull reliability method are improved. The methods are corrected to calculate the optimal calibration intervals and the result show more reasonability and creditability.In fact, the main work of this paper has passed the system validation and technical evaluation with product mode, and put into use. With the appraisement of user, they play an important role in the examining maintenance and metrology support.