Research On The Construction Of Digital Twin Of Ultra-supercritical CFB Boiler Reheater For Operation And Maintenanc

It is the general trend to promote the digital transformation of traditional industries and realize the integration of digital revolution and energy revolution.In the future energy system with renewable energy and new energy as the main body,the high parameter circulating fluidized bed boiler will become the main force of flexible peak shaving and rotating reserve with its advantages of strong fuel adaptability,low emission cost and large load reduction ratio,and has great commercial potential.With the improvement of boiler operating parameters,the problems of over-temperature and tube explosion of heating surface are becoming more and more obvious.Therefore,it is of great significance to establish an accurate,effective and rapid temperature measurement or calculation model of heating pipe wall to ensure the safe and efficient operation of boiler.As a new digital simulation technology based on physical entity model,digital twin integrates multi-scale,multi-disciplinary and multi-physical quantities by using multi-source data such as sensor perception,operation history and simulation application.It has the advantages of expansibility,real-time and fidelity,and has the ability of calculation,analysis and prediction.It is one of the key technologies to realize intelligent power plant.In this paper,the low-temperature reheater of Weihe 660 MW ultra-supercritical circulating fluidized bed boiler is taken as the research object.Through computer programming,a digital twin mechanism model of low-temperature reheater that can form a bidirectional dynamic mapping between physical entity and digital model is constructed.The finite element method based on the control volume and the finite difference method are combined to calculate the temperature distribution of the steam and the tube wall.Empirical formula with high precision of the convective heat transfer coefficient is used to replace the calculation of the radial temperature distribution inside the steam,so the coarse grids can be used in the tube wall and the steam domain.The maximum calculation error of outlet steam temperature under each design condition is1.63°C.This method greatly reduces the number of grids and reduces the computational time without affecting the computational accuracy.By introducing the thick-walled cylinder mechanics theory and the equivalent stress model into the digital twin mechanism model,the structural stress,thermal stress and equivalent stress of the pipe wall can be calculated while calculating temperature of steam and the pipe wall.Then the equivalent heat transfer coefficient is introduced.By changing the heat transfer coefficient in the boundary conditions of the finite element mesh unit,the impact of oxidation on the inner wall or ash deposition on the outer wall of the heating pipe on the operation of the reheater can be simulated.Based on the twin mechanism model data,the flue gas flow rate,flue gas temperature,tube outer wall temperature and steam flow rate are used as input variables.Through the support vector regression after optimizing the parameters by genetic algorithm,the prediction model of oxide layer thickness and ash layer thickness of low temperature reheater tube is established to realize the interconnection between physical entities,digital twins,twin data and intelligent applications in digital twin.The calculation results show that the temperature difference between the inner and outer walls of the reheater tube increased by the development of output of boiler unit.Meanwhile the equivalent stress on the tube wall increases accordingly.The maximum equivalent stress is 40.88 MPa under BMCR condition.When the inner wall of the tube is oxidized for 1mm,the outlet steam temperature of the low temperature reheater decreases by 13.08℃,while the wall temperature increases by 11.54 ℃.When the thickness of the uneven ash layer outside the tube is 3mm,the outlet steam temperature of the low temperature reheater is reduced by 24.57°C,and the outer wall temperature of the tube is reduced by 33.94 °C.In order to ensure the quality of reheat steam in actual operation,it is necessary to improve the flue gas parameters to strengthen heat transfer in order to reduce the influence of metal oxidation of the inner wall or ash deposition of the outer wall on the heat transfer process.In addition,when other conditions remain unchanged,the lower the flow rate in the tube,the higher the steam temperature at the outlet of the low temperature reheater,and the increase of steam temperature will aggravate the oxidation of the inner wall.With the increase of oxidation thickness,the probability of oxide layer falling off increases.In severe cases,the pipeline will be blocked,resulting in a decrease in the local flow area of the branch pipe and a decrease in the flow.When the flow deviation coefficient is reduced to 0.2,the temperatures of the steam at the outlet of the low temperature reheater and the inner and outer walls of the tube reach 741.60 ℃,774.26 ℃,and 775.21 ℃,respectively,exceeding the maximum allowable service temperature of the tube wall material.Finally,according to the metal creep theory,the Larson-Miller parameter method is selected as the creep life model of the low temperature reheater.Through MATLAB software,the high temperature creep data of metal materials are fitted,and the life calculation formula with a fitting degree higher than 0.99 is obtained.The intelligent application part of the reheater digital twin is realized,which can provide reference for the actual operation and maintenance of the power plant.