Research On Curve Surface Feature Recognition And Digital Reconstruction Of Minifog Impulse Nozzles

With dieing out of halon fire extinguishing agent, Water mist fire suppression technology shows wide application prospects because of its merits, just as no pollution, high speed, low water consumption, small destructiveness. Now, this fire fighting technique has became one of the most hot topics. In all kinds of water mist fire suppression systems, one of the most important part is the atomizing nozzle. Its property has direct influence on the performance of the whole system. At present, the main problem of our home products include: the structure of the nozzle is too complex and it is hard to control the operation parameters. So, it has big significance if we can develop a kind of nozzle with smaller volume, lower weight, simple mechanical structure and better aerosol performance. The Minifog impulse nozzle from the MINIMAX of Germany has these merits above. So in this paper, we use the Reverse Engineering (RE) as the main method to do some research on the feature extraction and curve surface reconstitution.In this paper, we choose one kind of Minifog impulse nozzle as the research object. After comparing with several kinds of data measurements, we choose the ATOS from GOM Corporation as the main equipment to obtain relative data. After that, we also do some pretreatment to the point cloud According to the problems existing in the data.By studying existing methods of feature recognition and extraction, a new method which is suitable for spiral scanning curved surface and can be used to recognize the boundary characteristic is raised. After cutting the points into slices, we do the boundary characteristic points extraction and contour line fit for every slice points. After doing that, we can get boundary line by fitting adjacent layer points, then, we can realize our purpose of feature extraction.After analyzing existing methods, we do comparison on these methods. A new method which is suitable for spiral scanning curved surface and can be used to recognize the boundary characteristic is raised. By using this method, we got the CAD model of this part. After doing the error analysis, we find the CAD model can satisfy our requirements.