This paper also optimizes the support vector machine ( svm ) parameters and controller scale coefficients based on genetic algorithms ( ga ) , in order to get the optimal performance of the controller 利用遺傳算法來優(yōu)化支持向量機(jī)參數(shù)和控制器比例因子參數(shù),以期實(shí)現(xiàn)最優(yōu)的控制性能。
To improve the controlling quality of this system , it permit user modify controlling parameters . user can input integral coefficient - kt . scaling coefficient can be adjusted by controlling system according to e and adjusting scale inputted by user 為進(jìn)一步改善控制系統(tǒng)的控制品質(zhì),允許用戶對(duì)控制參數(shù)進(jìn)行修改, k _ i由用戶輸入, k _ u則根據(jù)用戶設(shè)定的調(diào)整步長(zhǎng)和溫度偏差的情況由控制系統(tǒng)自調(diào)整。
A scale coefficient method based on the fem has been proposed to predict the optimum blank in sheet metal forming in this thesis . in order to improve the precision of results and reduce iterate time , at adjusting the original curve , this method is n ' t giving the same adjustment , but giving the corresponding adjustment based on calculation of scale coefficient and shape error of each node to the curve 本文在有限元仿真的基礎(chǔ)上提出了一種新的板料優(yōu)化的方法? ?比例因子法,該方法在調(diào)整初始輪廓線時(shí),不是給各個(gè)節(jié)點(diǎn)一個(gè)相同的調(diào)整量,而是依各個(gè)節(jié)點(diǎn)的比例因子及形狀誤差值計(jì)算出相應(yīng)的調(diào)整量,這樣調(diào)整的針對(duì)性強(qiáng),計(jì)算結(jié)果更精確,需要迭代的時(shí)間也更少。
In the last chapter of this thesis , we introduce two kinds of sheet metal parts . one is forming part which require very accurate prediction of blank shape , the other is drawing part which need just approximate prediction of blank shape . we apply the scale coefficient method to predict the blank shape of these two kinds of sheet metal parts 文中最后一章,給出了兩類典型的沖壓零件,一類是落料后直接成形的零件,該類零件初始板料的形狀及尺寸要求非常精確;另一類是兩次拉延成形的零件,這類零件的初始板料要求適合的形狀及尺寸即可。