本文選題：特征加工方法 + 典型主干工藝路線��； 參考：《浙江大學(xué)》2013年碩士論文

【摘要】：零件工藝設(shè)計直接關(guān)系到產(chǎn)品的生產(chǎn)效率與質(zhì)量。在激烈的市場競爭環(huán)境下,企業(yè)如何提高零件的工藝創(chuàng)新能力,已成為產(chǎn)品設(shè)計制造的一個重要問題。本文結(jié)合企業(yè)需求,研究了零件特征加工方案決策技術(shù)、典型主干工藝路線提取技術(shù)和零件工藝排序技術(shù),開發(fā)了鍛壓裝備工藝設(shè)計系統(tǒng),并且系統(tǒng)在鍛壓裝備產(chǎn)品的零件工藝設(shè)計中得到應(yīng)用,取得了良好的效果。 全文主要內(nèi)容包括： 第1章分析了傳統(tǒng)零件工藝設(shè)計方法存在的問題,介紹了零件工藝設(shè)計的關(guān)鍵技術(shù)及其研究現(xiàn)狀,闡述了本文的研究背景、內(nèi)容及組織構(gòu)架。 第2章提出了一種基于支持向量機的零件特征加工方案決策方法。介紹了支持向量機的二值分類算法和多值分類算法；建立了特征加工方法選擇的集成決策模型,并以內(nèi)圓特征為例,介紹了子支持向量機的模型結(jié)構(gòu)、輸人輸出參數(shù)的預(yù)處理、懲罰參數(shù)及核函數(shù)的選擇；利用LIBSVM軟件與Matlab平臺實現(xiàn)了零件特征參數(shù)的訓(xùn)練與預(yù)測。 第3章提出了應(yīng)用聚類分析技術(shù)提取零件典型主干工藝路線的方法。構(gòu)建了主干工藝路線的相似度度量因子,提出了對主干工藝路線進(jìn)行相似度計算的多級相似度綜合度量方法,在相似度計算基礎(chǔ)上,構(gòu)建了主干工藝路線設(shè)計結(jié)構(gòu)矩陣并對矩陣數(shù)據(jù)進(jìn)行降噪處理；為降低聚類劃分的難度和復(fù)雜性,運用迭代聚類算法實現(xiàn)了主干工藝路線設(shè)計結(jié)構(gòu)矩陣的聚類劃分,并從聚類簇中提取到典型主干工藝路線；以鍛壓裝備制造企業(yè)工藝數(shù)據(jù)的典型主干工藝路線提取為例,驗證了該方法的有效性。 第4章提出了一種基于典型主干工藝路線的零件加工工藝排序方法。結(jié)合工步優(yōu)先權(quán)系數(shù)和工步約束矩陣,建立了工藝排序的數(shù)學(xué)模型；利用粒子群算法來進(jìn)行工步排序,并對所得的最優(yōu)工步元序列進(jìn)行組合,形成工序,以獲得完整的工藝路線；以鍛壓機的主軸零件工藝排序進(jìn)行了應(yīng)用驗證。 第5章結(jié)合具體的企業(yè)項目,開發(fā)了鍛壓裝備工藝設(shè)計系統(tǒng),并且該系統(tǒng)在企業(yè)工藝設(shè)計中得到成功運行與應(yīng)用,驗證了本文所提出的理論和方法的可行性及實用性。 第6章對全文的主要研究內(nèi)容和創(chuàng)新點進(jìn)行了總結(jié),并展望了以后相關(guān)領(lǐng)域工作的努力方向。
[Abstract]:Part process planning is directly related to the production efficiency and quality of products. In the fierce market competition, how to improve the technological innovation ability of parts has become an important problem in product design and manufacture. According to the requirements of the enterprise, this paper studies the decision technology of feature processing scheme, the extraction technology of typical trunk process route and the technology of part process sequencing, and develops the process planning system of forging equipment. The system has been applied in the process design of forging equipment, and good results have been obtained. The main contents of this paper are as follows: The first chapter analyzes the problems existing in the traditional part process planning method, introduces the key technology of the part process planning and its research status, and expounds the research background, content and organization structure of this paper. In chapter 2, a part feature processing scheme decision method based on support vector machine (SVM) is proposed. The binary classification algorithm and multi-valued classification algorithm of support vector machine are introduced, the integrated decision model of feature processing method selection is established, and taking the inner circle feature as an example, the model structure of sub-support vector machine and the pretreatment of input output parameters are introduced. The selection of penalty parameters and kernel functions, and the training and prediction of feature parameters of parts by using LIBSVM software and Matlab platform are realized. In chapter 3, a method of extracting typical trunk process route by cluster analysis is presented. The similarity measure factor of the trunk process route is constructed, and a multi-level similarity comprehensive measure method is proposed to calculate the similarity degree of the trunk process route, which is based on the similarity calculation. In order to reduce the difficulty and complexity of clustering, the main process route design structure matrix is realized by iterative clustering algorithm. The typical trunk process route is extracted from the cluster, and the effectiveness of this method is verified by taking the typical trunk process route extraction from the process data of forging equipment manufacturing enterprise as an example. In chapter 4, a part processing scheduling method based on typical trunk process route is proposed. Combined with step priority coefficient and step constraint matrix, the mathematical model of process scheduling is established, and the process is formed by combining the optimal step element sequence with particle swarm optimization algorithm. In order to obtain a complete process route, the main shaft parts of the forging press process sequencing has been applied to verify. In chapter 5, the process planning system of forging equipment is developed, and the system is successfully run and applied in the enterprise process planning, which verifies the feasibility and practicability of the theory and method proposed in this paper. Chapter 6 summarizes the main contents and innovations of the paper, and looks forward to the future work in related fields.
【學(xué)位授予單位】：浙江大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2013
【分類號】：TH162;TG315

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 王志紅;杜平安;郭志龍;梁山虎;;基于遺傳算法與動態(tài)規(guī)劃法的工藝過程優(yōu)化[J];電子科技大學(xué)學(xué)報;2007年01期

2 王玫;王卓;王杰;;回轉(zhuǎn)類零件的人工神經(jīng)網(wǎng)絡(luò)工序選擇方法研究[J];電子科技大學(xué)學(xué)報;2010年03期

3 徐平;高奇;王明旭;;基于遺傳算法的計算機輔助工藝計劃工序決策優(yōu)化[J];中國工程機械學(xué)報;2006年03期

4 肖偉躍;楊繼榮;蔡悅?cè)A;;基于模糊Petri網(wǎng)考慮制造約束的零件表面加工方案決策方法[J];工程設(shè)計學(xué)報;2010年02期

5 張恒文;丁霞慶;蔡穎;吳俊杰;;基于實例推理的工藝規(guī)劃[J];航空制造技術(shù);2006年09期

6 孔波;鄭喜英;;支持向量機多類分類方法研究[J];河南教育學(xué)院學(xué)報(自然科學(xué)版);2010年02期

7 陳奇;相似字符串的比較方法[J];華南師范大學(xué)學(xué)報(自然科學(xué)版);1999年02期

8 朱海平;肖詩旺;黃剛;;基于遺傳算法的工藝過程排序研究[J];華中科技大學(xué)學(xué)報(自然科學(xué)版);2006年03期

9 孫麗,王秀倫,景寧;CAPP系統(tǒng)中基于實例的推理及檢索方式的研究[J];機床與液壓;2001年06期

10 黃嵐,王康平,周春光,龐巍,董龍江,彭利;粒子群優(yōu)化算法求解旅行商問題[J];吉林大學(xué)學(xué)報(理學(xué)版);2003年04期

相關(guān)博士學(xué)位論文 前1條

1 秦寶榮;智能CAPP系統(tǒng)的關(guān)鍵技術(shù)研究[D];南京航空航天大學(xué);2003年

相關(guān)碩士學(xué)位論文 前3條

1 劉曉陽;基于多色集合理論的加工工藝規(guī)劃建模與推理技術(shù)的研究[D];河北科技大學(xué);2011年

2 曹小波;基于優(yōu)先工序序列實例的CAPP策略及其關(guān)鍵技術(shù)研究[D];西北工業(yè)大學(xué);2005年

3 徐正;三維CAPP中零件特征提取及基于遺傳算法的工藝排序研究[D];華中科技大學(xué);2005年

，

本文編號：1947334