Research On Measurement Model And Control Method Of The Weight Of Zinc Coating In Continuous Hot-Dip Galvanizing Line

Hot-dip galvanized products have good machining performance and strong corrosion resistance,and therefore have been widely used in many fields,such as machinery industry,steel structure construction,automobile manufacturing,home appliance manufacturing,communication and power.With the growing demand for hot-dip galvanized sheet in the automobile industry,the requirements for steel plate quality are also increasing.The thickness and uniformity of the zinc layer are important quality standards for galvanized steel sheets and strips,and they also affect production costs to a certain extent.In the continuous hot-dip galvanizing line,air knife and zinc coating weight gauge are the core devices for controlling zinc layer weight.Both of them form a closed loop through a control algorithm.This thesis carried out a systematic study of zinc coating weight control arround the zinc coating weight measurement model,air knife control influencing factors,and control algorithm.At present,all the hot-dip galvanizing lines of major steel plants have installed zinc layer measurement systems produced by foreign companies,but the internal continuous hot-dip galvanizing industry has insufficient theoretical research on the zinc coating weight measurement method.To solve this problem,this thesis proposed a simulation method of zinc coating weight gauge based on Monte Carlo N-Particle Transport Code(MCNP)and introduced how to use MCNP to calculate and analyze the weight of the zinc layer evenly distributed on the sample.Based on the XRF measurement principle analysis and actual production equipment investigation,a physical model based on ZAIDs and atomic fractions was constructed.The changes of X-ray fluorescence spectrum and zinc fluorescence detection efficiency under different incident energies of monochromatic photons,detection distance and incident spot size were compared.The results show that:the smaller the difference between the incident photon energy and the characteristic X-ray energy of the target element,the greater the fluorescence intensity generated;The larger the detector window area,the greater the collected fluorescence intensity,and when the detector distance is 20?40mm,the fluorescence intensity reaches the maximum.The model constructed in this thesis established a good linear relationship between the fluorescence intensity of zinc K?and the zinc coating weight,effectively simulated the measurement process based on X-ray fluorescence,and provided a new method for the design and simulation of zinc coating weight gauges.In the hot-dip galvanizing by air wiping,there are various factors affect the weight of the zinc layer,among which several process parameters related to the air knife have a greater impact.Aiming at the object characteristics of non-linear and multi-variable in the zinc coating weight control process,a new method for simulating and optimizing the influence factors of zinc coating weight control based on orthogonal experimental design,numerical simulation and response surface method was proposed.Use experimental design and field data collection methods to obtain experimental sample data,use range analysis to obtain the primary and secondary order of each single factor on the response indicator,derive a linear regression formula,and optimize the model using orthogonal experiments with interactive columns.The results show that the coefficient of determination of the quadratic regression equation model obtained by fitting is 0.9976,P<0.0001.According to the experimental results,the relationship between the primary and secondary factors affecting the weight change of the zinc layer is:blowing pressure,strip speed,interaction of blowing pressure and strip speed,nozzle distance,nozzle gap,blowing pressure and nozzle distance.The result reveals the important influence of the intersection of factors on the weight of the zinc layer.On this basis,the control method of the weight of the zinc layer under the two modes of maximizing production efficiency and minimizing production energy consumption is determined.In order to further solve the complex problems of continuity,real-time,time-varying and hysteresis in the continuous hot-dip galvanized zinc coating weight control process,a new online zinc layer weight prediction method based on improved Dynamic Fuzzy Neural Network is proposed.The learning algorithm of the Dynamic Fuzzy Neural Network is introduced,and the prerequisites for Schmidt orthogonal decomposition in the learning algorithm are improved.At the same time,combining with orthogonal arrays,which can represent of the input space,the generalization ability of dealing with the high dimensional small sample is significantly improved.Taking the actual production data of the hot-dip galvanizing production line as the research object,the samples were trained and tested using radial basis neural network,BP neural network and Dynamic Fuzzy Neural Network respectively.After comparison and contrast,the effectiveness of the improved Dynamic Fuzzy Neural Network was verified.Based on the improved Dynamic Fuzzy Neural Network,the X-ray fluorescence measurement value was corrected using the alloying rate.The maximum deviation between the correction result and the gravimetric method is 1.73 g/m~2.2 sets of values in 10 sets of tests can be accurately predicted,10 sets predicted RMSE is 1.2737.Aiming at the problem that the measurement data of imported equipment cannot be obtained automatically,a zinc layer thickness gauge data acquisition system was developed to realize the automatic capture of the relevant data required,the final inspection report is formed after analysis and calculation,and is automatically uploaded to the production and marketing system.Aiming at the problem of manual compensation compensation experience and no theoretical basis,based on the data acquisition system and the above research,a control optimization support system was developed to realize the selection and prediction of control parameters.