随着设备向大型化、连续化和复杂化的方向发展，企业对设备的安全性、可靠性要求不断提高。一旦发生突发故障，将导致不可预料的巨大经济损失。为保证设备安全可靠运行，需要及时获取设备运行状态的信息，做到故障的提前预测。能够便捷、准确的实现数据采集和故障诊断成为研究的热点。 数据采集是故障诊断的前提，故障特征信息的提取是故障诊断的依据。本文以设备运行中的信息采集为目标，依托故障诊断机理、虚拟仪器技术和信号分析技术，提出了以PDA作为硬件平台的数据采集与诊断系统的设计方案，将独立分量分析(ICA)算法应用于故障特征信号的分离和提取中，设计开发了基于PDA的数据采集与诊断系统。 论文主要完成了以下工作： 第一：阐述机械设备中的关键部件——齿轮和轴承的故障诊断机理，分析了传动装置振动信号的故障特征，确定故障监测的参数类型和采集系统中的关键数据参数，设计基于PDA的数据采集和诊断系统的总体结构。 第二：在故障信号提取中，对独立分量分析(ICA)方法的优化算法进行深入的研究，运用独立分量分析(ICA)方法对盲信源信号（采集到的混杂信号）进行提取和分离，得到更为真实的特征信号。 第三：在LABVIEW PDA环境下，完成基于PDA的数据采集与诊断系统的设计与开发，将独立分量分析(ICA)算法在PDA环境中实现，实现数据采集、显示、分析和存储等功能。 第四，在减速器故障诊断实验台上，通过采集振动信号进行测试分析。结果表明，经ICA分离后的信号故障信息明显增强；基于PDA的数据采集系统可以摆脱空间上的限制，能够更便捷、准确的实现数据采集

Along with the developing of mechanical equipments towards enlargement, serialization and complication, the security and reliability of equipment requirement enhances unceasingly. If the unexpected accident happened, it will cause unpredictable huge economic losses. In order to ensure the safe and reliable operation, we need to provide equipment running status information to accomplish the fault early. Therefore, achieving the portable and accurate data acquisition and fault diagnosis becomes a research hotspot. Data acquisition is the premise for fault diagnosis; fault feature extraction is the basis of fault diagnosis. Putting information collection in equipment operation as target, relying on fault diagnosis mechanism, the virtual instrument technology and signal analysis technology, this thesis puts forward a data acquisition and diagnosis system design scheme,completes the design and development of data acquisition and diagnosis systems based on PDA.Independent component analysis (ICA) algorithm is applied to fault characteristic signal separation and extraction. The main work of this thesis is as followings: Frist：elaborated fault diagnosis mechanism of the key components—gear and bearing in machinery equipment,analysised the vibration signal fault characteristics of transmission device,determined the fault monitoring parameters of key types of collection system,and designed the general structure of data acquisition and diagnosis system based on PDA. Second：in the fault signal extracted, Emphasis is the independent component analysis (ICA) methods of optimization algorithm. Using independent component analysis (ICA) method to get more real characterization signals from the blind source signals (the collected mixed signals) which are extracted and separated. Third：combined with the virtual instrument technology and PDA technology,completed the design and development of data acquisition and diagnosis system based on PDA, which independent component analysis (ICA) algorithm is carried out in the PDA environment, also achieved the function of data acquisition, display, analysis and storage,etc. Finally, there is a testing analysis in gear fault diagnosis experiment platform, by collecting and analyzing the vibration signal.Results show that the fault features of signal are more obvious after the separation of ICA .The data acquisition system based on PDA can get rid of the restrictions on the space, and it can be more convenient and accurate to complete the data acquisition.