Font Size: a A A

The Modeling And Control Of Blast Furnace Burden Surface For The Index Optimization

Posted on:2019-03-10Degree:DoctorType:Dissertation
Country:ChinaCandidate:H G ZhangFull Text:PDF
GTID:1311330542953262Subject:Control Science and Engineering
Abstract/Summary:PDF Full Text Request
The iron and steel industry is the pillar of China's national economy,which is also with high energy consumption and many emissions.The iron-making process makes most of the energy consumption and emissions.In' The 13th Five-Year Plan for Iron and Steel Industry',the government asks to speed up the development of resource-saving and environment-friendly iron and steel industry,and actively develop and promote the use of energy-saving and low-carbon technology.It is inline with national development needs to develop the operations in blast furnace with intelligence,informatization and precise control.The charging operation belongs to one of the four operating patterns in blast furnace,which is the most commonly used measure to improve the status and ensure the stable production of blast furnace.The operation mechanism in burden charging process is complex in hash environment.It is difficult to detect the parameters in charging process.Thus,the optimization process of charing operation has always been a hot topic in academia and industry research.In this study,we focus on the charging operation in blast furnace in order to ensure the stable operation and index optimization.The burden surface reconstruction model is established,and the definition of burden surface is proposed.The clustering model based on the burden surface features is established to make analysis for the distributions of historical surface features.The index optimization model of burden surface is established,which can obtain the optimum burden surface shape suitable for the current operation condition.In the end,the strategy of precise charging is propsosed to improve the charging process and achieve the optimum burden surface shape.The main research contents and contributions are as follows:(1)This paper proposes and establishes the definition model of burden surface.The reconstruction and visualization model of burden surface is established based on the radar detection technology.Based on the charging mechanism,the '1D-2D-3D' definition strategy of burden surface is proposed,and the feature extraction model is established.The shape of the burden furnace is characterized by features,which corresponds to understanding habit of the operators to the burden surface shape.It can convert the burden surface research to multi-parameter control problems.The definition model of burden surface is the foundation of the research of the optimization of burden surface and charging operation.(2)This paper establishes the clustering model of burden surface features.Based on the extracted features of burden surface,the clustering analysis model is established.A clustering model suitable for the redundant characteristics of burden surface data is proposed,after the in-depth study of the Density and distance-based clustering(DD)algorithm.The clustering analysis of burden surface would help to analyze the historical surface feature information and establish the historical burden surface library.(3)This paper establishes the optimization model of burden surface for index optimization.Extreme Learning Machine algorithm is employed to establish the correlation model betweent the features of burden surface with the comprehensive performance index.The dynamic optimization strategy is set to obtain the optimal burden surface features information,which is suitable for the current operation condition.Based on the analysis of distribution characteristic of burden surface data,the imporved offline and online Extreme Learning Machine algorithms are proposed in order to improve the accuracy of optimization model.The model takes into account the impact of the lower adjustment parameters on performance indicators.The stable operaion and index optimization is the main purpose of model optimization,which would obtain the features of optimal burden surface and provide support for precise charging operation.(4)This paper proposes and establishes the precise charging strategy in charging process.Based on the 3-D definition of burden surface and charging mechanism,the charging stall division mechanism is established,which can calculate the required volumn of burden.And we put forward the concept of charging unit.The volume of required burden on the charging unit is calculated.Two precise charging strategies are put forward based on the real charging rules in blast furnace:the adaptive adjustment models of the throttle opening degrees and the rotation speed of charging chute.The precise charging operation improves the traditional burden rules,which can promote the development of automation and intelligence of charging operation.
Keywords/Search Tags:Optimization production, charging process, burden surface reconstruction, precise charging, machine learning
PDF Full Text Request
Related items