您好,欢迎访问三七文档
当前位置:首页 > 商业/管理/HR > 其它文档 > 试验设计方案书与数据处理方法解析版
1/17单因素方差分析:用同种原料织成的布,用不同的染整工艺处理后进行缩水率试验,试用方差分析考察不同工艺对缩水率是否有显著影响。这里采用不同的染整工艺,每种工艺处理四块布,记A为染整工艺这一因子。它的五个水平记为,,,,AAAAA试验结果如下:染整工艺布样号AAAAA14.36.16.59.39.527.87.38.38.78.833.24.28.67.211.446.54.18.210.17.8解:n5,m41A2A3A4A5A14.36.16.59.39.527.87.38.38.78.833.24.28.67.211.446.54.18.210.17.821.821.731.635.337.5147.92111nmijijCxnm1093.72052475.24470.89998.561246.091406.254597.032111nmijijRxm1149.257518.4937.2142.2586.4990.2560.8453.2968.8975.6977.4410.2417.6473.9651.84129.9642.2516.8167.24102.0160.84131.82124.95252.34316.03358.49211nmijijWx1183.632/172111nmijijCxnm=1093.72052111nmijijRxm=1149.2575211nmijijWx=1183.63则组内偏差平方和:eSSWR=34.3725组间偏差平方和:tSSRC=55.537总偏差平方和:TSSWC=89.9095方差来源偏差平方和自由度方差估计值F值0.05,AEFff0.01,AEFff显著性分组因素55.537413.884256.0593.064.89**随机误差34.3725152.2915总和89.9095193/17双因素无重复试验设计方差分析:为了考察材质和淬火温度对某种钢材淬火后的弯曲变形的影响,对四种不同材质分别用五种不同的淬火温度进行试验,测得其淬火后试件的延伸率数据如下:材质甲乙丙丁温度8004.45.24.34.98205.35.05.14.78405.85.54.84.98606.66.96.67.38808.48.38.57.9试对表中数据作方差分析来回答:不同材质对延伸率有显著影响吗,不同温度对延伸率有显著影响吗?解:a4,b5甲乙丙丁0jT20jT8004.45.24.34.918.8353.448205.35.05.14.720.1404.018405.85.54.84.9214418606.66.96.67.327.4750.768808.48.38.57.933.11095.610iT30.530.929.329.7T120.4201bjjT3044.822011bjjTa761.20520iT930.25954.81858.49882.09201aiiT3625.642T14496.162011aiiTb725.1282Tab724.8084/1719.3627.0418.4924.0188.928.092526.0122.09101.1933.6430.2523.0424.01110.9443.5647.6143.5653.29188.0270.5668.8972.2562.41274.11211abijijx763.16TSS2211abijijTxab38.352ASS22011aiiTTbab0.32BSS22011bjjTTaab36.397eTABSSSSSSSS1.635方差来源偏差平方和自由度方差估计值F值0.05,EFff0.01,EFff显著性A因素0.3230.1066670.7828753.495.95不显著B因素36.39749.0992566.783493.265.41**随机误差1.635120.13625总和38.352195/17双因素有重复试验设计方差分析:试确定三种不同的材料和三种不同的使用环境对蓄电池输出电压的影响,为此,对每种水平组合重复测输出电压4次,测得数据100V列入下表,试分析各因素及因素之间交互作用的显著性。温度C每行总计00iT101827材料1130,155,74,18034,40,80,5020,70,82,589732150,188,159,126136,122,106,11522,70,58,4512973138,110,168,160174,120,150,13996,104,82,601501每列总计00jT173812667673771解:a3,b3,r4111abrijkijkTx=3771220011aAiiTSSTbrabr=114881939-14220441=118161236220011bBjjTSSTarabr=115211689-14220441=3929512362222000001111111ababABijijijijTSSTTTrbrarabr=11111829261-4881939-5211689+14220441=111924121236220111111abrabeijkijijkijSSxTr=1475281-1829261=179664方差来源偏差平方和自由度方差估计值F值0.05,EFff0.01,EFff显著性A因素11816259088.883.355.49**B因素39295219647.529.533.355.49**交互作用AB11192427984.202.734.11**误差1796627665.46/17正交试验设计直观分析法和方差分析法:自溶酵母提取物是一种多用途食品配料,为探讨外加中性蛋白酶的方法,需作啤酒酵母的最适自溶条件试验,为此安排如下试验,试验指标为自溶液中蛋白质含量(%),取含量越高越好。因素水平表如下:水平因素A(单位:℃)B(pH值)C(加酶量:%)1506.52.02557.02.43587.52.8试验结果如下,试进行直观分析和方差分析,找出使产量为最高的条件。试验号ABC空列含量111116.25212224.97313334.54421237.53522315.54623125.507313211.408321310.90933218.95解:直观分析试验号因素含量ABC空列1K15.7625.1822.6520.742K18.5721.4121.4521.873K31.2518.9921.4822.971K5.258.397.556.912K6.197.147.157.293K10.426.337.167.66极差R5.172.060.40.75主次顺序ABC优水平3A1B1C3D优组合311ABC7/17方差分析n9,r3处理号ABC空列试验结果1jK1155..77662255..11882222..66552200..77441niiTx65.582TCTn21nTiiQx=TTSSQPm2ij1Kji1QrjjSSQCT2jK1188..55772211..44112211..44552211..88773jK3311..22551188..99992211..44882222..997721jK224488..3388663344..0033551133..0022443300..115522jK334444..8844445588..3399446600..1100447788..330023jK997766..5566336600..6622446611..3399552277..6622jSS45.46.490.310.83ABCedfdfdfdf3-1=22AAASSMSdf45.422.72,2BBBSSMSdf6.493.232,2CCCSSMSdf0.310.1552,2eeeSSMSdf0.830.4152因为222CeMSMS,所以因素C的偏差平方和、自由度并入误差的偏差平方和、自由度根据以上计算,进行显著性检验,列出方差分析表,结果见表:方差来源偏差平方和自由度方差估计值F值0.05,EFff0.01,EFff显著性A因素45.40222.7079.66.9418.00**B因素6.4923.2411.46.9418.00*C因素0.3120.16误差e0.8320.41误差e△1.1440.285总和53.03因素A高度显著,因素B显著,因素C不显著。本试验指标越大越好。对因素A、B分析,确定优水平为3A、1B;因素C的水平改变对试验结果几乎无影响,从经济角度考虑,选1C。优水平组合为311ABC。即温度为58℃,pH值为6.5,加酶量为2.0%。8/17一元回归分析:某水稻品种在5月至8月期间播种,测定播种至齐穗的天数x和播种至齐穗的总积温y(单位:日·度)的数据,如下表,ix5055606570iy1441.81494.91547.11600.91646.3设y关于x有线性关系ˆyabx,试用最小二乘法确定回归系数,相关系数,并进行方差分析。解:序号kkxky2kxkkxy2ky1501441.82500720902078787.242551494.9302582219.52234726.013601547.13600928262393518.414651600.94225104058.52562880.815701646.349001152412710303.69∑30077311825046643511980216.1611niixxn=60,11niiyyn=1546.222111nnxiiiiSSxxn=2118250-30025051111nnnxyiiiiiiiSPxyyxn=1466435-7731300=2575522111nnyiiiiSSyyn=2111980216.16-773126543.965xyxSPbSS=257510.3250aybx=1546.2-10.360928.2ˆ928.210.3yx9/172xyRxSPSSSS=26522.5,2xyryxSPSSSSSS=21.46方差来源偏差平方和自由度方差F值0.05F0.01F显著性回归26522.5126522.53709.4410.1334.12**残差21.4637.152121ˆnixyiRnyxyiiyySPSSSSSSSSyy=0.9996120.05,30.05311,3F0.878120.01,30.01311,3F0.9590.01,3,所建立的回归方程是高度显著的。10/17粗大误差判别:假设以下数据不含有系统误差,试用3、肖维勒、t检验、格拉布斯、狄克松准则来判别该数据中是否含有粗大误差。0.01序号12345数值1866987755序号678910数值935446239序号1112131415数值6756759126解:序号xv2vv2v118-39.531562.62-43.51892.252668.4771.744.520.2539840.471637.8236.51332.2547719.47379.0815.524
本文标题:试验设计方案书与数据处理方法解析版
链接地址:https://www.777doc.com/doc-3270860 .html