Parameter Estimationinthestate-space Modelforlifeprediction

The prediction of the remaining life of mechanical products can not only reduce the occurrence of accidents,avoid unnecessary losses,but also be important information for preventive maintenance and remanufacturing decisions.With regard to the long-term experience of the research on the remaining life prediction of mechanical products,in recent years,researches on the establishment of state-space models based on on-line monitoring data for the prediction of residual life have become more and more popular.However,the parameter estimation method of the state space model as the main link in the remaining life prediction needs further research.The parameter estimation method of the main link state space model needs further research.The prediction of the remaining life of mechanical products is a long-term and very complicated process.It requires us to track and observe the equipment for a long period of time in order to collect and obtain the corresponding more accurate historical data,so that we can do the development of the future.Evaluation and estimation.Especially for some relatively large and complex systems,which indicators are used to measure the mechanical remaining life of the whole machine is also an issue we need to urgently explore.Based on some of the prediction methods described in this article,there are certain deficiencies in the prediction of the remaining life of mechanical equipment.This article mainly introduces the method for predicting the remaining life of mechanical products,and puts forward some new prospects for the forecasted technology.This paper selects the vibration data of Rexnord's ZA-2115 double row cylindrical roller bearing.During the data acquisition,the NI DAQCard-6062E data acquisition card,PCB 353B33 acceleration sensor and data acquisition software developed by Lab VIEW are applied.The sampling frequency is 20 kHz,the sampling length is 20480,and it is collected every 10 min.The vibration intensity is obtained for every 10 min of sampling data.In order to verify the performance of the model,we analyze and analyze 200 vibration intensity data.In this paper,the state space model is established for the vibration intensity series of rolling bearings,and the trend of vibration intensity of rolling bearings is predicted and estimated.By using the Kalman filter algorithm and the particle filter algorithm,the reliability of the corresponding time is obtained separately.The prediction results fully prove the feasibility and effectiveness of the model.It is used for the prediction of the remaining life of mechanical equipment.It has broad prospects for improving the early detection capability of mechanical equipment,avoiding the operation of equipment in fault conditions,improving the safety of mechanical equipment,and reducing the maintenance of mechanical equipment.In this paper,the method of parameter estimation of the state space model is summarized and compared with the advantages and disadvantages.We analyze and compare the most commonly used six parameter estimation methods,discuss the implementation methods of various methods,and discuss their advantages and disadvantages.Integrate to select the best parameter estimation method.This not only contributes to the implementation of parameter estimation of the state space model,but also helps to propose improved algorithms for specific application examples.In this paper,two parameter estimation methods(Kalman filter algorithm and particle filter algorithm)are mainly used to establish the state space model for the monitored vibration intensity sequence of rolling bearings,and the variation trend of vibration intensity is predicted,and corresponding calculations are made.The reliability.The obtained prediction results prove the feasibility and effectiveness of the state space model and obtain satisfactory results.