Research On PCB Model Region Clustering Fusion Algorithm And Its Application

PCB is one of the most important basic components in electronic products.It is the result of collaborative design between structural engineers and electronic engineers.Its design quality and efficiency will directly affect the product performance and R&D cycle.Structural engineers and electronic engineers plan the available space of PCB by drawing restricted region according to the design requirements of their respective fields.The current market puts forward higher requirements for the function,performance and appearance size of electronic products,which leads to the increasing number of restricted region in PCB,the diversity of geometric outline,and the increasing complexity of PCB model,the complex PCB model reduces the efficiency of PCB design.This thesis focuses on the research of PCB model region clustering fusion algorithm.Multiple regions with the same type and similar geometry are fused into a multi outline region.MCAD is used to model the fused regions uniformly to improve the efficiency of collaborative design for Engineers.The main research works of this thesis are as follows:（1）Analyzes the business process of PCB electromechanical collaborative design,focuses on the process of data exchange and model conversion in collaborative design,analyzes the reasons for the low efficiency of PCB collaborative design,and points out that the key to improve the design efficiency is to improve the efficiency and quality of model conversion in change design,and reduce the extra design cycle.（2）Studies the data organization of PCB model files,analyzes the geometric and non geometric information of regions in PCB model files,and designs a region geometric feature description and extraction scheme with small amount of calculation,low feature dimension and strong significance,combined with the characteristics and requirements of region fusion,analyzes the applicability of clustering algorithm.（3）Aiming at the problem that the number of clustering centers k of K-means clustering algorithm affects the clustering results and is difficult to calculate,this thesis proposes a curve fitting method based on the basic information of regions and the modeling characteristics of MCAD software,which estimates the optimal k adaptively through the total number of regions of PCB model,improves the region fusion ratio of PCB model.（4）In order to solve the problem that the clustering results are unstable and may not be the global optimal solution due to the random selection of clustering centers in K-means algorithm,CFSFDP（Clustering by fast search and find of density peaks）algorithm is used to optimize it.Firstly,the algorithm is improved,and a method is proposed to adaptively calculate the cut off distance(9₍（8） for each data point,which replaces the global cut off distance set manually in the traditional algorithm.Finally,the clustering center is automatically calculated with the decision value of the data point,so as to improve the stability and accuracy of regional clustering.The experiment of PCB model regions clustering fusion shows that the algorithm is more accurate and independent of the location and angle of the region.The display efficiency of PCB structure model is increased by 1.67-4.92 times by fusion modeling.The clustering experiments of public data sets show that the accuracy and iteration times of the proposed algorithm is better than that of the same type of K-means algorithm and Fuzzy C-means clustering algorithm,and the clustering time is between them.（5）Creo Protookit,Tiny XML and other toolkits are used to research and develop electromechanical collaborative prototype system based on regional clustering fusion.The regional clustering algorithm is applied to open PCB model module to improve the display efficiency of PCB model and the design efficiency of engineers.Finally,the effectiveness of the regional clustering fusion algorithm and the functional integrity of the prototype system are verified by a collaborative example.