【作者】 孙志伟；

【导师】 周驰； 王跃进；

【作者基本信息】 华南理工大学 ， 机械工程， 2011， 硕士

【摘要】 电火花成形加工已经成为机械行业中一种必不可少的重要加工方法,拥有独特的加工优势,目前已经广泛地应用于模具制造、电子、仪器、汽车等领域。在电火花成形加工中,选择合理的电参数对于放电加工效率、加工质量以及电极的损耗等加工工艺效果有很大的影响。传统的电参数的选择仍旧依靠操作员工的经验,效果很差。耐普罗机械(广州)有限公司,是一家以生产手机模具而驰名的模具生产厂商,常年为Nokia、Motorola、NEC、博世、高露洁等国际大公司生产模具。论文针对公司电火花成形加工效率低的情况,对电火花成形加工中参数的选择优化进行了研究。分析电火花成形加工原理,将电火花成形加工的放电过程分为放电通道的形成、放电能量的转换、电蚀物抛出、极间介质冷却四个阶段,并对正常火花放电、过渡放电、电弧放电作了比较,电弧放电是对电加工有害的放电:分析脉冲宽度、脉冲间隙、峰值电流、伺服电压与加工速度、表面粗糙度、电极损耗之间的关系,为进行电火花成形加工单因素试验提供理论基础。在日本牧野(MAKNO)公司所生产的EDNC65机床上进行单因素影响试验,分别对加工精度、加工速度、工具电极损耗这三方面设计单因素试验,得出脉冲宽度、脉冲间隙、峰值电流、伺服电压与加工质量之间的直观关系。电火花成形加工中电参数的优化,是多目标优化。分析传统的多目标优化的方法,总结传统方法的缺点。介绍基于遗传算法的多目标优化方法及其特点,遗传算法具有抗干扰性强、善于解决非线性问题等优点,适合解决电参数优化问题。带精英策略的快速非支配遗传算法,解决了非支配遗传算法计算复杂度高、优良个体被破坏等问题。建立基于带精英策略的快速非支配排序遗传算法的电火花成形加工电参数优化模型。论文针对东洋碳素石墨电极材料IS063,常用模具钢材8407的电火花成形加工,进行了基于带精英策略的非支配排序遗传算法的电火花成形加工电参数选择优化,并在此模型基础上进行了MATLAB编程,计算结果显示,电火花加工平均加工速度随着迭代次数逐渐增大,迭代25次后趋于稳定。采用优化前和优化后的电参数进行实例加工,结果显示运用此方法进行电参数的优化可以提高电火花成形加工的效率、加工质量、降低电极的损耗,在粗加工中加工速度提高了22.8%;在精加工中,粗糙度减低了?13%,同时加工速度降低了10.17%。更多还原

【Abstract】 Electrical discharge machining(EDM)，an indispensable important processing method，has a unique processing advantages，EDM is now widely used in mold manufacturing，electronics，instrumentation，automotive and other fields．In EDM，electrical parametersselection optimization of electrical discharge machining has greatly affected efficiency，processing quality and electrode wear and tear．T he traditional electrical parameters selectionsstill relied on the operators’experience and were very poor effetely．Nypro tool(Guangzhou)Co.，LTD was a wellknown mobile phone mold manufacture．Which producedmolds for large inte．inational companies suchu as Colgate，Nokia，Motorola，NEC，Boschand so on．According to low efficiency of EDM inthe nypro tool(Guangzhou)Co.，LTD，electrical parameters selection optimization of electrical discharge machining was researchedin the thesis．EDM principle was analyzed．E DM process was divided into fourstage，the formation ofthe discharge channel，energy conversion of discharge，electric corrosion were dished out，intermedium cooling，and the normal spark discharge．The transitional discharge，arcdischarge were compared，arc discharge is harmful to electric discharge machining．Therelationship between pulse width，pulse gap，peak current，servo voltage and processingspeed，s urface roughness，e lectrode wear and tear were analyzed．W hich provide a theoreticalfoundation for single factor experiment in EDM．Single factor experiments were designed and done for the accuracy of processing，processing speed，tool electrode wear and tear in EDNC65 which produced by the companyof Makino．T he relationship between pulse width，p ulse gap，p eak current，s ervo voltage andthe quality of processing were analyzed．Electrical parameters selection optimization of electrical discharge machining wasmultiobjective optimization．Some approaches of traditional multiobjective optimizationwere analyzed and summed up the shortcomings．Multi-objective optimization methods basedon genetic algorithm were introduced，which has characteristics of strong anti-interferenceand solve nonlinear problem．Multiobjective optimization methods based on geneticalgorithm were suitable to solve electrical parameters selection optimization of EDM． The method of nondominated genetic algorithm with elite strategy has solved the problemof high computational complexity and destroyed good individual．Set up model of electricalparameters selection optimization of EDM based on the nondominated genetic algorithmwith elite strategy．Thesis the model of electrical parameters selection optimization of EDMbased on nondominated sorting genetic algorithm with elite strategy was conducted for Toyocarbongraphite ISO63 electrode and 8407 steel and Matlab programmed based on thismodel．The calculation results were showed that the average processing speed is graduallyincreasing as the number of iterations．The average processing speed was stabilized afteriterative 25 times．Experimental based on original of electric parameters and the optimizedelectric parameters showed that the model of electrical parameters selection optimization ofEDM based on the nondominated genetic algorithm with elite strategy can improved theefficiency of EDM，processing quality and lower electrode wear，the process was increased22.8％in the rough processing；the roughness was reduced 13％，while the process wasreduced 10.17％in the finish processing．更多还原