A new grey forecasting model based on BP neural network and Markov chain

A new grey forecasting model based on BP neural network and Markov chain was proposed. In order to combine the grey forecasting model with neural network, an important theorem that the grey differential equation is equivalent to the time response model, was proved by analyzing the features of grey forecasting model（GM（1,1））. Based on this, the differential equation parameters were included in the network when the BP neural network was constructed, and the neural network was trained by extracting samples from grey system＇s known data. When BP network was converged, the whitened grey differential equation parameters were extracted and then the grey neural network forecasting model （GNNM（1,1）） was built. In order to reduce stochastic phenomenon in GNNM（1,1）, the state transition probability between two states was defined and the Markov transition matrix was established by building the residual sequences between grey forecasting and actual value. Thus, the new grey forecasting model（MNNGM（1,1）） was proposed by combining Markov chain with GNNM（1,1）. Based on the above discussion, three different approaches were put forward for forecasting China electricity demands. By comparing GM（1, 1） and GNNM（1,1） with the proposed model, the results indicate that the absolute mean error of MNNGM（1,1） is about 0.4 times of GNNM（1,1） and 0.2 times of GM（I, 1）, and the mean square error of MNNGM（1,1） is about 0.25 times of GNNM（1,1） and 0.1 times of GM（1,1）.

Prediction Model of Sewing Technical Condition by Grey Neural Network

The grey system theory and the artificial neural network technology were applied to predict the sewing technical condition. The representative parameters, such as needle, stitch, were selected. Prediction model was established based on the different fabrics’ mechanical properties that measured by KES instrument. Grey relevant degree analysis was applied to choose the input parameters of the neural network. The result showed that prediction model has good precision. The average relative error was 4.08% for needle and 4.25% for stitch.

Establishment of Neural Network Prediction Model for Terminative Temperature Based on Grey Theory in Hot Metal Pretreatment

In order to improve the accuracy of model for terminative temperature in steelmaking, it is necessary to predict and control before decarburization. Thus, an optimization neural network model of terminative temperature in the process of dephosphorization by laying correlative degree weights to all input factors related was used. Then sim- ulation experiment of model newly established is conducted utilizing 210 data from a domestic steel plant. The results show that hit rate arrives at 56.45~~ when error is within plus or minus 5%, and the value is 100% when within ~10%. Comparing to the traditional neural network prediction model, the accuracy almost increases by 6. 839o//oo. Thus, the simulation prediction fits the real perfectly, which accounts for that neural network model for terminative tempera- ture based on grey theory can reflect accurately the practice in dephosphorization. Naturally, this method is effective and nraeticahle.

Wind-power estimating model based on the experimental data in laboratory

Wind-power （WP） estimation is necessary for power system in several operations, which are as the optimal power flow between conventional units and wind farms, generators scheduling, and electricity market bidding. Estimating the output power of a wind energy conversion unit （WEC） mainly bases on the incident wind speed at the unit site by using the power characteristic curve. In addition, several time-series models have been using in wind speed forecasting. These models are characterized with requiring a large set of data. In order to prevent from the wind speed measurement and the need of a precise wind turbine model, an novel method basing on neural network and the grey predictor model GM （1,1） is proposed. Though the method, the estimating model can be built only by using the experimental data, which are obtained from the WP system in laboratory. The effectiveness of the estimating model is confirmed by the simulation results.

基于改进粒子群优化算法的灰色神经网络模型

根据灰色神经网络的参数随机选择类似于粒子群算法中的粒子初始空间位置,采用改进粒子群算法代替梯度修正法,对网络参数进行了处理,并通过寻找粒子群算法中的最优个体,建立了基于改进粒子群算法的灰色神经网络,提高了预测模型的稳健性和精度.通过解决短期订货量问题,与反向传播(BP)神经网络、灰色神经网络、没有改进的粒子群灰色神经网络算法和基于遗传算法的灰色神经网络等方法进行了比较.分析结果表明,基于改进粒子群算法的灰色神经网络计算更为方便,并具有更好的逼近能力和预测精度.为优化网络模型参数提供了一种新方法,并拓展了预测模型的研究思路.

基于灰色神经网络的公路物流需求量预测模型

以浙江省公路货运量历史数据为例,考虑到影响货运量主要因素,采用灰色神经网络模型GNNM(1,N)进行预测,并与灰色模型GM(1,N)和神经网络预测结果相比较。计算结果表明:该方法在预测公路物流需求量具有有效性;在灰色模型GM(1,N)预测时,通过比较紧邻均值生成序列的生成系数α对预测精度的影响,选取了最优值进行计算从而提高了灰色模型的预测精度。

基于SAPSO优化灰色神经网络的空中目标威胁估计

针对目标威胁估计有很多不确定性的特点,分析了传统目标威胁估计方法和灰色神经网络初始参数随机选择的不足。采用模拟退火改进的粒子群算法代替梯度修正法,对网络参数初始值进行寻优,并通过该方法搜寻到的最优粒子,建立了基于模拟退火粒子群算法优化的灰色神经网络模型,以提高预测模型的稳健性和精确度。与灰色神经网络和没有改进的粒子群灰色神经网络等方法进行比较,仿真实验结果表明,模拟退火粒子群优化的灰色神经网络具有很好的预测能力,可以准确地完成空中目标威胁估计。

Analysis Methods of SrTiO_3 Ceramic's Electricity Performance

The effect of oxidizing-heat-treatment conditions on the electricity performance of doped SrTiO3 ceramic is analyzed by using the theory of grey neural network. Based on the number of main parameters, the model of GNNM （1,1）, GNNM （1,2）, GNNM （1,3） is used to analyze and construct the corresponding model of GNNM （2,1） gray neural network. It can reach the required precision by calculating.

灰色神经网络模型及其应用

灰色建模要求的样本点少,不必有较好的分布规律,而且计算量少,操作简便。而BP网络学习样本时,会反馈校正输出的误差,具有并行计算、分布式信息存储、强容错力、自适应学习功能等优点。本文将灰色预测建模和神经网络技术融合起来,建立灰色神经网络模型(GNNM)。提出计算残差序列和新的预测值的公式。用于发酵动力学预测,结果表明,灰色神经网络模型在预测精度方面优于常规灰色模型。该模型的算法概念明确,计算简便,有较高的拟合和预测精度,拓宽了灰色模型的应用范围。