表 因素水平表VIP免费

表因素水平表水平因素A(g·mL-1)B（℃）C（℃）11.041256521.061357031.0814575表正交试验结果NABCD综合评分%1111186.762122289.023133390.094212395.205223195.046231296.047313292.168321389.589332191.58K1265.87274.12272.38273.37K2286.28273.64275.80277.23K3273.32277.71277.29274.87R20.414.074.913.85表方差分析表变异来源SSVMSFPA71.132.0035.5628.270.03B3.312.001.651.310.43C4.222.002.111.680.37D2.522.001.26总变异81.178.00由方差分析结果可知，A因素对综合指标的影响显著，结合方差分析，选取A2水平；B、C对指标影响无显著，随着温度的升高，HSYA的损失率升高增大。结合工艺成本考虑，最后确定最佳工艺为A2B1C1，即：浓缩液相对密度为1.06g·mL-1，进风温度为125℃，出风温度为65℃。表试验因素水平表水平因素A乙醇浓度（%）B乙醇用量(倍)C提取时间(h)15581.0265101.5375122.0表正交试验及结果No列号评价指标ABCD芍药苷含量（mg/g）红景天苷含量（mg/g）总皂苷含量（mg/g）醇浸出物量（%）综合评分1111132.838.8928.8230.0085.862122230.5810.3129.4631.8689.233133330.2710.3932.5529.8991.364212337.859.9431.6129.7895.205223137.9110.0031.5532.0396.066231233.3510.6133.7031.7996.007313235.219.6230.5332.7892.178321330.5810.4530.0630.0989.609332129.1310.3333.7929.8891.38K1266.45273.23271.46273.29K2287.26274.89275.81277.41K3273.15278.74279.59276.16R20.815.528.134.11表方差分析表方差来源离均差平方和自由度均方FP处理因素A75.242.0037.6225.35<0.05B5.342.002.671.80C11.032.005.513.71误差2.972.001.48总变异94.578.00直观分析可知，影响提取效果的因素顺序为：A(乙醇浓度)>C(乙醇用量)>B(提取时间)。经方差分析知：乙醇浓度对提取效果有显著影响，而乙醇用量、提取时间对提取效果无显著性影响，结果最佳提取工艺为A2B3C3，鉴于乙醇用量和提取时间对提取效果无显著性影响。因此，为降低成本、节省工时，将B3改为B1，将C3改为C1，由此得提取工艺为：A2B1C1。确定黄芪、赤芍、红景天醇提条件为：65%乙醇8倍量，提取2次，每次1小时。表Box-Behnken设计及实验结果时间挤出转速（A）滚圆转速（B）滚圆时间（C）f-2Φ13030551.1724030555.0633040556.3844040555.3153035352.6264035357.5673035751.8884035750.0593530354.20103540350.86113530748.20123540755.31133535559.09143535559.24153535559.00表估计回归系数编号系数系数标准误TP常量59.11000.632993.3980.000A0.74130.38761.9130.114B1.15370.38762.9770.031C-1.22500.3876-3.1610.025A2-1.87250.5705-3.2820.022B2-2.75750.5705-4.8340.005C2-4.21000.5705-7.3800.001AB-1.24000.5481-2.2620.073AC-1.69250.5481-3.0880.027BC2.61250.54814.7670.005表方差分析结果来源自由度SeqSSAdjSSAdjMSFP回归9166.457166.456818.495215.390.004线性327.05027.04979.01667.500.027平方394.49894.497831.499326.210.002交互作用344.90944.909314.969812.460.009残差误差56.0086.00821.2016失拟35.9795.97881.9929135.570.007纯误差20.0290.02940.0147合计14172.465从方差分析表中可以看出，回归项的P值为0.004，表明应该拒绝原假设，认为本模型总的来说是有效的。看方差分析表中的失拟现象，本例中，失拟项对应的P值为0.007，明显小于显著性水平0.05，拒绝原假设，认为本模型中存在失拟现象。线性关系R2=0.9295，R-Sq与R-Sq（调整）比较接近，认为模型的拟合度较好；R-Sq（预测）大于R-Sq值，说明用这个模型进行预测的效果比较可信。从表中可以看到，挤出转速、滚圆转速，滚圆时间以及它们的平方项对应的概率值都小于显著性水平，说明这些效应都是显著的；而挤出转速和滚圆转速的交互效应项对应的概率值为0.073，大于显著性水平，认为该效应项是不显著的。用Mintab软件得回归方程：Y=7.97350+8.38150+8.38150+6.69250-0.0749000A2-0.110300B2-1.05250C2-0.0496000A*B-0.169250A*C+0.261250B*C