On-line Monitoring Of Liquid Composite Molding Based On Digital Image Processing

The Liquid Composite Molding(LCM) is one kind of high performance and low cost advanced for technique of manufacture composite manufacturing process. LCM process attracts much attention due to good product quality,less influence of the operator's skill in the quality of final product,high repeatability,and environment-protection.The permeability is the measure for the resistance of the fibre reinforcement to the resin flow.Accurate permeability data is essential for the optimization of composite manufacturing process,arrangement of the inlet and outlet in the mold, reducing the production time,and improvement of the composite part quality. However,traditional methods of measurement have many limitations in monitoring of the permeability.Therefore,develop an up to date detection method is valuable.Machine vision is a new measuring technology.It acquires images by using CCD,then obtains the parameters measured by image processing method,this non-contact method is used in various kinds of fields widely because of its advantages such as on-line,real-time,rapid,suitable precision and strong scene anti-disturbance.An automatic measuring system based on the technology of digital image processing is designed for on-line monitoring of LCM process.Using the data of the image,viscosity,injection pressure,and time,this system obtains the permeability automatically.Firstly,the image processing method of the system is studied in detail, such as filter,threshold,improvement of the binary image and edge detection.Image processing algorithms are chosen according to the requirements.Compared to the result of traditional methods,the results obtained by the new method are proved suitable.Then,this paper designs the structure of whole system according to machine vision and the principle of image processing.After that,suitable CCD camera,data acquisition card,and press sensor are chosen and configured.Finally, the error of the whole system is analyzed and some methods to reduce the error are raised.The method overcomes the shortcoming of traditional methods such as lacking of real time quality and difficulty in monitoring.The efficiency of the method is demonstrated by experiment results...