Research And Development Of Sintering Endpoint Prediction And Operation Instruction System Based On Python

Sintering process plays an important role in iron-making process.The quality and output of sinter directly affect the benefit of iron-making production.Among them,the sintering end point is an important index to detect the state of sintering production,and the stability of the sintering end point is the precondition to produce high-quality sinter.Due to the characteristics of sintering process,the state of sintering end point can not be perceived in real time,and there is a strong hysteresis.In order to accurately predict the end point of sintering and take appropriate measures to guide the abnormal working conditions after abnormal sintering end point,based on Python,the sintering endpoint prediction and operation guidance system is researched and developed in this thesis.In order to solve the problem of strong hysteresis in sintering process,a deep neural network prediction model based on Python language is designed,and the trained model is applied to the prediction of sintering end point,the mean square error loss of the final model reached 0.375,with high accuracy and meeting the requirements of the site.At the same time,the intelligent optimization rules are made according to the important parameters that affect the sintering end point.When the sintering end point is abnormal,the intelligent regulating model of trolley speed is y=2n/xup-xlow[x-xup+xlow/2].The intelligent regulating model of ignition temperature is（?）.The intelligent optimization model of one-mix and two-mix moisture is|H₁^’-H₁|=|W_MB1^’-W_MB1|·|M₀+H₁-θ₁·(A₀+H₁+H_1S+H_1B)|,|H₂^’-H₂|=|W_MB2^’-W_MB2|·|M₀+H₁+H₂-θ₂·(A₀+H₁+H₂+H_1S+H_1B+H_2S+H_2B)-θ₁·（1-θ₂）·(A₀+H₁+H_1S+H_1B)|,The TFe content,FeO content,SiO₂ content,MgO content,Al₂O₃ content,CaO content and basicity of sinter were analyzed by data analysis,and the optimization model of chemical composition of sinter was established,the dynamic optimization interval is obtained,which can be used to monitor the state of each chemical composition in real time,combine the abnormal chemical composition closely with the mixture configuration,and formulate the optimization strategy for the mixture when the different chemical composition is abnormal,and using Python language to complete the development of the sintering end-point Prediction and Operation Guidance system,the accurate prediction of the sintering end-point and the all-round,multi-angle intelligent processing after the abnormal sintering end-point have been realized,an intelligent operation guidance system integrating point monitoring,terminal prediction,terminal prediction model optimization,anomaly alarm and anomaly handling is constructed,it provides useful guidance for sintering production personnel to realize the whole process optimization of sintering end point.