Research On Intelligent Maintenance System Of Large Equipment Based On Optimized Hidden Markov Model

With the continuous innovation of science and technology,mechanical equipment is developing in the direction of precision,large,complex and intelligent.These equipment play a key role in the quality and efficiency of manufacturing,and the cost is often very expensive.The aging of equipment,the long maintenance time and the damage of equipment will cause great loss.Therefore,the intelligent maintenance of large equipment and the smooth operation of large equipment have become an important factor to improve the core competitiveness of the enterprises.Intelligent maintenance technology is a preventive maintenance technology,which can predict the performance decline of large equipment and enable the equipment to be effectively maintained.That is,intelligent maintenance technology can effectively support the effective use of equipment.Therefore,intelligent maintenance for large equipment is a hot topic in the industry.Based on the research background and significance of intelligent maintenance,the large-scale equipment package equipment is used as the research object,the intelligent maintenance of large equipment is carried out,the performance decline prediction of large equipment is predicted,and the feasibility of the scheme is determined.Based on system requirements,the general framework of intelligent maintenance system for large equipment is proposed.The real-time monitoring of the running state of large equipment is the basis of the intelligent maintenance system,and the prediction and evaluation of the performance decline process of the equipment and products is the core of the intelligent maintenance system.Firstly,the hardware system is set up to collect acceleration signal and acoustic emission signal of large carton packaging machinery.Through pre-filtering,zero average position processing,eliminating trend item processing and error-point elimination processing,signal preprocessing is carried out.The principal component analysis method is used to reduce the dimension of the data,and the distance based feature evaluation method is used to measure the importance of each feature,and the data is transmitted through the network.The establishment of appropriate models is crucial to predict the performancedecline of large equipment.In this paper,a weighted Hidden Markov Model method is proposed,which uses multiple weighted Hidden Markov models to predict the decline process of large mechanical equipment.In order to verify whether the performance degradation prediction algorithm of large equipment can meet the maintenance requirements,the system is tested in this paper,and the complete set of equipment for large carton packaging machinery is taken as the experimental object.The parameters of the weighted hidden Markov model are obtained by training data.The experimental results show that the weighted hidden Markov optimization algorithm can effectively predict the performance decline of the equipment and provide an effective support for the development of maintenance mode of large equipment.