A System For Injection Molding Quality Monitoring Based On Cavity Pressure

Cavity pressure is considered to be the "fingerprint" of injection molding process, and therefore, a cavity pressure monitoring technology has been developed. The technology, which is based on cavity pressure monitoring systems, is utilized to achieve some purposes including process optimization and quality monitoring. However, the existing cavity pressure monitoring systems have shortcomings in some aspects: first, some data processing functions like feature extraction in such systems are insufficient; second, such systems are not equipped with multivariate statistical monitoring methods; third, the high cost of the systems prohibits their widespread use. In consideration of the situation above, a system for injection molding quality monitoring based on cavity pressure is developed in this paper, which has advantages in data processing and quality monitoring.As a preliminary step, requirements analyses of hardware and software are conducted, and further, the architectures and modules of both hardware and software are designed.To overcome the deficiency in data processing, wavelet analysis is employed in noise reduction and feature extraction. In the data noise reduction, a noise reduction method based on wavelet transform is put forward. In the feature extraction, a feature extraction method based on wavelet decomposition is introduced, which enables the extraction of node energies and other features from time domain.As to quality control methods, a quality monitoring method based on statistical patterns of cavity pressures is proposed. The differences of this method and statistical pattern analysis(SPA) lie in the fact that statistical patterns of the presented method are more abundant than those of SPA method.Finally, the hardware and software implementation is finished, which achieves the function of data acquisition at millisecond level, curve visualization, feature extraction and quality monitoring. By conducting experiments on the presented monitoring system, the effectiveness of statistical monitoring methods and the reliability of the system are verified.