Comparative Analysis Of Fault Diagnosis Methods For Catalyst Production And Regeneration Equipment

With the continuous development of industry,the automobile has become the most important means of transportation for people’s daily travel.The demand for petrochemical products is increasing day by day.As a key material in petroleum refining,catalyst production has experienced rapid development in the decades of industrial development,especially the production equipment has also undergone several generations of product innovation.With the renewal and increasing of production and regeneration equipment,higher requirements are putting forward for reliability and fault diagnosis of catalyst production equipment.Once the equipment failure occurs suddenly,it will have a great impact on the production and reuse of refinery catalyst.In view of this,this paper carried out the study of catalyst production and regeneration equipment fault diagnosis methods.Catalyst production and regeneration equipment involves more than 20 devices,a large number of different types.According to the characteristics of various types of equipment for its classification,each category selected a representative of the equipment in-depth study.Fault diagnosis methods have been developed for many years,resulting in dozens of methods.According to different fault information processing methods,fault diagnosis methods are divided into three categories: knowledge-based,analytical model-based and signal processingbased.With the equipment of catalyst production and regeneration as the research object and the method of fault diagnosis as the means,a number of fault diagnosis application tests are carried out.The limitations and advantages and disadvantages of different fault diagnosis methods are analyzed according to the characteristics of the equipment and the problems in the process of fault diagnosis,it provides examples and references for the selection of fault diagnosis methods and the introduction of plant fault diagnosis methods.Specifically,the first part studies the background of refinery catalyst production and regeneration unit,the importance of safe and reliable operation of equipment.Studying the characteristics of catalyst production and regeneration equipment,classifying the equipment,selecting flue gas turbine,circulation pump,motoring three representative equipment,analyze the working principle,fault characteristics and impact of each equipment.Then the domestic and foreign fault diagnosis methods and reliability research are introduced.According to the difference of fault information processing methods,the fault diagnosis methods are divided into three categories.A fault diagnosis method with characteristics is selected in each large class.The principles and procedures of three diagnostic methods are analyzed.These methods have been used for many years in Lanzhou petrochemical catalyst production and regeneration equipment or are planned to be introduced.In the second part,the process of fault information collection and processing is studied.Then the spectrum analysis method is used as the fault diagnosis method,and the three representative equipments are selected as the experimental objects,and the fault diagnosis experiments are carried out on each equipment.The PEAKVUE technology is studied in the experiment of frequency spectrum analysis of motor,and it is applied in the experiment of motor and circulation pump.Finally,the paper collects and sorts out the capital and manpower investment in fault diagnosis methods of Lanzhou Petrochemical company over the years,and summarizes the effect of the experiment and the problems encountered.In the third part,the learning and topological structure of the neural network method is studied,and the fault diagnosis experiment of motor,circulation pump and flue gas turbine is taken as the center,and the diagnosis effect is constantly analyzed and evaluated,LM algorithm is used to improve the learning rate of BP neural network,which improves the global searching ability and convergence speed of the network.Finally,the limitations of the application of neural network method are summarized.In the fourth part,firstly,the improvement of FMECA in ISO is introduced,then the basis of catalyst production and regeneration equipment classification is analyzed,and FMECA is applied to flue gas turbine,the circulating pump and the motor were analyzed independently.This paper analyzes the effect and problems of FMECA in Lanzhou petrochemical company,and puts forward some suggestions for improvement.In the fifth part,based on the nine fault diagnosis experiments in the second,third and fourth parts,the application effects of different diagnosis methods on different types of equipment are analyzed and compared,in-depth study of the major types of fault diagnosis methods advantages and disadvantages and scope of application.This has a strong guidance and reference significance for other chemical enterprises to select and introduce fault diagnosis methods.