Research On The Key Technology Of Large Aspheric Ultra-precision Optical Grinding Machine Design

The rapid development of space optics make a growing requirements on large aperture mirrors.Limited by the transmission and installation conditions,SiC has become the primary material of large opital mirrors because of its light weight and strong stability.Conventional lapping techniques for SiC forming have the shortcomings of low efficiency and long working hours,which is difficult to meet the challenges of current market.Thus,replacing conventional lapping process with effective ultra-precision grinding to acquire ultra-precision polished surfaces directly has become the hotspot.During the whole process chain,ultra-precision grinding,the processing effect of which determines the machining efficiency and feasibility of subsequent polishing,has great influence on the quality of its surface and sub-surface.Thus,it becomes one of the key technologies of developing ultra-precision grinding machine for large aperture optical mirror for large aperture reflectors manufacturing.The coupling of ultra-high stiffness,large damping and long-lasting stability makes the design of large ultra-precision optical grinding machine one of the most complicated highest risk engineering problems.With the fast development of computer aided design(CAD)and finite element method(FEM)techologies,computers are playing more important roles in product design because of its strong abilities of massive data processing,calculating and simulating.However,Limited by current artificial intelligence levels,the inherent defect of computer in evaluating and decision-making makes human knowledge and experience still irreplaceable,especially when handling complicated tasks with qualitative analysis,incomplete conditioning or mutual contraction.Aiming at the human-machine integrated intelligent design of large ultra-precision optical grinding machine,the basic principles and methods of intelligent design and its application in grinding machine are discussed in this paper.A few-axis configuration and a novel force/position control strategy are intensively studied to solve the prominent multifactor coupling and behavior contradiction existed during the design process.Furthermore,the mechanics,electronics and detection of the grinding machine and the design parameters and technical requirements have been detailed designed and optimized.Finally,the large ultra-precision optical grinding machine with 1.2m diameter machining capability has been developed successfully.The main works and achievements of this paper are as follows:Firstly,a systematic framework for human-machine integrated intelligent design and a computer-centered intelligent design prototype are proposed.The ontology nature of intelligent design is explained from the view of knowledge operation.And the design cognitive inference based information flow process is analyzed asiterative optimizations with three levels.To deal with the influence of semantic heterogeneity in knowledge representation of intelligent design,we propose an ontology based knowledge representation and management system,and then develop a knowledge model which includes state ontology,resource ontology and process ontology in intelligent design.Then the integration of knowledge in various domains is realized through ontology mapping and fusion,which provide foundation for the reusing and sharing of knowledge and experience.Secondly,a systematic intelligent conceptual design framework and related design process composing of concept generation,evaluation and upgrading are propsed to deal with the problems of multa-factor coupling and behavior contraction in large ultraprecision optical grinding machine design.The concept trees of function requirement,design parameter and expected behavior for ultra-precision optical grinding machine design are proposed.And the five-axis configuration is generate and optimize using quotient kinematics.A virtual axis principle based few-axis configuration and a novel feedrate control based force and position hybrid control strategy are proposed according to the analysis of the contradiction among working range and static stiffness of large ultra-precision optical grinding machine,as well as the high stiffness of position servo and compliance of force control.Thirdly,we develop the 3D virtual prototype for the large ultra-precision optical grinding machine.Mechanical structure,electric control and measuring system of the machine are embodied and optimized completely.After that,the static and dynamic stiffness,anti-vibration performance and thermal deformation of the grinding machine and key components as well as the parameter of machine control system are analyzed and optimized to choose the best design parameter for the machine tool.Fourthly,during the detailed design of the key componets of large ultra-precision optical grinding machine,an ontology-based conceptual design method for precision spindle is proposed,and an ontology-based knowledge modelling,fuzzy inference rules and example searching based intelligent selection for spindle bearing and actuating methods are developed.After that,the geometric and finite element model of spindle is constructed to analyze and optimize its statics,dynamics and thermal features.Furthermore,a novel large self-compensation precision hydrostatic rotary with surface restriction is designed.The mechanism of the surface restriction and self-compensation principle is introduced,and then its structure features and performance parameter are analyzed.In conclusion,the successful development of large ultra-precision optical grinding machine not only proves the effectiveness of the human-machine integrated intelligent design in the theoretical level,but also lays firm foundations for increasing the manufacturing ability of large optical components in China.It also contributes greatly to the development of the key fields such as the space optics and military remote sensing in China.