Parameters Identification And Ride Comfort Optimization Based On CANPSO-BP

Mining equipments occupies two-thirds of the market share in large open-pit mine transportation,but because of the lagging research and design in domestic industry,the mines relied on the imported mining dump truck in a very long time.In order to change this situation,the mining dump truck is classified as one of the 16 major technical equipment by State Congress in key areas,and put forward higher requirements on the performance of mining equipments.The development of mining dump truck has entered a rapid development stage driven by policy and market demand.Due to the mining working conditions,dump truck vibration effect the driver comfort in the process of driving,so that it is easy to get tired.Therefore,it is very necessary to study and optimize the mine dump truck ride comfort.The establishment of dynamic model of the vehicle is the key issues of ride comfort simulation analysis and research,accurate parameter acquisition is one of the primary problems of dynamics model.Due to the special nature and various aspects of the conditions restricting,the parameters which have a greater impact on vehicle ride,like seat air spring suspension,hydro-pneumatic suspension damping characteristics of the nonlinear stiffness and mine soft soil road roughness,is not easy to get through the test,or acquired high cost.Therefore,based on ride comfort experimental data,this article put forward the new particle swarm neural network algorithm to identify the parameters,it is great significant for reducing vehicle development costs and improving the level of development.And finally based on the identification parameters vehicle dynamics model,considering the impact of uncertainty on the optimization results,using improved PSO suspension nonlinear parameter uncertainty interval Optimize,effectively improve vehicle ride comfort.Main contents of this paper are as follows:1?This study combine niche technology with chaotic mutation evolutionary strategy and adaptive inertia weight adjustment to improve particle swarm optimization mechanism,improve the search speed and precision of algorithm.The improved particle swarm algorithm embedded neural network topology,to replace the network BP learning algorithm to create a new particle swarm neural network system(CANPSO-BP neural network).2 ? Based on the establishment of vehicle rigid coupling model,using CANPSO-BP neural network on seat air spring suspension and pneumatic suspension to identify and analysis the nonlinear stiffness and damping,by comparison with the standard BP network and PSO-BP network,verify CANPSO-BP network identification capability.According to the vehicle road test data,the nonlinear stiffness damping parameters of the seat air spring suspension and the oil suspension suspension are identified and verified respectively.3 ? Combined with ground-based vehicle mechanics theory,deduced the interaction force formula of the elastic tire interaction on the soft soil road.Write subroutine suitable for ADAMS TIRSUB tires,to set up road vehicles-soft soil coupling system.And on the basis of consideration of mining deformation of soft soil road,using CANPSO-BP algorithm to mine road roughness recognition.4?On the basis of vehicle dynamics model of identification parameter,using the method of interval uncertainty optimization,with the pneumatic suspension and seat air spring suspension nonlinear stiffness and damping coefficient as the optimization variables,the cargo capacity,centroid position and speed as uncertain variables,optimizing the seat root mean square value of weighted acceleration and comparing with the deterministic optimization,the results showed that: the optimal design parameters to enhance the dump truck vehicle riding comfort and has good robustness.