Estimation of GM (1, 1) model parameter usually adopts the least square criterion, but test of model precision often USES average relative error criterion.
估计GM(1,1)模型中的参数通常采用最小二乘准则，而在模型精度检验时又常采用平均相对误差。
Firstly, an average performance index containing tracking error and control energy over a class of additive model errors is defined.
首先针对一类相加模型误差的描述，定义了一个平均意义上的包含跟踪误差和控制能量的性能指标。
The effluent treatment process was predicted with this BP neural network model with the average relative error of 19%, which indicates that the generalization power of the network is not so desirable.
利用BP神经网络模型实现了对造纸废水处理过程的预测，平均相对误差为19%，表明网络泛化能力不是很好。
The average value of absolute error is 2.05% and the biggest error is 5.12% in the model of loom efficiency. The results indicate that ANN model has high forecast precision.
对于织机效率模型，预报误差绝对值的平均为2.05%，最大误差为5.12%，可知模型具有较高的预报精度，人工神经网络对此模型的预报可行有效。
The proposed statistical SS2M model has an average error of 4.16% with respect to SPICE Monte Carlo simulations, with an average error of standard deviation of only 3.06%.
同时，提出的统计ss2 M模型与SPICE蒙特卡罗分析结果比较，均值的平均误差在4.16%以内，而方差的平均误差在3.06%以内。
For RBF neural network model, which is more effective to monitoring weld quality than the others, the average error validated is 2.28% and the maximal error validated is under 10%.
径向基函数神经网络模型的监测准确率高于其他两种模型，其平均验证误差为2.28%，最大验证误差低于10%。
Simulation results showed that the ANN model gave better predictions than the regressive model. The average relative error of ANN was 14.9% and that of linear regression was 25.8%.
模拟的结果显示ANN模型比线性回归模型有更好的预测能力，预测的平均相对误差：ANN模型为14.9%，线性回归模型为25.8%。
The estimation model was tested by industrial practical data, its average error is 7.5%. So this model could be used as a guide in practical operation in sintering process.
工业实际数据验证表明，智能集成模型的残硫估计误差平均值仅为7.5%，而且真实反映了烧结块残硫的变化趋势，可以为生产操作提供有益的指导。
It is proved in use that the new method has an advantage over conventional ones, the prediction model has higher precision, the maximum average error is within 1%.
应用结果表明，该方法优于一般预报方法，具有较高的精度，最大相对误差在1%以内。
The average prediction error by this model is 0.959% for the refractive index of 95 amorphous homopolymers.
由该模型对95个聚合物的折光率进行预测，平均相对误差为0.959％。
The average error of the leachate concentration is less than 6%, which is verified the model is reliable and practical.
淋滤液浓度拟合误差小于6%，从而验证了模型的可靠性和实用性。
This paper presents the residual error forecast model of average-growing function by using its residual error data sequence to adjust the model based on the finished forecast model.
在均生函预报模型的基础上，利用其残差数据序列对均生函数预报模型进行校正，提出了均生函数残差预报模型。
GA-BP neural network model is applied in matching and predicting the production of the gas Wells with 5.1% of the average relative error. It proves th...
利用GA - BP神经网络模型对气井产量进行了拟合和预测，拟合的平均相对误差为5.1%，表明新模型适用于洛带气田的产量递减预测。
The results of simulation indicated that, the biggest relative error was 1.3%, and the average relative error was 0.7%. So this model could reflect the change trend exactly, …
仿真结果表明，该模型的最大相对误差为1.3%，平均相对误差为0.7%，可以较准确地反映冰铜温度的变化趋势，为生产操作提供有益的指导。
Some data were chosen to train the network model. The results show that the convergence rate was faster, the model had higher accuracy, the average absolute error can reach 0.002 7%.
对现场收集的数据进行仿真学习，结果表明，该预测模型收敛速度快，具有较高的预测精度，平均绝对误差可达到0.002 7%。
In seismic inversion, the result is non-unique because of the finiteness and error of measured data the obtained model is the average value of the real model.
在地震反演中，由于观测数据的有限性和误差，使得地震反演结果的非唯一性广泛存在，求得的模型是实际模型在一定范围内的平均值。
Generally GM (1,1) model takes the average relative error between restored value of the model and real value as the criterion to evaluate the simulation precision.
模型一般以模型还原值与实际值平均相对误差检验模型的模拟精度。
The average translation error is reached to 0.165 pixels compared with the traditional affine model.
与传统的仿射变换相比，平均平移误差精度达到了0.165个像素。
The results show that the relief would lead to an average-area error calculated by the real and the assumed of the numerical model.
实测地形资料计算结果表明，地形的起伏使得山区的实际有效面积和网格区水平假设面积相对误差可达10%以上。
The results show that the relief would lead to an average-area error calculated by the real and the assumed of the numerical model.
实测地形资料计算结果表明，地形的起伏使得山区的实际有效面积和网格区水平假设面积相对误差可达10%以上。
应用推荐