Research On On-line Monitoring Technology Of Laser Cladding Quality Based On Optical Signals

Laser cladding uses high-power laser beam to melt the metal powders layer by layer onto the substrate,which can obtain small heat affected zone and high surface quality.However,laser cladding is a complicated multi-parameter process,and the small disturbance of parameters may result in defects.In addition,the working condition is prone to deterioration,which seriously affects the product quality.Therefore,it is of great significance to monitor the product quality on line in laser cladding.The optical emissions contain abundant information.Spectrometer is useful for spectral analysis,but it is expensive.For this reason,a set of low-cost acquisition system is established.Through the self-developed photoelectric sensor,two optical signals can be collected at the same time.The photodiode and signal amplifying circuit are integrated,which makes signal amplify first and then transmit in a long distance,reduces noise interference,and improves acquisition accuracy.Computer network technology provides an effective means for remote monitoring and diagnosis of laser cladding quality.The software based on C/S framework is divided into codec module,monitoring module,server module and client module.The monitoring module is responsible for collecting and analyzing data.The application of multi-thread technology enables the monitoring terminal to complete the high-speed acquisition and recording of two signals.Other modules and network unit in monitoring module realize the function of network communication.To provide data support for the on-line quality evaluation of the monitoring system,the influences of process parameters and working conditions on optical signal are analyzed systematically.To achieve good product quality,the cladding parameters can be optimized by following four requirements: RMS of the IR signal should be 4.0V–4.5V;RMS of the VIS signal should be 0.80V–0.95V;SD of the IR signal ought to be 0.6V–0.7V;SD of the VIS signal ought to be 0.13V–0.20 V.Under abnormal condition,the signal changes sharply and the 1/f noise is much higher.The software evaluation algorithm is improved by experiment law,which makes monitoring system judge the quality in real time.In order to verify the reliability and stability of the system,two experiments were carried out.One experiment is under normal conditional and the other is affected by water pollution.The results show that the system meets the expected functional requirements and runs well.