| 1. | Ann is trained through fast bp algorithm with variable learning rate that mixed with momentum factor . this algorithm can improve network ' s reliability
算法上采用含有动量因子的自适应调整学习率的变学习率算法对网络进行学习训练，以提高网络的学习速度，并且可以增加算法的可靠性。 |
| 2. | The actual prediction ability of bp network model is improved and training velocity is increased by adding momentum factor and using the weight control calculation in the bp network model
Bp网络模型在加入动量项和引入权值控制算法进行改进后，不但预测能力有所改善，网络的训练速度也有了很大的提高。 |
| 3. | Ann is trained through fast bp algorithm with variable learning rate that mixed with momentum factor and levenberg - marquardt algorithm . these algorithms can improve network ' s convergence speed
算法上采用含有动量因子的自适应调整学习率的变学习率算法或levenberg一marquardi优化方法对网络进行训练，以提高网络的收敛速度。 |
| 4. | Besides the adaptive adjust of learning rate and momentum factor , this algorithm can give appropriate permitted convergence error with adaptive adjust in the course of learning . so it overcomes the faults of the controlling means used in traditional bp algorithm
该算法在学习过程中，除了对学习率和动量因子进行自适应调整外，还能根据网络的实际训练情况自适应确定允许均方误差的值，从而克服了传统的bp算法在学习过程中采用的控制方法存在的缺陷。 |
| 5. | Section of ann theory and its application study : after comparing study various model ' s algorithm , an improved bp algorithm has been adopted , i . e . a momentum factor a has been inducted . it can utilize the learning result of previous step during the training process , which can accelerate the training process of traditional bp algorithm
神经网络理论及应用研究部分:算法研究?对各种网络模型算法作出比较后，采用一种基于传统bp算法的改进算法，即在计算过程中引入动量项a项以利用上一步学习结果，加快了传统bp算法的学习速度。 |
| 6. | This paper focuses on the theories and controller designs of forward neural netwoks ? bp network . at first , the structure and algorithms of bp network are deeply researched , the ralations between momentum factor and convergence speed 、 convergence accuracy are revealed and a kind of improved bp algorithm is presented . then the identification method based on bp network with adaptive learning rate is studied and the simulaton indicates it can adaptively track the plant
本课题主要针对前向神经网络? ? bp网络理论与控制器设计进行研究。首先重点对bp网络的结构和学习算法进行了深入研究，揭示了动量因子与网络收敛速度、收敛精度之间的关系，并提出了一种改进的算法。然后研究了采用自适应学习率bp网络的辨识方法，仿真说明其可以自适应地跟踪辨识被控对象。 |
| 7. | Using the first order and the second order derivative information of the network error function to the learning rate factor / / and the momentum factor ? , dynamic optimization of the learning rate factor / / and the momentum factor ? is obtained during the network training process , which efficiently speeds up the network learning rate
并利用网络误差函数对训练速率系数和冲量系数的一阶和二阶导数信息，在网络训练过程中动态优化调整训练速率系数和冲量系数，有效地加快了网络的训练速度。 |
| 8. | Realization of improved bp algorithm - single output three layers " artificial neural network generator base on improved bp algorithm has been developed by the author , and the generator has some functions that the number of neuron in first and second layer and theirs related training parameters such as learning rate . momentum factor a and the value of sum error e can all be self - defined by the users ; connection weights and threshold in each layer ' s neuron training data and teaching signals can also be input or modified in the friendly interface
生成器功能是:网络结构中的第一、二层神经元个数和训练参数中的学习速率粉，动量因子a和期望误差值:可由用户在一定范围内自定义;各层的权值、阀值、网络初始样本值及教师值可在友好的界面下输入、修改。 |
