The static and dynamic characteristic models of stage are built by ansys , which analyses the displacement , stress , natural frequency and vibration model of stage . furthermore , the influence of primary structural parameters on static and dynamic characteristic is analyzed by orthogonal experimental scheme comprehensively and the numerical results from experiment are regressed by stepwise regression method , thus the optimization regression equation is obtained . meanwhile , the neural network model is built and validates the regression model 應(yīng)用ansys軟件建立了工作臺(tái)靜、動(dòng)態(tài)特性的分析模型,并分析了工作臺(tái)的位移、應(yīng)力以及振動(dòng)的固有頻率和振型,而且還應(yīng)用正交試驗(yàn)表安排參數(shù)全面分析了結(jié)構(gòu)參數(shù)對(duì)微位移工作臺(tái)的靜、動(dòng)態(tài)特性的影響,并用逐步回歸分析方法對(duì)數(shù)值計(jì)算結(jié)果進(jìn)行回歸,獲得了微動(dòng)工作臺(tái)的最優(yōu)回歸方程,同時(shí)建立了人工神經(jīng)網(wǎng)絡(luò)模型并對(duì)回歸模型進(jìn)行了驗(yàn)證。
( ii ) this paper makes the annals data of the listed companies in 2004 as sample , and gets earnings per share and the regression equation of the four financial indexes which have noticeable effect on it via stepwise regression method . the four financial indexes which affect earnings per share are the net asset per share , unaccommodated profit per share , quick ratio and return on assets respectively . the equation provides a (二) 、以河南省上市公司2004年年報(bào)數(shù)據(jù)作為樣本,應(yīng)用逐步回歸方法,得到每股收益和對(duì)它有顯著影響的四個(gè)財(cái)務(wù)指標(biāo)的回歸方程,影響每股收益的四個(gè)財(cái)務(wù)指標(biāo)分別是:每股凈資產(chǎn)、每股未分配利潤(rùn)、速動(dòng)比率和總資產(chǎn)報(bào)酬率,為這些上市公司提高每股收益提供參考。