Research On On-Line Detection System Of Surface Roughness Of Steel Plate

The surface roughness is an important indicator to evaluate the quality of steel surface.And it has a direct impact on the performance of steel,so the surface roughness detection in industrial production has a wide range of applications.The off-line detection is the main method used in the traditional testing of the steel’s surface roughness.But this method is inefficient in detecting the surface quality of steel and it’s unable to achieve the real-time monitoring of steel surface quality.In recent years,the optical method in the on-line detection is developing rapidly.This non-connect detection method,with high efficiency and fast detection,can avoid the problems in the traditional contact detection and it is suitable for on-line detection of surface roughness.In the detection of steel surface roughness,how to ensure the accuracy and stability of measurement is the focus.Therefore,we need to use the appropriate method to improve the performance of the detection system based on the actual needs.In this paper,the relative research in the surface roughness detection is firstly analyzed and compared.Then,an on-line surface roughness detection system based on the actual measurement environment and requirements is designed.The main contents of this article are as follows:1.The paper introduces the existing roughness detection methods and their advantages and disadvantages.According to the actual testing environment and measurement requirements,the on-line detection based on light-section method is selected,and the mechanical structure and hardware structure of the system are designed to realize the system’s motion.At the same time,the host computer control software is compiled to complete the system motion control and data processing.2.This paper introduces the theory of surface roughness detection and adopts the Gaussian filter to extract the roughness information.Aiming at the boundary effect and robustness of Gaussian filter in filter processing,this paper studies a robust Gaussian filter algorithm and corresponding data processing method to improve the accuracy of the surface roughness signal.And it’s filtering effect can be verified through experiments.3.Build and debug the entire roughness on-line detection system,then inspect the surface of the standard film and the steel plate respectively.Based on the evaluation method of surface roughness,the evaluation of the roughness parameters of the surface is completed.Through detection of the standard film,the system performance is tested.Finally,the steel is tested and the surface roughness of the steel is judged based on the maximum rule.