基于BP神经网络的加药控制系统研究

浮选是最重要的选矿方法之一,煤矿浮选工艺自动加药控制问题是影响浮选效果的重要因素;传统加药控制主要是由加药工根据生产经验来控制,为了解决浮选精度低这一问题,文中提出采用BP神经网络作为加药控制系统加药量的预测模型,并采用动量项方法对其进行优化,通过仿真得出动量项法网络和测试样误差曲线,说明带动量项改进的神经网络更容易满足网络精度的要求,并且该方法能够有效地降低误差、减小浪费。     

基于PLC的火电厂电除尘器振打控制的研究

为了提高火电厂中静电除尘器的稳定性、可靠性、自动化水平,采用了高性能施耐德Twido可编程控制器(PLC)以及Twidosoft编程软件,设计了电除尘器电机振打控制方案,编制了振打控制梯形图程序;当沉降室侧壁振打3分钟,停7分钟;第一电场振打2分钟,停8分钟;第二电场振打2分钟,停18分钟;第三电场振打2分钟,停28分钟时,振打效果较好;运行结果表明:采用Twido可编程控制器之后,实现了电机振打的自动运行,与传统继电器振打控制系统相比,可靠性高,稳定性好,除尘效率高,粉尘排放浓度低于50mg/m3,为火电厂静电除尘器电机振打系统的自动控制提供了新的方法。     

遗传算法模糊PID测控系统在环控试验台中的应用

飞机环控试验台须模拟流量0～14000kg/h、压力0～2.5MPa和常温～500℃的空气环境;项目要求测控范围广、精度±1%且不超调;空气状态具有非线性、时变等特点,且控制参数之间存在复杂耦合;针对以上难点,设计了分布式测控系统,提出了改进的智能PID控制方案;通过遗传算法分段整定PID参数,离线建立PID数据库,使系统能够根据控制目标值选择最优PID初值;在此基础上,结合模糊推理在线调整PID参数,使系统具有了自适应性,能在具体工况和干扰下达到很好的控制效果;实际应用中完全满足了指标要求,解决了传统PID的控制难点,对类似的复杂系统有一定借鉴意义。     

基于图像处理的片式元件缺陷检测算法

表面缺陷检测是电子元器件生产过程中的关键步骤,通过分析表面缺陷检测的现状问题,结合片式元件图像的自身特征,提出一种基于改进的自适应阈值canny算子的片式元件缺陷检测算法;该算法把片式元件灰度图像进行区域分割,得到不同值域范围内的灰度区域,利用带有敏感度的自适应双阈值canny算子进行边缘检测;算法实现结果表明,该算法能高效地检测并处理不同缺陷类型的片式元件,并且可以达到工业生产的检测速度。     

视觉测量中多光学靶点亮度自适应控制

提出了一种面向大视场高精度视觉测量的多红外目标靶点（红外LED）亮度自适应控制方法。以微控制器PIC16F873A和LED驱动芯片MAX6964作为核心器件,设计了可满足大空间范围内靶点亮度控制需求的多目标靶点控制驱动器。为了使LED在近距测量时的调节精度和远近距离测量时的发光亮度都能满足要求,根据MAX6964输出特点设计了PWM信号输出电路,并采用推挽电路驱动LED。实验证明,目标靶点在CCD相机上成像光斑稳定一致。     

μC／OS—Ⅲ在Cortex—M3处理器上的移植

为了将μC／OS-Ⅲ移植到Cortex—M3处理器上,选用RealView MDK作为软件开发平台,针对Cortex—M3处理器特性编写了移植所需的C语言和汇编语言源代码,并验证了移植的正确性。移植后的μC／OS-Ⅲ能够稳定运行于Cortex～M3处理器上。该移植对大部分Cortex—M3处理器具有通用性,对其他架构处理器的μC／OS-Ⅲ移植具有参考作用。     

基于tinyAVR和Nokia3310 LCD的温度绘图仪

采用14引脚封装的tinyAVR单片机、广泛应用于手机中的液晶显示模块Nokia3310LCD和深受欢迎的单总线3引脚数字温度传感器DS18B20,构建了一款具备实用性和趣味性的,既可读数、又可读图的双模式温度绘图仪。     

Do different reputation systems provide consistent signals of seller quality: a canonical correlation investigation of Chinese C2C marketplaces

In recent years, consumer-to-consumer (C2C) marketplaces such as eBay and Taobao have adopted a component rating system, and run it simultaneously with but independent of a binary rating system. This paper investigates the extent to which binary rating and component rating systems are able to provide consistent signals of sellers?? quality, focusing on the reputation system design under the Chinese context. Using field data from Taobao, we performed canonical correlation analyses and found that the reputation signals of the two systems are generally correlated. As expected, negative and neutral ratings accurately reveal buyer dissatisfaction. Our results, however, show that positive ratings exhibit negative correlations with the three component ratings (i.e., item-as-described, customer service, and on-time delivery), suggesting that large numbers of positive ratings on Taobao may encourage trust in the platform but do not help to choose credible sellers. Our results elucidate the role of cultural difference in explaining the negative relationship in China and provide important implications for the design of reputation systems.     

Customer knowledge discovery from online reviews

The explosive growth of Chinese electronic market has made it possible for companies to better understand consumers?? opinion towards their products in a timely fashion through their online reviews. This study proposes a framework for extracting knowledge from online reviews through text mining and econometric analysis. Specifically, we extract product features, detect topics, and identify determinants of customer satisfaction. An experiment on the online reviews from a Chinese leading B2C (Business-to-Customer) website demonstrated the feasibility of the proposed method. We also present some findings about the characteristics of Chinese reviewers.     

数字化校园综合查询服务平台的设计与研究

本文在对数字化校园中心共享数据库及公共数据交换平台的研究基础之上设计了一种综合查询服务平台架构及查询服务功能,同时分析了查询实现过程。