| In order to meet the requirements of power demand while being green and environmentally friendly,thermal power units in my country are developing towards super(super)critical units.However,during the long-term service,the internal pipe surface is oxidized.When the generated oxide scales fall off and accumulate to a certain degree,it will cause problems such as pipe blockage and bursting,and steam turbine main valve jamming,etc.,which seriously affects the safe operation of the thermal power plant.At present,the commonly used ultrasonic testing,radiographic testing,acoustic vibration testing and other methods all have certain shortcomings,which cannot achieve rapid and accurate measurement of the scale accumulation.Based on this background,this paper proposes to use the magnetic detection method to study the soft measurement technology of scale accumulation in the supercritical boiler tube,to achieve rapid and accurate measurement of scale accumulation.This article takes the power plant of CHINA HUANENG as the background,and takes the scale produced in the supercritical boiler tube as the research object.First of all,combined with the formation mechanism of boiler pipeline oxide scale,a non-stop state of the main pipeline particle separation technology is adopted,and an oxide scale detection scheme is proposed.Secondly,the feasibility analysis of oxide scale magnetic detection was carried out by ANSYS simulation software;simulation experiments were used to determine the effective structural parameters of the experimental device,and the rationality of the magnetic induction strength value around the austenitic stainless steel pipe as the auxiliary amount of scale accumulation in the pipe was verified.Then,an experimental device was built;in view of the abnormality of the original data collected,a data preprocessing method with improved threshold segmentation was proposed.Finally,the soft measurement method of scale accumulation is studied.In order to meet the requirements for rapid measurement of accumulation amount on site,a soft-sensing modeling method for scale accumulation amount based on magnetic field gradient is proposed by analyzing the spatial magnetic field distribution model of pipeline scale.In order to meet the requirements for accurate measurement of the accumulation amount,a particle swarm gray wolf-support vector machine(PSO-GWO-SVM)is a soft-sensing method for scale accumulation.In order to solve the problem of setting support vector machine hyper parameters,a particle swarm optimization algorithm(PSO)and gray wolf optimization algorithm(GWO)combined optimization strategy is used.The research results have verified the feasibility of the magnetic detection of oxide scale.In order to meet the different requirements on the site for rapidity and accuracy,a complete set of solutions for soft measurement of scale accumulation is given to better achieve scale accumulation in supercritical boiler tubes The need for real-time detection. |
