On Load Voltage Regulation Design And Fault Diagnosis Of Electric Melting Magnesium Furnace

With the continuous improvement of the degree of industrial automation,the equipment has more intelligent,integrated,multi complex change trend,and available data types are more and more with the use of advanced sensors.A large number of industrial data can provide rich materials for the fault diagnosis technology based on data driven.It is an important research direction how to mine data,extract data feature effectively,and make full use of the valuable information.At the same time,the valuable knowledge and experience of the technical personnel is of great significance to the rapid troubleshooting and fault diagnosis.It is an important opportunity and challenge for researchers to study how to combine the traditional methods of multivariate statistics and knowledge data to better serve the industrial safety.In this work,a semi supervised kernel discriminant analysis method based on knowledge data is proposed.At first,they are induced that the background and significance of the industrial process,the development of process monitoring,diagnosis methods and the status quo of domestic and foreign.Secondly,it introduces the traditional methods of fault monitor and diagnosis and Manifold Regularization method,as well as the method based on the feature extraction of knowledge information.Aiming at the current stage of fused magnesium furnace with constant current strategies encountered in the production process of high energy consumption and frequent faults,OLTC design is proposed,and through the multi-objective optimization model is solved to demonstrate the effectiveness and efficiency of the design.In terms of the high frequency breakdown of electrical melting furnace,the method is proposed based on the knowledge of frequent data semi supervised kernel discriminant analysis on the current method of kernel discriminant analysis.The production process data of electric magnesium alloy are simulated,and the validity of the method was verified to prove the validity and generalization ability of the data is strong.