再励学习在卫星姿态控制中的仿真研究

为了满足卫星姿态控制系统对控制精度、抗干扰和鲁棒性要求的不断提高，将模糊神经网络结合再励学习算法应用到卫星姿态控制系统中，即可以在不需要被控卫星的精确数学模型的前提下解决网络参数在线调整的问题，又可以在无需训练样本的前提下实现控制器的在线学习。最后同传统PID控制相比的仿真结果表明，基于再励学习的三轴稳定卫星姿态控制系统不仅可以达到卫星姿态控制任务对控制精度的要求，还可以有效地克服干扰，从而达到了在线学习的目的。     

具有混沌激励函数的BP网络算法

该文讨论了BP网络学习过程中的假饱和现象，并给出了一种改进的算法，有效地解决了假饱和的问题。仿真结果表明，该方法不但可以提高网络学习的快速性，而且具有一定的避免权值落入局部极小点的能力，从而提高了网络的收敛精度，同时，该算法还能提高网络的泛化能力。     

Application of Fractal Theory in Simulation of Ferromagnetic Elements‘Hysteresis Loop in Transformer

The simulation of the transformer transient is one of the indispensable qualifications for improving the performance of transformer protection, the key technique of the transformer‘ s transient simulation is the treatment of ferromagnetic elements‘ loop. Thus the shapes of the primary hysteresis loop and each internal secondary hysteresis loop in the identical magnetism conducting are analyzed, and then it is proposed that there are somefractal characteristics in the relation between them. The fractal phenomenon of the ferromagnetic dements‘ hys-teresis loop in the transformer‘‘ s transient simulation is first brought forward, the mutuality between the ferro-magnetic dements‘ primary hysteresis loop and its secondary hysteresis loops is revealed in mechanism by usingthe fractal theory. According to the iterated function system of fractal theory, the secondary hysteresis loops canbe generated by the iterative calculation of the primary loop. The simulation results show the validity of this idea.     

Multi-Valued Associative Memory Neural Network

A novel learning method for multi-valued associative memory network is introduced,, which is based on Hebb rule, but utilizes more information. According to the current probe vector, the connection weights matrix could be chosen dynamically. Double-valued and multi-valued associative memory are all realized in our simulation experiment. The experimental results show that the method could enhance the associative success rate.     

均方可镇定系统的随机状态反馈控制策略

提出网络控制系统随机最优控制律中随机R iccati方程在一定条件下存在稳态解.给出了稳态解存在的条件,并在此基础上提出了均方可镇定系统的随机状态反馈控制策略.该方法在保持控制性能的基础上,减小了控制律的复杂性,具有实际应用价值.     

一种新的双向联想记忆的学习算法

提出了一种新的用于双向联想记忆的学习算法，该算法利用了输入向量各元素之间的关联信息，在联想的过程中，动态地调整权值矩阵，增强了网络适应能力，利用了更多的已知信息，从而提高了网络的性能．     

80C196KC使用注意事项两则

我们在80C196KC单片机的开发过程中,发现80C196KC单片机使用时有2点需要特别注意:“复位”信号的使用和16位“字”操作。为避免各位同事走弯路,特写此文。     

利用BP神经网络进行伺服系统间隙非线性补偿

目的为了消除普遍存在于伺服系统中的间隙非线性的影响,提出一种利用BP神经网络进行非线性补偿的方法.方法以某武器跟踪伺服系统为例,采用一个3层BP神经网络对其间隙非线性特性进行离线辨识,然后根据辨识结果设计一个非线性补偿器.结果仿真结果表明所提出的方法能够有效消除间隙特性引起的系统自振荡(极限环),并且能够提高系统精度.结论利用BP神经网络进行间隙非线性补偿的方法能够有效解决伺服系统中间隙特性带来的影响,且易于在工程中实现.     

ATMS Based Information Fusion Target Recognition Method

The non-monotonic problem exited in information fusion systems is solved. Through theintroducing of non-monotonic reasoning method, which was realized with ATMS, into the informationfusion system, it gains the ability to process insufficient information with flexibility and non-monotonicbehavior.In the simulation test of our system, our system manifests its ability of dealing the insufficient