Vibration Monitoring And Bp Neural Network-based Continuous Casting Slag Automatic Detection Method

The dissertation mainly introduces the research and development work of ladle-tundish slag detection. Based on lots of investigation of the internal and external slag detection technology, ladle-tundish slag detection method using vibration measurements, BP-ANN and analyzes the characteristic of the molten steel vibration signal has been developed. According to the characteristic the method of slag characteristic parameter identification has been developed. The method of the signal processing includes Time analysis, Frequency analysis, the Wavelet analysis and BP-ANN. Based on the dissertation, the software of slag detection using vibration measurements has been developed. And the method of slag detection using vibration measurements and BP-ANN is proved true and achieves the prospective requirement.This dissertation presents holistic design of the system, hardware design and software design. The software is easy to be operated and runs reliably, which is developed strictly under the software engineering. Module programming which can shorten the developing time and make it easy to be rebuilt is adopted. This system design achieves prospective requirement by designing and testing system.