Research And Development Of Mechanical Fault Diagnosis System For Medical Equipment Based On Information Fusion

Mechanical fault diagnosis system for medical equipment includes equipment fault diagnosis,daily maintenance and management,which can assist equipment administrators and maintenance personnel to grasp the dynamic information of equipment in time.In order to improve the success rate of maintenance personnel's diagnosis of equipment mechanical failures,this paper analyzes and studies the application of particle swarm optimization algorithm,BP neural network and information fusion to mechanical failures of medical equipment,and designs and develops medical equipment mechanical failure diagnosis systems.Improve the accuracy of mechanical fault diagnosis and assist maintenance personnel to make further judgments on equipment faults.Using UCI's device signal data set to verify the experiment,The main research contents of this article are as follows:(1)Based on the problems of data loss,redundancy and noise elimination in mechanical fault signals,this paper proposes wavelet analysis to reduce noise and improve the PSO algorithm(particle swarm optimization)Combined with the characteristics of multi-scale analysis when processing signal data with wavelet analysis,this paper makes use of the wavelet analysis to perform noise reduction processing on the signal data and takes advantage of non-linear functions in place of random functions to simplify the impact of randomness of functions on the optimal subset,and efficiently improve the selection of the optimal feature subset.Finally,the experimental comparison among the improved PSO,the original particle swarm algorithm and IPSO(Improved Particle Swarm optimization)proves that the IPSO has a better improvement in convergence than the other two,and solves the problem of serious randomness of the original algorithm.At the same time,the selection efficiency of the optimal feature subset is also improved.(2)Aiming at the problem of equipment mechanical fault diagnosis,this paper studies the theory and method of BP neural network and information fusion,and establishes a mechanical fault diagnosis model combining BP neural network with D-S evidence theory fusion algorithm.use BP neural network to initially diagnose the mechanical fault,DS evidence theory algorithm is used to make the initial diagnosis results of BP neural network fused in a decision-making,and finally the result of fault diagnosis is obtained.The fusion of the two theories takes advantage of the self-learning-characteristics of the BP neural network,reduces the error of the BP neural network in the diagnosis of mechanical faults,and further improves the accuracy of the mechanical fault diagnosis.(3)According to the design requirements of mechanical fault diagnosis system for medical equipment,the system development was completed,and the relevant functional tests and performance tests were performed on each functional module of the system.The results show that the system improves the success rate of the diagnosis of equipment mechanical fault bearing failure and gear failure provides an objective basis for the maintainers to carry out in-depth inspections of equipment mechanical fault.