Prediction Research Of The Straddle-type Monorail Vehicle Running Tire Partial Wear Based On Support Vector Machine

The operation of the straddle-type monorail vehicle line 3 in Chongqing shows that the straddle-type monorail vehicle running tire partial wear is serious and that replacement of tires is frequent,and that operating cost is very high.What's more,the running tire partial wear will cause the tire adhesion ability to decline,and make the straddle-type monorail vehicle driving force lower,and the braking distance and braking time longer,which seriously affect the traffic safety of straddle-type monorail vehicle.Therefore,this paper,with support vector machine(SVM)as the theoretical analysis tool,combined with the actual operation of the Chongqing Light Rail Line 3 trains,for the purpose that the prediction of the straddle-type monorail vehicle running tire partial wear,analysis of the monorail vehicle structure and kinetic parameters of the straddle-type monorail vehicle running tire partial wear,sets up the support vector machine regression prediction model with these parameters.The main contents and innovations are as follows:1)With the analysis of straddle-type monorail vehicle running tire partial wear and support vector machine status,pointing out that it is feasible that the prediction of straddle-type monorail vehicle running tire partial wear is used with the support vector machine.The support vector machine theory is introduced,and the support vector classifier machine and support vector regression machine is derived.2)Based on multi-body system dynamics theory,the single vehicle model as a subsystem,use MD ADAMS build train "Vehicle-Tyres-rail beam" coupling dynamics simulation model.3)With the combination of multi-discipline software modefrontier sensitivity analysis module and Adams,the sensitivity analysis model of parameters about straddle-type monorail vehicle running tire partial wear is established,screening the parameters of straddle-type monorail vehicle tire partial wear,for the foundation of the selection of independent variables about support vector machine(SVM)in the prediction model.4)With the combination of multi-discipline software modefrontier and Adams,taking the straddle-type monorail vehicle running tire partial wear parameters as independent variables,and regarding the straddle-type monorail vehicle running tire partial wear of evaluation index as the dependent variable,taking the operation safety of vehicles as the prerequisite,the multi-objective simulation model is established.Based on datas by modefrontier and Adams co simulation,the running tire partial wear support vectormachine prediction model is established,and the accuracy of the model was evaluated.