均方根误差

The following conclusions have been reached, milk has an obvious absorption in the mid-infrared band, and the locations of the absorption peak are consistent with the attributes of molecule functional group analysis and the information from FOSS Company; the effects of the factors on the repeatability are below 3‰; For the quantitative analysis, the method of factorial design is applied to optimize the calibration set. The outliers are determined considering the residual variance distribution. The correlations are 0.990 and 0.986 for fat and protein, respectively. The Root Mean of Square Error of Prediction is 0.121 and 0.111 for fat and protein, respectively. With the application of the methods of wavelengths selection, the numbers of wavelength is decreased greatly from 3301 to 100. The speed is accelerated and the accuracy of the calibration is improved to some extent.

结果表明：牛奶在中红外的吸收特性明显，吸收峰的位置与分子结构的谱带归属以及乳品分析仪器的成功厂商FOSS公司提供的资料基本一致；各种因素对光谱测量影响的CV值基本可以控制在3‰以下；在定量检测中，通过析因设计方法优选校正集样品，结合样品的残差图确定奇异点，脂肪模型和蛋白模型的相关系数分别为0.990和0.986，预测均方根误差分别为0.121和0.111，而在采用波长选择对模型进行优化后，波长从3301个减为100个以内，加快了运算的速度，同时还在一定程度上提高了模型的精度。

Correlation coefficient of prediction set for the correction model of fat and protein are 0.990 and 0.994, and root mean square error prediction is 0.007 or 0.004, respectively.

所建立的脂肪和蛋白质校正模型的预测集相关系数分别为0.990和0.994，预测均方根误差分别为0.007与0.004。

Model parameters were iteratively identified untilthe root mean square error between the respiratory and model-predictedimpedance was minimum.

模型参数进行重复识别直到呼吸阻和模型预测阻间的均方根误差达到最小。

It was found that both the root mean square error and the mean error were minimal when using the neural network and fuzzy controller.

在临床试验部分，本研究针对四位受测者进行不同控制器的控制品质比较，结果发现以类神经网路加上模糊控制器的控制方法，其均方根误差及平均差皆为最小。

Experimental results show that it outperforms the discrete wavelet transform approach with 33% to 36% reduction in root mean square error, 0.9% to 53% improvement in standard deviation of gray level, 39% to 42% enhancement in mutual information, and 0.3% to 38% improvement in space frequency.

实验结果表明，与小波变换融合方法相比，均方根误差减小了33%~36%，灰度标准差提高了0.9%~53%，互信息提高了39%~42%，空间频率提高了0.3%~38%，客观评价准则与目视效果吻合良好。

The experimental results show that the predicting result of RBFNN is better than that of BPNN. Root mean square error of prediction decreases from 0.72 0Brix to 0.65 0Brix.

研究结果表明：径向基函数神经网络比反向神经网络获得的预测结果更好，校正模型的预测均方根误差从0.720Brix降低到0.650Brix。

The performance of the final model was evaluated according to the Root Mean Square Error of Prediction and the Correlation coefficient in prediction and calibration sets.

研究了利用联合区间偏最小二乘法和遗传偏最小二乘法等筛选特征谱区的方法，通过交互验证法确定偏最小二乘模型的主成分因子数和筛选区间，并以预测均方根误差和相关系数R作为模型的评价指标。

Monte Carlo simulation results show that the particle filter with threshold de-noising can effectively reduce the filter root mean square error and improve filter accuracy, and that in the corresponding linear and non-linear systems are reduced by 14% and 12% respectively.

在所采用的线性及非线性系统中，均方根误差均值分别降低了14%和12%。

Then, the effect of PM filter was evaluated by comparing the root mean square error of the standard FA maps and the processed FA maps.

通过比较标准FA图与处理后FA图间的均方根误差，评价PM滤波的效果。

The root mean square error of prediction was 0.006 03. The regression coefficient was 0.99456and the average recovery of isoniazid was99.772%.

对预测集样品的预测均方根误差为0.00603；回归系数为0.99456；加样平均回收率为99.772%。

According to the clear water experiment, aeration performance of the new equipment is good with high total oxygen transfer coefficient and oxygen utilization ratio.

曝气设备的动力效率在叶轮转速为120rpm～150rpm时取得最大值，此时氧利用率和充氧能力也具有较高值。

The environmental stability of that world - including its crushing pressures and icy darkness - means that some of its most famous inhabitants have survived for eons as evolutionary throwbacks, their bodies undergoing little change.

稳定的海底环境─包括能把人压扁的压力和冰冷的黑暗─意谓海底某些最知名的栖居生物已以演化返祖的样态活了万世，形体几无变化。