Research On Manufacturing Data Normalization And Automatic Check Technology For PCB

In the electronic product design and manufacture, manufacturing process of PrintedCircuit Board (PCB) is one of the important factors affecting performance and qualityof the electronic product. The designers tend to pay more attention on design, ignoringthe actual process. So, it is necessary to check the PCB manufacturing data; the currentcheck methods of PCB always use manual methods, but the efficiency is low and hashigh undetected rate, although some Electronic Design Automation (EDA) designsoftwares have check function, the check contents is not comprehensive and the filetype for check is relatively small. Therefore, it is of great significance to research themanufacturing data normalization and automatic check technology for PCB.By analyzing the heterogeneous EDA file structure, the paper designs anormalized data structure, achieves the feature of extracting and conversing themanufacturing data from EDA design files. It establishes the knowledge check rulesdatabase, which bases for intelligent automatic check on manufacturing data;it proposesthe deductive reasoning methods for check rules and the sub-block check idea,andestablishes the model of the intelligent automatic check.Experiments show that the normalized file can effectively store the manufacturingdata in EDA files. The proposed check method is applicable to a variety of data sources,and it can find design errors quickly, reduce the undetected rate, and have goodinteractivity, flexibility and scalability.The next step will study the expert experience values and find law in them, andupdate and improve the design specification and check rules.