| The integrated transmission device is the core power component of the armored combat vehicle,which can realize driving and steering,the performance status directly determines the maneuverability of the armored fighting vehicle on the battlefield and affects its combat capability.With the rapid development of our country’s economy,our army’s logistics support system has been continuously strengthened,weapon maintenance technology is developing vigorously.At present,there are relatively few studies on the health prediction of integrated transmission device,the software system related to it is in the initial stage of development.In order to solve this problem,based on actual engineering projects,this dissertation conducts in-depth research on integrated transmission device,constructs a health prediction method for integrated transmission device,and designs its software system,provide convenience for the maintenance staff.First of all,this dissertation analyzes the working principle of the integrated transmission device,and finds that it has the working characteristics of variable working conditions,and the health status lacks a fixed evaluation standard;This dissertation divides the monitoring signals into variable-condition signals and fixed signals according to whether the standard working value follows the changes of working conditions;And determine the functional components corresponding to each type of signal,analyze the mutual influence between them;and divide the integrated transmission device into four functional modules according to the use functions: power receiving mechanism,,transmission mechanism,steering mechanism,and action mechanism;A health prediction method is proposed,which first realizes the health prediction of a single component,then realizes the health prediction of its upstream module,and finally realizes the health prediction of the entire integrated transmission device.Second,transform the monitoring data into health indicators through the inferiority degree models,and use the inferiority degree fuzzy matrix to determine the health status distribution of the components,and use the analytic hierarchy process to build a health recognition method for upstream functional modules;Aiming at the working characteristics of the integrated transmission device with variable working conditions,with the help of transfer learning ideas,the two continuous working conditions are divided into the source field and the target field,substitute the one-dimensional performance trend features extracted from the source domain into the target domain for analysis,construct a public feature set,and realize the feature migration from the source domain to the target domain;Combining the characteristics of gray prediction model with strong trend expression and BP neural network fitting performance,using the idea of error correction,construct a combinatorial optimization model to include the advantages of another model.And take the transmission structure of the integrated transmission device as an example to verify,combine monitoring data to identify its health status,in order to verify the sustainable working time,the remaining life of a single shift clutch is predicted,substituting the feature migration and non-feature migration data into three models for verification,and comparing the experimental results,it is verified that the feature migration method can effectively realize the trend characteristic migration,improve the accuracy of prediction,and realize the reuse of historical data,and verify the prediction accuracy of the combined optimization model is better than the two single prediction models,error stability is better,and it is more suitable for long-term prediction.Finally,with Visual Studio 2013 as the development environment and SQL Sever2008 as the database carrier,and use the C# language to construct the health prediction software system for integrated transmission device,and display its functions. |
PDF Full Text Request