微型智能低空速传感器的设计

利用大气中的两个物理参量大气静压力、大气动压力和空速的关系,分别用两个高精度压阻式绝对压力传感器对大气静压力和大气动压力进行测量,再用温度传感器对两个绝对压力传感器的测量值进行温度修正,将修正值作为多层感知机神经网络的输入,空速值作为输出.对标准或实验数据采用动批量法进行网络训练,得到网络各层的权值和偏置值,由此可以确定在任意静压力和动压力输入作用下,网络输出估计的空速值,最后借助数字信号处理器芯片实现微型智能低空速传感器的设计.实验结果表明通过训练、学习的传感器具有更高的精确度,同时能够很好地抑制时间漂移.并且很好地解决了低空速测量误差大的问题.

Dynamic programming algorithm for economic lot-sizing problem with bounded inventory and out-sourcing

This paper addresses a single item dynamic lot-sizing model with inventory capacity and out-sourcing. The goal is to minimize the total costs of production, setup, inventory holding and out-sourcing. Two versions of an out-sourcing model with time-varying costs are considered: stock out case and conservation case. Zero Inventory Order property has been found and some new properties are obtained in an optimal solution. Dynamic programming algorithms are developed to solve the problem in strongly polynomial time respectively. Furthermore, some numerical results demonstrate that the approach proposed is efficient and applicable.