Research And Design Of Three-dimensional Energy Monitoring System For Sintering Process Based On Intelligent Prediction Algorithm

Energy problems and dynamic monitoring of industrial processes has always been the concern of the steel industry.Sintering is one of the first three processes of steel industry,the production process of which is a typical complex process with continuous,non-continuous changes of material and energy.Some process parameters are of time-lag and cannot be measured accurately and so on.Therefore,strengthening the sintering process supervision and control can effectively control the energy consumption and sinter quality.At present,the use of three-dimensional monitoring in the industry is rare.Most of companies use of two-dimensional plane monitoring.In order to realisticaly display sintering process,this thesis uses three-dimensional technology for monitoring.The whole idea of this thesis is that the dynamic process of the sintering process is established,and the neural network is used to build the model to perform the dynamic prediction of the process.At the same time,the 3D model is used to build the 3D part of the process.Specific research and design are as follows:1.Parameter selection: a detailed analysis for the sintering process is conducted.According to the characteristics of the process and the related various types of industrial parameters,select the 9 parameters which are most relervant to energy consumption,and collect the data.2.System design and Database design: based on the requirements of the entire system,the system and the Database were designed.Through the analysis of the requirements of the database to determine the function of the basic table.Then the structural design,the E-R diagram,and the design of the table are carried out.3.Energy consumption forecast : according to the characteristics of the sintering process and the parameters,the Wavelet Neural Network is used to predict the energy consumption.Based on the data characteristics,the neural network is trained to get a good static prediction.The static prediction is to predict the average energy consumption(kgce / t)in one day.Due to the continuity and timeliness of the sintering process,dynamic predictions are made on the basis of static prediction,which can predict the energy consumption at a certain short time slot.In the dynamic prediction,the input of the neural network is the influence parameter of the current sampling time to be predicted and the influence parameter of the previous sampling time.The output is the energy consumption value at the current sampling period.Thereby the time delay of the process can be fully taken into account for accuracy.4.The research and design of 3D monitoring system are carried out for monitoring.The system's features include process display,data display,scene switching,alarm prompts and so on.This system consists of the following 4 main modules: 1)3D model design module: The realization of industrial processes in the three-dimensional reproduction of the device,according to the establishment of the three-dimensional model.Design and create these models including sintering machine,ring cooler and others;2)Interface and display modules: Importing the created 3D model into Unity3 D.Using the Particle System and GUI System to add the three-dimensional animation effects and display interfaces.Finally,complete the initial display interface's production;3)Alarm module: Comparing the forecast value and the real value of the energy consumption.When the difference between these two values is too large,the alarm light flashes,reminding the operator to adjust the parameters;4)Data-display module: Connecting Unity3 D and the Database to achieve the data interaction.This thesis finally implements the design of 3D monitoring system,which provides a good basis for the monitoring of sintering industrial process.