Neural network-based control for the fiber placement composite manufacturing process

At McDonnell Douglas Aerospace (MDA), an artificial neural network-based control system has been developed and implemented to control laser heating for the fiber placement composite manufacturing process. This neurocontroller learns the inverse model of the process on-line to provide performance that improves with experience and exceeds that of conventional feedback control techniques. When untrained, the control system behaves as a proportional-integral (PI) controller. However, after learning from experience, the neural network feedforward control module provides control signals that greatly improve temperature tracking performance. Faster convergence to new temperature set points and reduced temperature deviation due to changing feed rate have been demonstrated on the machine. A cerebellar model articulation controller (CMAC) network is used for inverse modeling because of its rapid learning performance. This control system is implemented in an IBM-compatible 386 PC with an A/D board interface to the machine. Work supported by the McDonnell Douglas Independent Research and Development Program Export Authority: 22CFR 125.4(b)(13)     

Cover Feature: Superior Proton-Transfer Catalytic Promiscuity of Cytochrome c in Self-Organized Media (7/2021)

Evolutionarily elderly proteins commonly feature greater catalytic promiscuity. Cytochrome c is among the first set of proteins in evolution to have known prospects in electron transport and peroxidative properties. Here, we report that cyt c is also a proficient proton-transfer catalyst and enhances the Kemp elimination (KE; model reaction to show proton transfer catalytic property) by ∼750-fold on self-organized systems like micelles and vesicles. The self-organized systems mimic the mitochondrial environment in vitro for cyt c. Using an array of biophysical and biochemical mutational assays, both acid–base and redox mechanistic pathways have been explored. The histidine moiety close to hemin group (His18) is mainly responsible for proton abstraction to promote the concerted E2 pathway for KE catalysis when cyt c is in its oxidized form; this has also been confirmed by a H18A mutant of cyt c. However, the redox pathway is predominant under reducing conditions in the presence of dithiothreitol over the pH range 6–7.4. Interestingly, we found almost 750-fold enhanced KE catalysis by cyt c compared to aqueous buffer. Overall, in addition to providing mechanistic insights, the data reveal an unprecedented catalytic property of cyt c that could be of high importance in an evolutionary perspective considering its role in delineating the phylogenic tree and also towards generating programmable designer biocatalysts.     

高温熔融-电感耦合等离子体原子发射光谱法测定含铁尘泥中5种有害元素

准确、快速地测定含铁尘泥中有害元素,对含铁尘泥节能降废项目研究及综合利用有着重要的指导意义。实验针对含铁尘泥试样因含碳量高,导致常规酸溶解方法如果不增加除碳步骤则很难完全分解试样的问题,采用高温熔融法将试样制成熔融液后倾倒于稀酸中溶解的方法,解决了碳含量高难溶解的问题。采用四硼酸锂-碳酸锂混合熔剂在高温下将试样熔融,熔融液高温倾倒于稀酸中制成澄清溶液,由于骤冷,试样炸裂成微小玻璃珠,无迸溅,操作简单,使得后续溶解过程快速完成,从而建立了电感耦合等离子体原子发射光谱(ICP-AES)测定P、Pb、K、Na、Zn共5种有害元素的方法。按照实验方法测定含铁尘泥中P、Pb、K、Na、Zn,测定值的相对标准偏差(RSD,n=9)在1.4%～5.7%之间,加标回收率在96%～110%之间。研究结果表明方法能够满足含铁尘泥样品中5种元素的定量分析要求。     

在线监测结晶器调宽换向阀阀芯卡阻方法的实现

对某炼钢厂铸机结晶器在线调宽液压系统的工作原理进行了分析,并重点分析了调宽过程中换向阀阀芯卡阻的问题,针对该问题对其液压系统进行了改进,改进后可以在线监测换向阀阀芯是否卡阻。该改进的实施在减少结晶器在线调宽时发生漏钢故障方面起到了重要作用。     

RTM工艺中气泡形成及消除研究

总结了关于RTM（树脂压铸）工艺中气泡缺陷的文献报道，对RTM中气泡缺陷的形成机理、影响因素及各研究者提出的消泡方法、理论和模型进行了概述，同时对RTM工艺中气泡缺陷的研究现状给予充分关注，并对RTM缺陷研究的未来方向进行了展望。     

A novel method for camera external parameters online calibration using dotted road line

On board camera is the most important information source for driver assistant application which is based on computer vision. One problem for on board camera is that the external parameters are easy to be changed when moving on the road. Traditional method either calibrates external parameters offline or calibrates external parameters semi-online, which both need human intervention. By observation, it is found that corner connection of dotted road lane can form two groups of parallel lines. Using this geometric characteristics and through theoretical derivation, a novel online camera external parameter calibration method is proposed which focus on the situation when vehicle is moving. The pro of this method is that it maintains relatively high calculation accuracy and more important, it does not require any human intervention in whole calibration process. Experimental and comparison results show that this method is simple and have accurate results which fully meet the requirements of practical application.     

面向在线社区用户的群体推荐算法研究

结合现有两种主要群体推荐算法的优势, 建立新的算法框架, 并引入差异度因素对模型进行优化。另外, 考虑到在线社区用户的特点, 定义互动度指标来描述群体成员间的互动程度, 通过分析其与推荐精度之间的关系, 探讨互动度对群体推荐的影响。选取豆瓣网数据进行实验, 并与传统方法进行比较, 结果表明, 融入差异度的算法具有更好的推荐效果, 且有效的互动机制能够保证较高的推荐精度。     

单光源多组分红外在线分析仪设计

基于红外光谱吸收原理,设计了一种单光源单光路的在线分析仪。在实验室环境下,利用SO2进行实验,并对实验结果进行分析。实验结果表明:该分析仪具有很好的准确性、重复性和稳定性,实现了多种气体的在线测量。     

变压器油色谱分析与故障判断及处理

阐述变压器油气体色谱分析与变压器内部故障之间的内在联系，通过色谱分析监测变压器的运行状况，并介绍变压器油气体色谱超标在线脱气处理方法。     

树突细胞算法在线分析组件的研究

原始树突状细胞算法（DCA）的离线分析过程,将会导致时间差异,从而产生假警报,增加了虚警率,也会导致攻击的成功发生,这对一个人侵检测系统来说是致命的。因此,文中的目的就是在不影响检测精度的前提下提高检测速度。于是文中提出了分片思想的在线分析组件与DCA相集成的方法,即根据抗原采样数量或者时间将一系列已处理的信息分割成为更小的部分,使得每个分片独立地进行实时的、周期性的分析,这样在每个分片内的入侵攻击就能及时地被识别出来。文中给出了DCA在线分析模块的伪代码描述,并且将其应用于SYN端口扫描的检测实验中。结果表明,DCA在线分析模块在不影响检测精度的前提下有效地提高了检测速度。