| Turbine lubrication system is an important part of large-scale steam turbines in petrochemical power industry.However,due to the difference of working conditions,the complexity of structure and other factors,the lubrication system of steam turbines is prone to various problems.If the problems are just identified by judging the mechanical appearance,they cannot be solved.So it will bring great difficulties to fault diagnosis.Therefore,if advanced analysis technology can be used to effectively identify and solve the problems existing in the steam turbine equipment and its lubrication system,unnecessary losses can be reduced and the service life of the equipment can be prolonged.Firstly,the article describes the advantages and disadvantages of the current mainstream analytical methods and the research progress at home and abroad.On this basis,the lubrication system of steam turbine is taken as the research object,and the wear of the lubricating oil system of steam turbine is studied based on the oil monitoring technology and grey prediction theory in statistics.Specific research contents include the following aspects:1.Analyse the structure and composition of the lubrication oil system of steam turbines in depth;classify and elaborate the wear types and common abrasive particles of friction pairs in the lubrication system,as well as the common elements and main wear devices in the system;compare the commonly used analysis methods in oil monitoring technology and determinate the analysis methods used in this subject.2.The lubrication system of 200MW turbogenerator unit in a power plant is sampled regularly at the oil tank,and the experimental research is carried out.Physical and chemical properties analysis,elemental spectrum analysis and ferrography analysis are used in the experiment.The results show that the values are normal and the lubricating oil system has no obvious faults.Experiments show that a variety of analytical techniques can effectively monitor the lubricating oil system.3.The lubrication system of 350MW turbogenerator unit in a power plant is sampled regularly and experimentally studied.By means of physical and chemical analysis,elemental spectrum analysis and Ferrography analysis,the wear and faults of the equipment are judged according to the experimental results,and the faults of seals and shaft parts in the lubrication system are predicted.After the maintenance of the equipment,it is found that the judgement results are basically consistent with the actual maintenance results,which proves the superiority of the combination of various oil analysis techniques.It further illustrates the importance of regular maintenance of machinery and equipment.4.Select the physical and chemical characterization data and elemental spectral data of Shell L-TSA 46~#turbine oil used in the above experiments as characteristic information.Determine the moisture content in physical and chemical properties as the maximum correlation parameter for iron element in elemental spectral analysis,which laid a foundation for wear prediction.In addition,a prediction model of iron content based on GM(1,1)model was established by means of MATLAB system.The model is verified by posterior ratio and small error probability.The prediction effect of the model is good and it is a first-order accuracy.In addition,the method of generating sequence residual is used to optimize the basic prediction model,and the correction factor?is introduced into the original prediction theory to establish the revised prediction model.The optimum correction factor?is determined to be 0.990,which further improves the prediction accuracy of the model. |