首页 | 本学科首页   官方微博 | 高级检索  
相似文献
 共查询到20条相似文献,搜索用时 93 毫秒
1.
Conventionally, direct tensile tests are employed to measure mechanical properties of industrially pro- duced products. In mass production, the cost of sampling and labor is high, which leads to an increase of total pro- duction cost and a decrease of production efficiency. The main purpose of this paper is to develop an intelligent pro- gram based on artificial neural network (ANN) to predict the mechanical properties of a commercial grade hot rolled low carbon steel strip, SPHC. A neural network model was developed by using 7 x 5 x 1 back-propagation (BP) neural network structure to determine the multiple relationships among chemical composition, product pro- cess and mechanical properties. Industrial on-line application of the model indicated that prediction results were in good agreement with measured values. It showed that 99.2 % of the products' tensile strength was accurately pre- dicted within an error margin of ~ 10 %, compared to measured values. Based on the model, the effects of chemical composition and hot rolling process on mechanical properties were derived and the relative importance of each in- put parameter was evaluated by sensitivity analysis. All the results demonstrate that the developed ANN models are capable of accurate predictions under real-time industrial conditions. The developed model can be used to sub- stitute mechanical property measurement and therefore reduce cost of production. It can also be used to control and optimize mechanical properties of the investigated steel.  相似文献   

2.
In order to manage and control semiconductor wafer fabrication system (SWFS) more effectively,the daily throughput prediction data of wafer fab are often used in the planning and scheduling of SWFS.In this paper,an artificial neural network (ANN) prediction method based on phase space reconstruction (PSR) and ant colony optimization (ACO) is presented,in which the phase space reconstruction theory is used to reconstruct the daily throughput time series,the ANN is used to construct the daily throughput prediction model,and the ACO is used to train the connection weight and bias values of the neural network prediction model.Testing with factory operation data and comparing with the traditional method show that the proposed methodology is effective.  相似文献   

3.
A new satellite orbit prediction method based on artificial neural network (ANN) model is proposed to improve the precision of orbit prediction. In order to avoid the difficulty of amending the dynamical model, it is attempted to use ANN model to learn the variation of orbit prediction error, and then the prediction result of ANN model is used to compensate the predicted orbit based on dynamic model to form a final predicted orbit. The experiment results showed that the orbit prediction error based on ANN model was less than that based on dynamical model, and the ent satellites and different improvement effects for differtime were different. The maximum rates of improvement of predicting 8, 15, 30 d were respectively 80 %, 77.77 %, 85 %. The orbit prediction error control technique based on the method of back overlap arc compare was brought forward to avoid the risk that the precision of predicted orbit is even worse after it is compensated by ANN model. The phenomena of failure were basically eliminated based on this technique, and the rate of failure was reduced from 30 % to 5 %. This technique could ensure that the engineering application of ANN model could come true.  相似文献   

4.
High temperature heat treatments were conducted for as-cast N08028 alloy to obtain various microstructures with different amounts of σ-phase,and then hot compression tests were carried out using Gleeble-3500 thermo-mechanical simulator in deformation temperature range from 1100 to1200 ℃ and strain rate range from 0.01 to 1 s-1. For the same initial microstructure, the flow stress was observed to increase with increasing the strain rate and decreasing the deformation temperature, while for the same deformation condition, the flow stress was found to increase with increasing the amount of σ-phase in the initial microstructure. Moreover, dynamic recrystallization was found to be the main dynamic soften mechanism. On this basis, Arrhenius-type constitutive equations and artificial neural network(ANN) model with back-propagation learning algorithm were established to predict hot deformation behavior of the alloy. Furthermore, the parameters of constitutive equations were found to be dependent on the initial microstructure, which was also as one of the inputs for the ANN model. Suitability of the two models was evaluated by comparing the accuracy, correlation coefficient and average absolute relative error, of the prediction. It is concluded that the ANN model is more accurately than the constitutive equations.  相似文献   

5.
《矿物冶金与材料学报》2021,28(8):1321-1331
In the prediction of the end-point molten steel temperature of the ladle furnace, the influence of some factors is nonlinear. The prediction accuracy will be affected by directly inputting these nonlinear factors into the data-driven model. To solve this problem, an improved case-based reasoning model based on heat transfer calculation (CBR-HTC) was established through the nonlinear processing of these factors with software Ansys. The results showed that the CBR-HTC model improves the prediction accuracy of end-point molten steel temperature by 5.33% and 7.00% compared with the original CBR model and 6.66% and 5.33% compared with the back propagation neural network (BPNN) model in the ranges of [?3, 3] and [?7, 7], respectively. It was found that the mean absolute error (MAE) and root-mean-square error (RMSE) values of the CBR-HTC model are also lower. It was verified that the prediction accuracy of the data-driven model can be improved by combining the mechanism model with the data-driven model.  相似文献   

6.
This paper creates a LM (Levenberg-Marquardt) algorithm model which is appropriate to solve the problem about weights value of feedforward neural network. On the base of this model, we provide two applications in the oilfield production. Firstly, we simulated the functional relationships between the petrophysical and electrical properties of the rock by neural networks model, and studied oil saturation. Under the precision of data is confirmed, this method can reduce the number of experiments. Secondly, we simulated the relationships between investment and income by the neural networks model, and studied invest saturation point and income growth rate. It is very significant to guide the investment decision. The research result shows that the model is suitable for the modeling and identification of nonlinear systems due to the great fit characteristic of neural network and very fast convergence speed of LM algorithm.  相似文献   

7.
A Worsted Yarn Virtual Production System Based on BP Neural Network   总被引:6,自引:0,他引:6  
Back-Propagation (BP) neural network and its modified algorlthm are introduced. Two series of BP neural network models have been established to predict yarn properties and to deduce wool fiber qua/ides. The results from these two series of models have been compared with the measured values respectively, proving that the accuracy in both the prediction model and the deduction model is high. The experimental results and the corresponding analysis show that the BP neural network is an efficient technique for the quality prediction and has wide prospect in the application of worsted yam production system.  相似文献   

8.
Fashion color forecasting is one of the most important factors for fashion marketing and manufacturing. Several models have been applied by previous researchers to conduct fashion color forecasting. However, few convincing forecasting systems have been established. A prediction model for fashion color forecasting was established by applying an improved back propagation neural network (BPNN) model in this paper. Successive six-year fashion color palettes, released by INTERCOLOR, were used as learning information for the neural network to develop a reliable prediction model. Colors in the palettes were quantified by PANTONE color system. Additionally, performance of the established model was compared with other GM(1, 1) models. Results show that the improved BPNN model is suitable to predict future fashion color trend.  相似文献   

9.
In this paper,a new type of neural network model - Partially Connected Neural Evolutionary (PARCONE) was introduced to recognize a face gender. The neural network has a mesh structure in which each neuron didn't connect to all other neurons but maintain a fixed number of connections with other neurons. In training,the evolutionary computation method was used to improve the neural network performance by change the connection neurons and its connection weights. With this new model,no feature extraction is needed and all of the pixels of a sample image can be used as the inputs of the neural network. The gender recognition experiment was made on 490 face images (245 females and 245 males from Color FERET database),which include not only frontal faces but also the faces rotated from-40°-40° in the direction of horizontal. After 300-600 generations' evolution,the gender recognition rate,rejection rate and error rate of the positive examples respectively are 96.2%,1.1%,and 2.7%. Furthermore,a large-scale GPU parallel computing method was used to accelerate neural network training. The experimental results show that the new neural model has a better pattern recognition ability and may be applied to many other pattern recognitions which need a large amount of input information.  相似文献   

10.
A hybrid model of a subminiature helicopter in horizontal turn is presented. This model is based on a mechanism model and its compensated neural network (NN). First, the nonlinear dynamics of a sub-miniature helicopter is established. Through the linearization of the nonlinear dynamics on a trim point, the linear time-invariant mechanism model in horizontal turn is obtained. Then a diagonal recursive neural network is used to compensate the model error between the mechanism model and the nonlinear model, thus the hybrid model of a subminiature helicopter in horizontal turn is achieved. Simulation results show that the hybrid model has higher accuracy than the mechanism model and the obtained compensated-NN has good generalization capability.  相似文献   

11.
通过对水泥分解炉的特性进行分析,确定了出口气体温度的数学模型。用BP神经网络建立水泥分解炉的控制系统,通过控制煤量和风量来使出口气体的温度保持在规定的范围内。针对BP神经网络通常权值更新方法中出现的收敛速度慢、易陷入局部最优值问题,利用改进的免疫算法来优化神经网络的权值。仿真结果表明该温度控制系统优于通常的神经网络PID控制系统,并具有良好的可靠性、自适应性和鲁棒性。  相似文献   

12.
分析了传统系统辨识方法和神经网络方法各自的优缺点,提出一种应用在强非线性系统辨识中的遗传神经网络模型。针对神经网络的收敛速度和全局收敛性的缺陷,分别采用高速收敛算法和遗传算法进行了改进,取得了较为满意的结果。针对神经网络泛化能力差不易应用的缺点,使用了结构进化方法和结构设计方法中的剪枝算法来改善模型的泛化能力。最后使用2个模拟信号进行仿真试验,结果与分析的结论一致。  相似文献   

13.
目的解决溶解气体分析法在变压器故障诊断中准确率不高之难题。方法采用人工神经网络方法和基于动量因子技术的改进BP网络训练算法。结果建立了一BP神经网络模型,使网络具有较强的学习能力、泛化能力和适应能力。结论神经网络能较准确的对变压器故障进行诊断,具有一定的应用及推广价值。  相似文献   

14.
对作业车间调度问题的换位矩阵表示方法进行了改进,给出新的作业车间调度问题的 Hopfield 神经网络计算能量函数表达式,然后提出改进的 Hopfield 神经网络作业车间调度方法。为了避免 Hopfield 神经网络容易收敛到局部极小的缺点,将模拟退火算法应用于 Hopfield 神经网络求解,提出随机神经网络作业车间调度方法。与已有算法相比,改进算法能够保证神经网络稳态输出为可行的作业车间调度方案。  相似文献   

15.
针对CMAC神经网络的网络节点随输入维数的增大呈几何级数增加的问题,提出了基于模糊聚类的超闭球CMAC神经网络改进算法。该算法通过对输入数据进行模糊聚类确定网络节点数和节点值,并根据输入输出数据通过模糊推理优化算法计算神经网络初始权值。与原算法比较,该算法可有效降低神经网络节点数,提高系统的学习精度。对一个多步时延的非线性系统的辨识仿真结果表明了该算法的可行性与有效性。  相似文献   

16.
为提高 RBF 神经网络的交通流预测精度,提出基于混沌-RBF(Chaos-RBF,C-RBF)神经网络的交通流预测算法,该算法首先计算混沌相空间的嵌入维数和嵌入延迟,构造得到的相空间向量作为 RBF 神经网络的输入,其相空间次邻向量作为期望输出值,滚动训练得到神经网络的权值,然后以实际交通流作为输入,经由网络计算得到预测值。仿真结果表明该算法相比于 RBF 神经网络,预测精度提高 96%,证明了该算法的有效性。  相似文献   

17.
人工神经网络在地震分析预报中的应用   总被引:7,自引:0,他引:7  
为了探索人工神经网络应用于地震分析预报的可能性,以福建及其周边地区地震活动为例,采用b值、地震频次、地震能量释放、空间集中度4项地震活动性指标作为神经网络的输入,用具有S-型特性函数的BP网络对由每年地震的活动指标组成的标准样本进行训练,由训练结束后的权值和阈值及待预报样本的因子测值计算出网络输出值,作为地震活动性的预测.结果表明,用神经网络可以在一定精度范围内使震级预报的内检符合率达100%,在例子中,外推预报准确率达88%以上.  相似文献   

18.
为了克服传统BP神经网络在经济预测中的不足,引入遗传算法对BP神经网络的权值进行训练,在此基础上建立互推递进遗传神经网络的预测模型,经实例测试表明该模型是可行的.  相似文献   

19.
针对RBF神经网络的预测精度受样本数据随机性影响较大,而灰色理论能弱化数据随机性的特点,提出了差值结合法将灰色GM(1,1)模型和RBF神经网络模型有效地结合起来,构建了差值灰色RBF网络预测模型。并利用此模型进行股票价格预测,实证结果表明:该模型预测稳定性较好,预测精度高,平均预测误差为0.68%,与BP神经网络和RBF神经网络相比具有更好的泛化能力和更高的预测精度,在股票预测中具有一定的实用价值。  相似文献   

20.
港口船舶交通流量预测   总被引:2,自引:0,他引:2  
为更精确地对港口或航道内船舶交通流量进行预测,分别建立BP神经网络预测模型和RBF神经网络预测模型进行仿真,并以宁波港船舶交通流量为例进行验证.结果表明,在宁波港现有发展基础和港口设施状况下,RBF神经网络用于宁波港船舶交通流量预测误差较小,预测值与实际值相近.  相似文献   

设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号