Research On Process Quality Control Technology Of Aerospace Pyrotechnics Based On Intelligent Detection

Aerospace pyrotechnics,embedded with explosives,epitomize single-use components distinguished by their inherent "uninspectable and untestable" performance traits.These characteristics demand the strengthening of process data monitoring to assure product quality,where performance indicators are evaluated through the "inspectable and testable" facets of process data.Due to the aerospace pyrotechnics industry’s specific and somewhat isolated nature,compounded by scant investment and modest engagement with advanced manufacturing technologies,traditional pyrotechnic production faces persistent challenges.These encompass limited automation,prevalent manual interventions,inadequate quality detection capabilities,ineffective record traceability,and complexities in assessing process quality.Addressing the unique demands of small-scale,diverse,and batch-wise discrete production in aerospace pyrotechnics,this study leverages intelligent detection technologies to scrutinize critical process data segments and undertake vital quality control research,thereby instituting data-driven process quality oversight.Initial analysis elucidates the distinctive attributes and quality control hurdles of aerospace pyrotechnics,delineating the aims and execution strategies for intelligent detection technologies.The investigation further explores intelligent detection approaches,including machine vision,X-ray detection digital radiography(DR)enhancements,and concurrent testing,to enable the integration of intelligent detection technologies in pivotal processes such as assembly,X-ray detection,and electrical performance evaluation.Embracing a data-driven methodology,this research harmonizes the complete data management cycle,applying intelligent detection alongside additional process data for the forecasting,appraisal,and refinement of production techniques.This facilitates a process data-oriented quality management approach,substantially elevating the quality and production efficacy of aerospace pyrotechnics.The study methodically engages in research across four principal domains.(1)A comprehensive machine vision solution has been proposed for quality inspection across the entire pyrotechnic assembly process.This solution specifically addresses the myriad factors influencing quality and the unique aspects of the assembly process in aerospace pyrotechnics,encompassing appearance,dimensions,charge loading,and assembly accuracy in various scenarios.The detection speed has changed from 20-50 seconds for manual detection to less than 2 seconds,and the intelligent detection efficiency has been improved by more than 10 times compared with manual inspection.Through the identification of diverse machine vision inspection methods,this approach facilitates the intelligent inspection of items pending examination,thereby mitigating the challenges of low automation in quality inspection and data management,along with the significant reliance on manual processes for handling process data and multimedia records.(2)An enhanced intelligent X-ray inspection system,leveraging Digital Radiography(DR)retrofit,has been developed.This system is designed to overcome the limitations inherent in traditional X-ray film photography,including high personnel requirements,low efficiency in interpretation,and traceability issues.The introduction of a key point information-based system for the automatic interpretation of X-ray images marks a significant step forward in digitizing X-ray inspection.Compared with manual interpretation,the detection efficiency has been increased by more than 60 times.This innovation substantially elevates the level of intelligence in pyrotechnic X-ray inspection processes,enabling more efficient and reliable analysis.(3)A multi-channel electrical performance detection system based on parallel testing has been developed.Addressing the challenges in pyrotechnics’ electrical performance testing,such as numerous steps,low testing efficiency,non-real-time data,reliance on manual judgment,and difficulty in data traceability,this system employs a multi-channel parallel intelligent testing scheme.The development of multi-channel electrical performance automatic detection equipment significantly enhances testing efficiency compared to manual testing,Compared with manual testing,the efficiency is increased by more than 4 times,achieving rapid and intelligent electrical performance inspection.(4)Concurrently,research focused on whole process quality control,anchored in data-driven principles,has been initiated.By integrating with extant ERP and MES systems to forge a central control platform,a thorough data collection encompassing the quality index system is facilitated.This research explores the requisite collection,transmission,and storage of pyrotechnics production process data,culminating in the establishment of a foundational database aimed at enhancing data gathering,analysis,and iterative processes to bolster process enhancements and inform production decisions.Notably,two principal areas of data analysis and exploration are highlighted: firstly,the application of grey relational analysis for the prediction of key process quality knowledge in small pyrotechnic samples,aimed at bolstering and refining process knowledge,thereby facilitating process quality predictions.Secondly,efforts are dedicated to refining pyrotechnic quality assessment indicators through the optimization of quality,compilation of algorithms,and the formulation of models,thereby enabling a nuanced analysis and assessment of the production process alongside the visualization of operational conditions.